The Obstetric Hematology Manual. Manual

User Manual: manual pdf -FilePursuit

Open the PDF directly: View PDF .
Page Count: 278

Download
Open PDF In Browser	View PDF

This page intentionally left blank

The Obstetric Hematology
Manual

The Obstetric Hematology
Manual
Edited by
Sue Pavord
University Hospitals of Leicester NHS Trust

Beverley Hunt
Guy’s and St. Thomas’ NHS Foundation Trust and King’s College, London

CAMBRIDGE UNIVERSITY PRESS

Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore,
São Paulo, Delhi, Dubai, Tokyo
Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521865647
© Cambridge University Press 2010
This publication is in copyright. Subject to statutory exception and to the
provision of relevant collective licensing agreements, no reproduction of any part
may take place without the written permission of Cambridge University Press.
First published in print format 2010
ISBN-13

978-0-511-67748-9

eBook (NetLibrary)

ISBN-13

978-0-521-86564-7

Hardback

Cambridge University Press has no responsibility for the persistence or accuracy
of urls for external or third-party internet websites referred to in this publication,
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.

Contents
List of contributors
Preface ix
Acknowledgments

page vii
x

Section 1. Cellular changes
1.

Normal hematological changes
during pregnancy and the puerperium
Margaret Ramsay

Section 4. Thrombophilia and
fetal loss

Hematinic deficiencies
Jane Strong

Inherited red cell disorders 28
Emma Welch and Josh Wright

Maternal autoimmune cytopenias
Hamish Lyall and Bethan Myers

11. Antiphospholipid syndrome 131
Sue Pavord, Bethan Myers, and Beverley Hunt
45

12. Thrombophilia and pregnancy loss
Isobel D. Walker

Section 2. Feto-maternal
alloimmune syndromes
5.

13a. Management of obstetric
hemorrhage: obstetric management
Annette Briley and Susan Bewley
13b. Management of obstetric
hemorrhage: anesthetic management
Vivek Kakar and Geraldine O’Sullivan

Acute management of suspected
thromboembolic disease in pregnancy
Andrew J. Thomson and Ian A. Greer

Thromboprophylaxis 99
Sarah Germain and Catherine Nelson-Piercy

Prosthetic heart valves
Claire McLintock

109

151

158

13c. Management of obstetric
hemorrhage: hemostatic
management 166
Eleftheria Lefkou and Beverley Hunt

Section 3. Thromboembolism and
anticoagulation
7.

141

Section 5. Hemorrhagic disorders

Fetal/neonatal alloimmune
thrombocytopenia 63
Michael F. Murphy
Red cell alloimmunization
Alec McEwan

10. Management of anticoagulants
at delivery 120
Christina Oppenheimer and Paul Sharpe

13d. Management of obstetric
hemorrhage: radiological
management 171
Ash Saini and John F. Reidy
14. Inherited disorders of primary
hemostasis 176
Sue Pavord
15. Inherited coagulopathies
Sue Pavord

186

Contents

Section 7. Malignant conditions

16. Genetic counseling and pre-natal
diagnosis in hemophilia 194
Andrew Mumford

19. Myeloproliferative disorders 229
Claire Harrison and Susan E. Robinson
20. Effects of chemoradiotherapy for
hematological malignancy on fertility
and pregnancy 243
Seonaid Pye and Nina Salooja

Section 6. Microangiopathies
17. Pre-eclampsia 203
Eleftheria Lefkou and Beverley Hunt
18. Thrombotic thrombocytopenic
purpura and other microangiopathies
Marie Scully and Pat O’Brien

218
Index 253

Contributors

Susan Bewley
Women’s Services, Guy’s and St. Thomas’ NHS
Foundation Trust, London, UK

Claire McLintock
Natural Women’s Health, Auckland City Hospital,
Auckland, New Zealand

Annette Briley
Maternal and Fetal Research, Guy’s and St. Thomas’
NHS Foundation Trust, London, UK

Andrew Mumford
Bristol Haemophilia Centre, Bristol Haematology and
Oncology, Bristol, UK

Sarah Germain
Diabetes and Endocrine Centre, Guy’s and
St. Thomas’ NHS Foundation Trust, London, UK

Michael Murphy
National Blood Service, John Radcliffe Hospital,
Headington, Oxford, UK

Ian A. Greer
Hull York Medical Centre, University of York,
Heslington, York, UK
Claire Harrison
Department of Haematology, Guy’s and St. Thomas’
NHS Foundation Trust, London, UK
Beverley Hunt
Department of Haematology, Guy’s and St. Thomas’
NHS Foundation Trust and King’s College,
London, UK
Eleftheria Lefkou
Department of Haematology, Guy’s and St. Thomas’
NHS Foundation Trust, Lambeth Palace Road,
London, UK
Vivek Kakar
Department of Anaesthesia and Intensive Care, Guy’s
and St. Thomas’, NHS Foundation Trust, London, UK

Bethan Myers
Department of Haematology, Queen’s Medical
Centre, Nottingham, UK
Catherine Nelson-Piercy
Department of Obstetrics, Guy’s and St. Thomas’
NHS Foundation Trust, London, UK
Pat O’Brien
Department of Obstetrics and Gynaecology,
University College London Hospitals, London, UK
Christina Oppenheimer
Department of Obstetrics and Gynaecology, Leicester
Royal Infirmary, Leicester, UK
Geraldine O’Sullivan
Department of Anaesthetics, Guy’s and St. Thomas’
NHS Foundation Trust, London, UK
Sue Pavord
Department of Haematology, Leicester Royal
Infirmary, Leicester, UK

Hamish Lyall
Department of Haematology, Norfolk and Norwich
University, Norwich, UK

Seonaid Pye
Department of Haematology, Charing Cross Hospital,
London, UK

Alec McEwan
Department of Obstetrics and Gynaecology, Queen’s
Medical Centre, Nottingham, UK

Margaret Ramsay
Department of Obstetrics and Gynaecology, Queen’s
Medical Centre, Nottingham, UK

vii

List of contributors

John F. Reidy
Department of Radiology, Guy’s and St. Thomas’
NHS Foundation Trust, London, UK

Jane Strong
Department of Haematology, Leicester Royal
Infirmary, Leicester, UK

Susan E. Robinson
Department of Haematology, Guy’s and St. Thomas’
NHS Foundation Trust, London, UK

Isobel D. Walker
Department of Haematology, Glasgow Royal
Infirmary, Glasgow, UK

Nina Salooja
Division of Investigating, Imperial College London,
London, UK

Emma Welch
Department of Haematology, Royal Hallamshire
Hospital, Sheffield, UK

Marie Scully
Department of Haematology, University College
London, London, UK

Josh Wright
Department of Haematology, Royal Hallamshire
Hospital, Sheffield, UK

Paul Sharpe
Department of Anaesthesia, Leicester Royal
Infirmary, Leicester, UK

viii

Preface

This book aims to appeal to both those who have
already submersed themselves in the field of obstetric haematology and new-comers to the area. Many
have already discovered the numerous challenges and
dilemmas involved but also have found this area of
medicine to be both stimulating and rewarding. Others may be new to the field or have unwittingly found
themselves regularly involved in the care of these
women. We hope that all will benefit from this manual, which reflects up-to-date clinical management of
this complex group of patients as they present in clinical practice.
The impact of haematological disease on fertility,
pregnancy and the puerperium can be considerable. Thrombosis and haemorrhage are the leading
causes of maternal mortality and a large number of
haematological conditions are associated with fetal
loss. Advances in fetal maternal medicine and obstetric care has enabled high expectations of fetal survival and maternal wellbeing. However the stakes
are high, management can be complex and good
outcomes require excellent multidisciplinary team
work.
New challenges arise in the light of changing cosmopolitan populations, including rising birth rates
and improved survival and fertility from chronic ill-

nesses and life-threatening conditions. Thus in-depth
understanding is required to deal with this broad
range of disease. We are fortunate to have such a distinguished group of contributors, whose knowledge,
experience and opinions are invaluable, particularly in
an area where randomised clinical trials are scant and
good quality evidence hard to find.
This branch of medicine is gaining increasing
recognition as a subspecialist area, with the growth
of national and international specialist groups and
development of educational courses in the area.
Clinical problems have become an important feature in postgraduate examinations, both in hematology and obstetrics. This book is therefore not only
an important guide for practitioners in haematology, obstetrics, midwifery, and obstetric anaesthesia
but is invaluable for those studying for postgraduate
examinations.
Obstetric haematology is immensely rewarding,
and we hope this book provides encouragement,
particularly for those who are new to the speciality, to view it as both thought-provoking and
enjoyable.
Sue Pavord
Beverley Hunt

Acknowledgments

Thanks to our families for tolerating our time away
in writing and editing, and to the Staff of Cambridge
University Press, who guided us.

Section

Cellular changes

Section 1
Chapter

Cellular changes

Normal hematological changes during
pregnancy and the puerperium
Margaret Ramsay

Introduction
There are both subtle and substantial changes in
hematological parameters during pregnancy and
the puerperium, orchestrated by changes in the
hormonal milieu. A thorough understanding of these
is important to avoid both over and under-diagnosing
abnormalities. Appreciation of the time frame for
some of the changes allows sensible planning; this
is particularly true when considering thromboprophylaxis.
Some of the quoted reference ranges may differ
between centers, depending on laboratory techniques.
However, the principles of recognizing physiological
changes can still be applied.

Red cells
During pregnancy, the total blood volume increases by
about 1.5 l, mainly to supply the needs of the new vascular bed. Almost 1 liter of blood is contained within
the uterus and maternal blood spaces of the placenta.
Expansion of plasma volume by 25%–80% is one of
the most marked changes, reaching its maximum by
mid pregnancy. Red cell mass also increases by 10%–
20% but the net result is that hemoglobin (Hb) concentration falls.1 Typically, this is by 1–2 g/dL by the
late second trimester and stabilizes thereafter. Women
who take iron supplements have less pronounced Hb
changes, as they increase their red cell mass proportionately more than those without dietary supplements
(the increase is approximately 30% over pre-pregnancy
values).1
It is hard to define a normal reference range for
Hb during pregnancy and the limit for diagnosing
anemia. The World Health Organization has suggested that anemia is present in pregnancy when Hb

concentration is ⬍ 11 g/dL. However, large studies
in healthy Caucasian women taking iron supplements
from mid pregnancy found Hb values in the early
third trimester to be 10.4–13.5 g/dL (2.5th–97.5th centiles)2 . Studies from other ethnic populations have
documented lower third trimester Hb concentrations,
which may be attributable to the women entering pregnancy with poor iron stores or with dietary deficiencies
of iron and folic acid.
Red cell count and hematocrit (Hct) values are likewise lower in pregnancy, but the other red cell indices
change little (Table 1.1), although red cells show more
variation in size and shape than in the non-pregnant
state. There is a small increase in mean cell volume
(MCV), of on average 4 fL for iron-replete women,
which reaches a maximum at 30–35 weeks gestation
and occurs independently of any deficiency of B12 and
folate.2
Hemoglobin and hematocrit increase after delivery. Significant increases have been documented
between measurements taken at 6–8 weeks postpartum and those at 4–6 months postpartum, demonstrating that this length of time is needed to restore them to
non-pregnant values.1

Summary points

r Hb concentrations decrease in pregnancy.
r Hb ⬍ 10.4 g/dL suggests anemia.
r Hb ⬎ 13.5 g/dL is unusual and suggests
inadequate plasma volume expansion (which can
be associated with pregnancy problems including
pre-eclampsia and poor fetal growth).
r MCV is normally slightly increased.
r MCH and MCHC are normally unchanged in
pregnancy and do not change with gestation.

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

Section 1. Cellular changes

Table 1.1 Red cell indices during pregnancy and the puerperium

Gestation
Red cell indices

18 weeks

32 weeks

39 weeks

8 weeks postpartum

Hemoglobin (Hb) g/dL

11.9 (10.6–13.3)

11.9 (10.4–13.5)

12.5 (10.9–14.2)

13.3 (11.9–14.8)

Red cell count × 1012 /L

3.93 (3.43–4.49)

3.86 (3.38–4.43)

4.05 (3.54–4.64)

4.44 (3.93–5.00)

Mean cell volume (MCV) fL

89 (83–96)

91 (85–97)

91 (84–98)

88 (82–94)

Mean cell hemoglobin (MCH) pg

30 (27–33)

30 (28–33)

30 (27–32)

Mean cell hemoglobin concentration
(MCHC) g/dL

34 (33–36)

Hematocrit

0.35 (0.31–0.39)

0.35 (0.31–0.40)

0.37 (0.32–0.42)

0.39 (0.35–0.44)

Mean and reference ranges (2.5th–97.5th centiles). Samples were collected longitudinally from 434 women.
Adapted from Ref 2.

White cells
2

White cell count (WBC) is increased in pregnancy
with a typical reference range of 6 × 109 –16 × 109 /L.
In the hours after delivery3 , healthy women have been
documented as having WBC 9 × 109 –25 × 109 /L. By
4 weeks post-delivery, typical WBC ranges are similar
to those in healthy non-pregnant women (4 × 109 –10
× 109 /L).
There has been much discussion about the normal ranges for the different types of white cells.4 Neutrophils contribute most to the overall higher WBC.
There is an increase in immature forms and the cytoplasm shows toxic granulation. The count3,4 is relatively constant throughout gestation (3 × 109 –10 ×
109 /L), markedly elevated in the hours after delivery (up to 23 × 109 /L) and back to non-pregnant
values by 4 weeks post-partum (1.5 × 109 –6 ×
109 /L). Neutrophil chemotaxis and phagocytic activity are depressed, the latter being inhibited by factors
present in pregnancy serum. There is also evidence of
increased oxidative metabolism in neutrophils during
pregnancy.
Lymphocyte count3,4 decreases during pregnancy
through first and second trimesters, increases during
the third trimester, but remains low in the early puerperium as compared to normal non-pregnant values.
Typical pregnancy range for lymphocyte count is 1.1
× 109 –2.8 × 109 /L, compared with the non-pregnant
reference range 0.8 × 109 –4.0 × 109 /L. Lymphocyte
count is restored to normal range by 4 weeks after
delivery. Detailed studies of T and B lymphocyte subsets in peripheral blood and the proliferative responses

of these cells to mitogens found more helper and suppressor cells and less killer cells during pregnancy.
Lymphocyte proliferation in response to a variety of
agents was found to be impaired in pregnancy, suggesting that there is an immunosuppressant factor present
in the serum.
Monocyte count is higher in pregnancy, especially in the first trimester, but decreases as gestation
advances.4 Typical values3,4 in the third trimester are
0.2 × 109 –1.0 × 109 /L, as compared to non-pregnant
values 0.1 × 109 –0.9 × 109 /L. The monocyte to lymphocyte ratio is markedly increased in pregnancy.
Eosinophil and basophil counts do not change significantly during pregnancy.3
Myelocytes and metamyelocytes may be found in
the peripheral blood film of healthy women during
pregnancy and do not have any pathological significance.

Summary points

r WBC is elevated in pregnancy, mostly due to
neutrophilia.
r Lymphocyte count is lower and monocyte count
higher.
r During pregnancy, only WBC ⬎ 16 × 109 /L is
considered abnormal.
r Soon after delivery, only WBC ⬎ 25 × 109 /L is
considered abnormal.
r Eosinophil and basophil counts do not change in
pregnancy.

Chapter 1. Normal changes

Platelets
Large cross-sectional studies in pregnancy of healthy
women (specifically excluding any with hypertension)
have shown that the platelet count decreases during
pregnancy, particularly in the third trimester. 5 This is
termed “gestational thrombocytopenia.” Almost 12%
of women in one study5 were found to have a platelet
count of ⬍ 150 × 109 /L late in pregnancy. Of these
women, 79% had platelet counts 116 × 109 –149 ×
109 /L; none had complications related to thrombocytopenia and none of their babies had severe thrombocytopenia (platelet count ⬍ 20 × 109 /L). Thus, it
has been recommended that the lower limit of platelet
count in late pregnancy should be considered as 115 ×
109 /L. Only 1% of healthy women have platelet counts
⬍ 100 × 109 /L.
Platelet size is an indicator of the age of the
platelets; young ones are large and they become progressively smaller with age. Platelet volume has a
skewed distribution, tailing off at larger volumes. The
platelet volume distribution width increases significantly and continuously as gestation advances and the
mean platelet volume becomes an insensitive measure
of platelet size. Studies suggest that platelet lifespan is
shorter in pregnancy. The decrease in platelet count
and increase in platelet size in pregnancy suggests that
there is hyperdestruction of platelets.
Platelet function, as assessed by the time required
for whole blood to occlude a membrane impregnated
with either epinephrine or adenosine 5’diphosphate
(ADP), has been studied in late pregnancy.7,8 No correlation was found between platelet count and the “closure times” over a range of platelet counts 44 × 109 –
471 × 109 /L in healthy women.8 Another study found
that the closure times were increased in women with
severe pre-eclampsia, although they did not correlate
with clinical bleeding problems in these women.9 In
women with gestational thrombocytopenia, platelet
closure times are influenced by hemoglobin level,
being prolonged when there is both thrombocytopenia
and anemia.7 This is perhaps not surprising, given the
contribution of red cells to the hemostatic process, in
part due to ADP donation. The increase in fibrinogen
during pregnancy helps to maintain platelet function.

Summary points

r Platelet count decreases during pregnancy in
some patients.

r The lower limit of normal platelet count at term is
115 × 109 /L.
r There is evidence of platelet hyperdestruction in
pregnancy.
r Platelet closure times are not affected by absolute
platelet count in healthy women during
pregnancy.
r Platelet closure times are prolonged when there is
anemia in addition to a low platelet count.
r The increase in fibrinogen during pregnancy more
than compensates for the fall in platelet count.

Coagulation factors
Screening tests used to assess the coagulation pathways include the activated partial thromboplastin time
(APTT), which measures the intrinsic pathway, the
prothrombin time (PT), which measures the extrinsic pathway, and the thrombin time (TT) which measures the final common pathway. In pregnancy, the
APTT is usually shortened, by up to 4 seconds in the
third trimester, largely due to the hormonally influenced increase in factor VIII. No marked changes in
PT or TT occur.
Many coagulation factors are increased in pregnancy (Table 1.2). Von Willebrand Factor and Factors
VII, VIII, X, and fibrinogen increase substantially as
gestation advances. In one longitudinal study,10 Factor VII activity increased from the range 60%–206%
(compared to standard) at the end of the first trimester
to 87%–336% by term. The same study, found Factors II and V increased in early pregnancy, but then
reduced in the third trimester. Another cross-sectional
study found a 29% rise in Factor V from 6–11 weeks’ to
36–40 weeks’ gestation.11 Increased levels of coagulation factors are mediated by rising estrogen levels and
thought to be due to both increased protein synthesis and enhanced activation by thrombin. Coagulation
factors remain elevated in the early puerperium and
for assessment of true non-pregnant levels, it is best to
sample 8–12 weeks after delivery.

Summary points

r APTT is usually shortened in pregnancy.
r Von Willebrand factor and factors VII, VIII, X,
and fibrinogen increase.
r There is a variable change in factor XI levels.
r Coagulation factor levels remain high in the early
postpartum period.

Section 1. Cellular changes

Table 1.2 Coagulation factors during pregnancy and the early puerperium

6–11
weeks
N = 41

12–16
weeks
N = 28

17–23
weeks
N = 10

24–28
weeks
N = 19

29–35
weeks
N = 36

36–40
weeks
N = 23

3 days
post-natal
N = 87

Prothrombin fragments 1 + 2 nmol/l

1.1
⬍ 2.9

1.1
⬍ 1.5

1.3
⬍ 2.1

1.8
⬍ 3.4

2.0
⬍ 3.9

1.9
⬍ 3.5

2.2
⬍ 4.9

Fibrinogen activity
g/l

3.6
2.5–4.8

3.8
2.5–5.1

3.6
2.6–4.7

4.4
2.9–5.9

4.1
2.5–5.8

4.2
3.2–5.3

4.5
3.1–5.8

Prothrombin activity
iu/dl

153
107–200

160
111–209

153
41–265

172
92–252

153
100–211

162
107–217

169
108–231

Factor V activity
u/dL

99
39–159

101
39–162

111
47–175

108
50–166

111
43–179

129
65–194

141
71–211

Factor VIII activity
iu/dl

107
62–220

129
82–130

189
59–159

187
71–341

180
31–328

176
50–302

192
54–331

Factor IX activity
iu/dl

100
49–151

106
82–130

96
74–118

121
59–183

109
65–154

114
79–150

136
65–207

Factor X activity
iu/dl

125
88–162

129
78–180

128
50–206

159
52–263

146
81–212

152
113–191

162
69–254

Factor XI activity
iu/dl

102
50–154

103
58–147

86
58–114

102
45–162

100
31–169

92
36–181

96
46–146

Factor XII activity
iu/dl

137
70–204

160
52–268

186
64–247

170
54–286

178
78–278

179
62–296

174
86–262

Von Willebrand Antigen iu/dl

137
70–204

160
52–268

186
64–247

170
54–286

178
78–278

179
62–296

174
86–262

RCo
iu/dl

117
47–258

132
55–298

128
50–206

204
68–360

169
86–466

240
100–544

247
97–630

Mean and 2 standard deviation normal ranges. From a cross sectional study of 239 women, each of whom was only sampled once.
Adapted from ref. 11.
RCo: Ristocetin cofactor activity.

Table 1.3 Natural anticoagulant factors during pregnancy and the early puerperium

6–11
weeks
N = 41

12–16
weeks
N = 28

17–23
weeks
N = 10

24–28
weeks
N = 19

29–35
weeks
N = 36

36–40
weeks
N = 23

3 days
post-natal
N = 87

Total Protein S
u/dl

80
34–126

77
45–109

66
40–92

68
38–98

67
27–106

58
27–90

69
37–85

Free Protein S
u/dl

81
47–115

72
44–101

64
38–90

60
34–86

54
32–76

57
15–95

58
29–87

Protein C activity
u/dl

95
65–125

94
62–125

101
63–139

105
73–137

99
60–137

94
52–136

118
78–157

Antithrombin activity
u/dl

96
70–122

100
72–128

100
74–126

104
70–138

104
68–140

102
70–133

108
77–137

Mean and 2 standard deviation normal ranges. From a cross sectional study of 239 women, each of whom was only sampled once.
Adapted from ref. 11.

Natural anticoagulants
6

There are changes in the balance of the natural anticoagulants during pregnancy and the puerperium
(Table 1.3). Levels and activity of Protein C do not
change and remain within the same ranges as for non-

pregnant women of similar age.11 There are increased
levels and activity of Protein C in the early puerperium. Total and free (i.e. biologically available) Protein S levels decrease progressively through gestation.
Ranges for total and free Protein S are lower in the

Chapter 1. Normal changes

Table 1.4 Natural anticoagulants and markers of fibrinolysis

Number of patients
Weeks

41
11–15

48
16–20

47
21–25

66
26–30

62
31–35

48
36–40

61 Postdelivery

61 Postnatal

Fibrin degradation
Products ␮g/ml

Mean

1.07

1.06

1.09

1.13

1.28

1.32

1.66

1.04

Fibrinolytic activity
(100/Lysis time)

Mean

7.6

7.4

7.3

5.5

4.5

5.6

6.75

5.75

Lysis time in hours

Mean

13.25

13.5

13.75

18.25

22.25

17.8

14.8

17.4

Antithrombin III:C

Mean
Range

85
49–120

90
46–133

87
42–132

94
47–141

87
42–132

86
40–132

87
48–127

92
38–147

Antithrombin III:Ag

Mean
Range

93
60–126

94
56–131

93
56–130

97
56–138

96
59–132

93
50–136

95
58–133

100
64–134

␣ 1 Antitrypsin

Mean
Range

124
66–234

136
86–214

125
53–295

146
85–249

149
89–250

154
91–260

172
84–352

77
44–135

␣ 2 Macroglobulin

Mean
Range

176
100–309

178
98–323

170
92–312

160
88–294

157
85–292

153
85–277

146
81–265

142
82–245

Where no units are shown, values are expressed as per cent of standard. Where shown, range is 2.5th–97.5th centile. Samples were collected
longitudinally from 72 women. Post-natal samples were collected 2 weeks-12 months following delivery. The post-natal values were found
to be similar to those obtained from healthy pre-menopausal women who were not using oral contraceptives.
Adapted from ref. 10.

first trimester (34–126 and 47–115 iu/dL, respectively)
than in women of similar age, not using oral contraceptives (64–154 and 54–154 iu/dL, respectively).11
This makes it difficult to diagnose Protein S deficiency
in pregnancy. Antithrombin levels and activity are usually stable during pregnancy, fall during labor and rise
soon after delivery (Tables 1.3 and 1.4).
Acquired activated Protein C (APC) resistance
has been found in pregnancy, in the absence of Factor V Leiden, antiphospholipid antibodies or a prolonged APTT.11 This has been attributed to high
Factor VIII activity and may also be influenced by
high Factor V activity and low free Protein S levels. Similar acquired APC resistance has been found
in women using oral contraceptives and in association with inflammatory disorders. The changes in APC
resistance with gestation preclude use of APC sensitivity ratios as a screening test for Factor V Leiden during
pregnancy.

Summary points
r
r
r
r

Protein C is unchanged in pregnancy.
Protein S decreases in pregnancy.
Antithrombin levels decrease during labor.
There is acquired APC resistance during
pregnancy.

Fig. 1.1 Thromboelastograph analyzer.

Thromboelastography
Thromboelastography (TEG)(Fig. 1.1) provides an
overall assessment of coagulation by measuring the

Section 1. Cellular changes

viscoelastic properties of whole blood as it is induced
to clot in a low-shear environment. The parameters
derived from the automated TEG equipment define
the reaction time to initiation of a clot (R), the clot
formation rate (␣) and time (K), the clot strength or
maximum amplitude (MA) and clot lysis (reduction
in maximum amplitude after 60 minutes, LY60) (Fig.
1.2). The various parameters are correlated and are
affected by the availability of fibrinogen and platelet
function. The TEG coagulation index (TEG CI) is
derived from R, K, MA, and ␣, which has a normal
range of −3 (hypocoagulability) to +3 (hypercoagulability).
In healthy late pregnancy, there is increasing hypercoagulability and the TEG CI has been measured in
the range −0.6 to +4.3. Within the first 24 hours of
delivery, TEG CI values of −0.5 to +3.9 have been
found.12 The highest TEG CI values have been found
during active labor. Parameters return to baseline by 4
weeks postpartum13 (Fig 1.3). No differences have been
found in TEG parameters during pregnancy between
smokers and non-smokers. Significantly lower TEG
CI values were found in a large study of women who
took folic acid supplements14 during the first trimester
(−1.22 to +2.87), indicating that they were less
hypercoagulable than those who did not take supplements (−1.52 to +2.60).
Studies of TEG in pregnant women with thrombocytopenia are inconclusive to date. The TEG MA correlates with platelet count as well as fibrinogen, but it is as
yet unclear whether TEG parameters can be used clinically to predict the safety of regional anesthetic techniques in women with low platelet counts, especially
those with pre-eclampsia.8,9

Summary points

r TEG gives a global assessment of coagulation
status.
r TEG CI measurement demonstrates the tendency
to hypercoagulability in pregnancy.
r There is insufficient experience with TEG in
pregnant women with thrombocytopenia or
pre-eclampsia to judge its clinical usefulness.

Markers of hemostatic activity
8

Hemostatic activity can be assessed by measuring
markers of both clot formation and clot destruc-

tion.15 Many have been used in research settings,
but the ones that have clinical applications are
thrombin–antithrombin complexes (TAT) and prothrombin fragments (F 1+2), which reflect in vivo
thrombin formation, plus tests that demonstrate plasmin degradation of fibrin polymer to yield fragments,
namely D-dimers and fibrin degradation products
(FDP). Exact reference ranges depend on the reagents
and testing kits used for the assays. Increased levels of F 1+2 are shown in Table 1.2; by term, levels are approximately four times higher than those
from a healthy adult population. Likewise, TAT levels15 increase with gestation; in early pregnancy the
upper limit of normal is similar to the adult range of
2.63 ␮g/L, whereas by term, the upper limit of normal
is 18.03 ␮g/L.
D-dimer levels are very markedly increased in
pregnancy, with typical ranges tenfold higher in
late pregnancy than in early pregnancy or the nonpregnant state. In one study,15 where the healthy
adult range for D-dimers was ⬍ 433 ␮g/L, by mid
pregnancy the range was ⬍ 3000 ␮g/L and by
late pregnancy ⬍ 5300 ␮g/L. It is thought that the
increase in D-dimers reflects the increase in fibrin
during pregnancy, rather than increased fibrinolytic
activity.

Summary points

r Markers of thrombin production (TAT and F1+2)
are elevated in pregnancy.
r D-dimers are tenfold higher in late normal
pregnancy than typical levels from healthy
non-pregnant women.

Fibrinolysis
There is additional hemostatic control exerted by
lysis of the fibrin clot. This is achieved by plasmin, created from plasminogen by activators. The
fibrin mesh is lyzed to fibrin degradation products, including D-dimers. Tissue plasminogen activator is the most important endothelial cell derived
plasminogen activator. There is reduction in the
activity of the fibrinolytic system during pregnancy,
mostly due to increased levels of plasminogen activator inhibitors (PAI-1 and PAI-2), which are produced by the placenta. PAI-1 is also produced by
platelets and endothelium. There is an exponential

Chapter 1. Normal changes

(a)

(b)
Fig. 1.2 Thromboelastograph trace (a) pregnant (b) non-pregnant, showing shortened R and K times and increased maximum amplitude in
pregnancy.

increase in PAI-1 with gestation, from typical values ⬍ 50 ␮g/L in early pregnancy and the nonpregnant state, to values 50–300 ␮g/L at term.15 Old
studies of fibrinolytic mechanisms in pregnancy and
the puerperium demonstrated that levels of plasminogen activator decline through pregnancy, reach
their lowest levels during labor and increase soon
after delivery.16 The discovery of PAI-1 and PAI-2

provides the explanation for these changes, which
lead to maximum suppression of fibrinolysis during
labor.
There are a number of inhibitors of plasmin,
including ␣2 antiplasmin, antithrombin, ␣1 antitrypsin, ␣2 macroglobulin and C1 -esterase inhibitor. Levels
of ␣1 antitrypsin and ␣2 macroglobulin increase after
delivery (Table 1.4), as do Factor VIII and fibrinogen

Section 1. Cellular changes

95% CI for the mean
70

P<0.0001

MA (mm)

P<0.0001
P<0.05

60
55
50
Control

3
4
5
Weeks’ postpartum

7-9

10-12

Fig. 1.3 Interval plot of maximum amplitude vs. weeks’ postpartum after normal delivery.

activities (Table 1.2); this is an acute phase reaction,
similar to that seen after surgery. There are also
increased levels of thrombin activatable fibrinolysis inhibitor (TAFI) in pregnancy, which inhibits
fibrinolysis by various mechanisms.17 Overall,
although fibrinolytic activity increases after delivery,
it takes at least 6 weeks to be completely restored to
normal non-pregnant levels.
Clot lysis time is prolonged in pregnancy
(Table 1.4), particularly in the third trimester. In
one study,17 the median and interquartile range
for clot lysis time was 98 (90–111) minutes in the
first trimester, 110 (99–124) minutes in the second
trimester and 127 (107–171) minutes in the third
trimester, but 92 (80–99) minutes in the first 24 hours
after delivery of the placenta.
Increased circulating FDP levels (Table 1.4) and
D-dimers15 are found during pregnancy despite systemic suppression of fibrinolysis. It is thought that
there is increased fibrin generation and degradation
locally in the placental circulation. Differences have
been found in hemostatic and fibrinolytic processes

in blood samples from venous placental blood and
from forearm blood10 . It is also possible that clearance
of FDP and D-dimers may be altered in pregnancy.
Overall, there is a low level of intravascular coagulation, demonstrable from as early as 11–15 weeks
gestation.10 Levels of FDP, D-dimers and soluble fibrin remain high after delivery for at least the first
week.

Summary points

r Fibrinolysis is suppressed during pregnancy and
especially during labor.
r PAI-1 from endothelial cells is increased in
pregnancy.
r PAI-2 is produced in the placenta.
r Various factors continue to suppress fibrinolysis
soon after delivery.
r Raised FDP and D-dimers indicate clot formation
and destruction, possibly locally in the placental
circulation.

Chapter 1. Normal changes

Homocysteine
Homocysteine levels fall in early pregnancy and are
significantly reduced compared to the non-pregnant
state, in all three trimesters.18 This appears to be
multifactorial and related to the hormonal changes

in pregnancy, physiological hemodilution, increased
renal clearance of homocysteine, folic acid supplementation and enhanced remethylation of homocysteine due to increased demands for methionine by the
fetus.

Section 1. Cellular changes

References
1. Taylor DJ, Lind T. Red cell mass during and after
normal pregnancy. British Journal of Obstetrics and
Gynaecology 1979; 86: 364–370.
2. Milman N, Bergholt T, Byg K-E et al. Reference
intervals for haematological variables during normal
pregnancy and postpartum in 434 healthy Danish
women. European Journal of Haematology 2007; 79:
39–46.
3. Edlestam G, Lowbeer C, Kral G et al. New reference
values for routine blood samples and human
neutrophilic lipocalin during third trimester
pregnancy. Scandinavian Journal of Clinical
Laboratory Investigation 2001; 61: 583–592.
4. Valdimarsson H, Mulholland C, Fridriksdottir V
et al. A longitudinal study of leucocyte blood counts
and lymphocyte responses in pregnancy: a marked
early increase of monocyte-lymphocyte ratio.
Clinical and Experimental Immunology 1983; 53:
437–443.
5. Boehlen F, Hohfeld P, Extermann P et al. Platelet
count at term pregnancy: a reappraisal of the
threshold. Obstetrics and Gynecology 2000; 95:
29–33.
6. Fay RA, Hughes AO, Farron NT. Platelets in
pregnancy: hyperdestruction in pregnancy. Obstetrics
and Gynecology 1983; 61: 238–240.
7. Vincelot A, Nathan N, Collert D et al. Platelet function
during pregnancy: an evaluation using the PFA-100
analyser. British Journal of Anaesthesia 2001; 87:
890–893.
8. Beilin Y, Arnold I, Hossain S. Evaluation of the platelet
R
function analyzer (PFA-100
) vs. the
thromboelastogram (TEG) in the parturient.
International Journal of Obstetric Anesthesia 2006; 15:
7–12.
9. Davies JR, Roshan F, Hallworth SP. Hemostatic
function in healthy pregnant and preeclamptic

women: an assessment using the platelet function
R
R
analyzer (PFA-100
) and Thromboelastograph
.
Anesthesia and Analgesia 2007; 104: 416–420.
10. Stirling Y, Woolf L, North WRS et al. Haemostasis in
normal pregnancy. Thrombosis and Haemostasis 1984;
52: 176–182.
11. Clark P, Brennand J, Conkie JA et al. Activated protein
C sensitivity, protein C, protein S and coagulation in
normal pregnancy. Thrombosis and Haemostasis 1998;
79: 1166–1170.
12. Sharma SK, Philip J, Wiley J. Thromboelastographic
changes in healthy parturients and postpartum
women. Anesthesia and Analgesia 1997; 85: 94–98.
13. Maybury HJ, Waugh JJS, Gornall A, Pavord S. There is
a return to non-pregnant coagulation parameters after
four not six weeks postpartum following spontaneous
vaginal delivery. Obstetric Medicine 2008; 1: 92–94.
14. Deol PS, Barnes TA, Dampier K, Pasi KJ,
Oppenheimer C, Pavord SR. The effects of folic acid
supplements on coagulation status in pregnancy.
British Journal of Haematology 2004; 127: 204–
208.
15. Cadroy Y, Grandjean H, Pichon J et al. Evaluation of
six markers of haemostatic system in normal
pregnancy and pregnancy complicated by
hypertension or pre-eclampsia. British Journal of
Obstetrics and Gynaecology 1993;100: 416–420.
16. Bonnar J, McNicol GP, Douglas AS. Fibrinolytic
enzyme system and pregnancy. British Medical Journal
1969; iii: 387–389.
17. Mousa HA, Downey C, Alfirevic Z, Toh C-H.
Thrombin activatable fibrinolysis inhibitor and its
fibrinolytic effect in normal pregnancy. Thrombosis
and Haemostasis 2004; 92: 1025–1031.
18. Walker MC, Smith GN, Perkins SL et al. Changes in
homocysteine levels during normal pregnancy.
American Journal of Obstetrics and Gynecology
1999;180: 660–4.

Section 1
Chapter

Cellular changes

Hematinic deficiencies
Jane Strong

Introduction
Deficiency of any of the vitamins and minerals essential for normal erythropoiesis (hematinics) may be
associated with defective erythropoiesis and anemia.
Hematinics include iron, copper, cobalt, vitamins A,
B12 , B6 , C, E, folic acid, riboflavin, and nicotinic acid.
Iron, folate, and vitamin B12 deficiency are the most
common hematinic deficiencies. These are the focus of
this chapter.

Iron deficiency
Epidemiology
Iron deficiency anemia is the most common health
problem that women face worldwide. It affects about
20% of the world’s population and is a significant cause
of morbidity and mortality. Of anemias diagnosed in
pregnancy, 75% are due to iron deficiency.
On a worldwide perspective, the deficiency in iron
reflects poor nutrition resulting from widespread economic and social deprivation. Many women have
depleted or borderline iron stores due to menstruation and the demands of previous pregnancies, and
few women enter into pregnancy with sufficient iron
stores. Combined with the increased iron demands in
pregnancy due to the expansion in red cell mass and
the requirements of the developing fetus, many women
become iron deficient.
Worldwide, iron deficiency anemia in pregnancy
affects about 50% of women. In developing countries
the prevalence is 56% and in developed countries 18%.
The majority of these women are already anemic prior
to pregnancy. Prevalence studies in the United States
reveal iron store depletion in about 10% of women of
reproductive age, with anemia present in 5%.

The iron deficiency anemia rates in pregnancy
increase with each trimester – starting with 9% in
the first trimester, 14% in the second, and 37% in the
third.
It is of note that it takes 2 years of normal dietary
iron to replace the iron lost with each pregnancy. More
than 500 mg of storage iron are required to avoid iron
deficiency in pregnancy. This amount of storage iron
is present in only 20% of women with 40% having no
storage iron at the start of pregnancy.

Pathogenesis
Iron homeostasis
Dietary elemental iron is absorbed from the duodenum and jejunum. The typical western diet will contain
15 mg/day iron. The recommended daily allowance of
iron for pregnancy is 30 mg/day.
The dietary bioavailability of iron depends on the
iron content of the food and its form. Heme iron,
derived from meat is more readily absorbed than
non-heme iron. Absorption is facilitated by reducing
agents such as vitamin C, hence the recommendation
to take iron supplements with orange juice or ascorbic acid tablets. Absorption is inhibited by phytates
in cereals, tannins in tea and polyphenols in some
vegetables.
Only approximately 10% of dietary iron is
absorbed. This increases in pregnancy and triples
from the first to the third trimester peaking after 30
weeks.
The iron requirements of a pregnancy, labor, and
delivery are approximately 1240 mg (see Table 2.1).
Iron requirements in pregnancy rise sharply from
1–2 mg/day in the first trimester to 4 mg/day in the second trimester and peaking at 6 mg a day in the third
trimester. Lactation requires 0.5–1.0 mg/day of iron.

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

Section 1. Cellular changes

Table 2.1 Iron requirements for pregnancy, labor, and delivery

Source of increased iron requirement Iron demand
Increase in red cell mass

450 mg

Fetus and placenta

300 mg

Increase in basal maternal requirements

240 mg

Blood loss at delivery (normal vaginal
delivery)

250 mg

Iron requirements for pregnancy, labor,
and delivery

1240 mg

Absorption is regulated by the gastrointestinal
tract and is dependent on iron stores. In normal pregnancy a physiological hypervolemia occurs and this
results in a modified response to blood loss. The
plasma volume increases from 6 weeks gestation by
50%. The red cell mass has a slower rate of expansion.
By term, it has increased by 25%, but this is dependent
on iron status.
Iron is required for the red cell expansion and ferritin levels show a marked decline between 12 and
25 weeks. This results in a physiological reduction
in hemoglobin concentration that is maximal at 32
weeks. Hemoglobin concentrations return to normal
within 1 week in the postpartum period in iron-replete
women.
The increase in blood volume helps to compensate
for blood loss at delivery. A blood loss of 1000 ml can
be tolerated without a significant drop in hemoglobin.
Provided the blood loss at delivery does not exceed
25% of the pre-delivery blood volume, there is no
further increase in blood volume. The plasma volume decreases as a result of diuresis, the hematocrit increases, and the blood volume returns to nonpregnant values.

The placental regulation of iron transfer
to the fetus

The apical surface of the placental syncytiotrophoblast
has transferrin receptors that trap maternal transferrin by endocytosis, and the iron is bound to halotransferrin within the placental cell. Iron is released, bound
to ferritin within the placenta, and then actively transported to the fetus initially as fetal apotransferrin and
then as holotransferrin in the fetal circulation.
If maternal iron decreases, the placental tranferrin receptors increase and conversely placental iron
uptake is inhibited by placental synthesis of ferritin.
Transfer of iron to the fetus occurs predominantly in

the last 4 weeks of pregnancy. Two-thirds of fetal iron
is found in the fetal hemoglobin, the rest in the fetal
liver.
Maternal iron deficiency anemia affects both
mother and fetus. Iron-dependent enzymes in every
cell are affected and there are neuromuscular, gastrointestinal, and epithelial consequences that can influence
fetal mortality, growth, and programing.

Diagnosis of iron deficiency
Iron deficiency develops sequentially, with storage iron
becoming depleted initially. This is followed by a fall in
the amount of iron available for erythropoiesis. Subsequently, the peripheral blood hemoglobin drops and,
with that, there is a fall in the delivery of oxygen to
peripheral tissues, and the patient develops clinical
symptoms and signs.
Each phase in the development of symptomatic
iron deficiency anemia has various hallmarks outlined
below.

Decrease in storage iron
Tissue and bone marrow iron become deplete first.
Bone marrow samples can be specifically stained to
look for iron. Without iron supplementation, 80% of
women are deplete of iron stores at term with no stainable iron in their bone marrow samples. Although this
is a rapid and reliable method of assessing iron stores,
the invasive nature of the test means it is rarely done
to diagnose iron deficiency as there are several reliable
non-invasive tests. Bone marrow examination is generally reserved for severe anemias when the cause cannot be determined by other means and when there is
evidence of marrow failure.
Serum ferritin levels fall early in the development
of iron deficiency. This is one of the first abnormal laboratory tests. Ferritin levels are not affected by recent
ingestion of iron, but they are an acute phase reactant
rising if there is active infection or inflammation.
Transferrin levels increase early in the development of iron deficiency, but are rarely available as a laboratory measure. This transporter protein increases
in an attempt to deliver more iron to the tissues.

Decrease in iron for erythropoiesis
Serum transferrin receptors are transmembrane proteins present in all cells. They bind transferrin-bound
iron and transport it to the cell interior. Receptors
increase as the iron supply decreases. Small amounts

Chapter 2. Hematinic deficiencies

of transferrin receptors circulate in the plasma in
amounts proportional to the total. These soluble transferrin receptors can be measured by immunological
techniques. This test is reported as being 100% specific
in identifying iron deficiency in pregnancy and has significant advantages over ferritin and transferrin saturation.
Once tissue iron deficiency is established, serum
transferrin receptors increase in proportion to the
degree of iron deficiency. Serum transferrin receptor
level changes occur before a reduction in the mean
corpuscular volume (MCV) and mean corpuscular
hemoglobin concentration (MCHC) in red cells and
also before the rise in free erythrocyte protoporphyrin.
A reduction in MCV and MCHC are seen at an
early stage in the development of iron deficiency in
the non-pregnant state, but these are a poor indicator
of iron deficiency that develops during pregnancy. The
increased drive to erythropoiesis resulting in the physiological increase in red cell mass means that there are
a higher proportion of young large red cells and this
can mask the effect of iron deficiency on red cell MCV.
A normal MCV does not exclude iron deficiency and
the red cell indices in established iron deficient women
in pregnancy may be normochromic normocytic.
Iron replete pregnancies are associated with a physiological increase in red cell size – usually around 4fL
(femtoliters – 10−15 L).
Free erythrocyte protoporphyrin increases as iron
for erythropoiesis reduces. Iron addition to the porphyrin ring is the last step in heme biosynthesis. When
iron is low, free protoporphyrin increases. Zinc competes with iron and, if iron is unavailable, zinc protoporphyrin levels increase and these can also be measured. Both free erythrocyte and zinc protoporphyrin
increase in situations of acute infection or inflammation. These measurements are also elevated in lead poisoning.

increased by 50% and the red cell mass by 18%–25%
depending on iron status. These physiological changes
cause a dilutional decrease in hemoglobin and hematocrit. Increased hemoglobin in the second trimester
may represent poor maternal blood volume expansion
and is associated with maternal and fetal morbidity. A
hematocrit above 43% has been associated with a fourfold increased risk of fetal growth retardation.
Iron deficiency is often diagnosed retrospectively
after a good hemoglobin response to a therapeutic trial
of iron supplements. In populations where there is a
possibility of thalassemia that can present with full
blood count features similar to iron deficiency, iron
therapy should only be started after iron deficiency
is confirmed with a measure of iron stores such as
ferritin.

Clinical signs and symptoms
Patients with iron deficiency are often asymptomatic,
but symptoms may occur without an anemia. Irondependent enzymes in every cell are affected and there
are neuromuscular, gastrointestinal and epithelial consequences. Prior to the development of an anemia, the
signs and symptoms of iron deficiency are non-specific
and include reduced exercise tolerance and tiredness.
Severe iron deficiency is associated with pallor,
glossitis, angular chelitis, nail ridging, and when severe
nail spooning – koilonychia. Dsyphagia can develop
if a post-cricoid web occurs. Iron deficiency can
also affect cellular immunity and phagocytosis, with
women being increasingly susceptible to infection.
Pica can occur in as many as 50% of patients as a
symptom of severe iron deficiency and can take different forms – craving for earth, clay, starch, and ice. It
improves with iron replacement (Tables 2.2, 2.3, 2.4).
Table 2.2 Clinical signs and symptoms of iron deficiency

Symptoms

Signs

Iron
deficiency
without
anemia

Irritability
Poor concentration
Tiredness and fatigue
Reduced exercise tolerance

None

Iron
deficiency
with anemia

Tiredness and fatigue
Reduced exercise tolerance
Shortness of breath on exercise
Palpitations
Headache
Dysphagia
Pica

Pallor
Glossitis
Angular chelitis
Koilonychia

Decrease in peripheral hemoglobin
Anemia is defined as a hemoglobin level at least two
standard deviations below the median value for a
healthy matched population. The World Health organization defines anemia in pregnancy as a hemoglobin
below 11 g/dL. Some define a different cut-off in the
second trimester – the United States Centers for Disease Control (US CDC) use a value of 10.5 g/dL.
The maternal blood volume expands in the first and
second trimesters – the plasma volume expansion is

Section 1. Cellular changes

Table 2.3 Effects of iron deficiency

Effects of iron
deficiency

Mother

Fetus and pregnancy outcome

Neonate, infant, and older

See Table 2.2 above
Decreased cognitive function
Tissue enzyme malfunction
Effects on neuromuscular
transmission

Hb≤ 9 g/dL – increased risk of:
• Prematurity (doubles risk)
• Small for gestational age
• Spontaneous abortion

Lower Apgar scores
Low iron stores in newborn
associated with growth
restriction, neurological
and mental impairment

Low ferritin:
Placental hypertrophy – increase in
angiogenesis

Increased placenta:fetal ratio is
a predictor of cardiovascular
disease and diabetes in adult life

Table 2.4 Laboratory investigations in iron deficiency

Laboratory test and normal
non-pregnant female
range
Normal – pregnancy

Iron deficiency without Iron deficiency with
anemia – iron store
anemia –
depletion
Mild – severe

Bone marrow reticuloendothelial
iron 2+ – 3+

2+–3+
Difficult to maintain by third trimester
without iron

None

Serum iron
60–150 mcg/dL

⬎60 mcg/dL progressive fall over
pre-pregnancy values

Borderline low

low

Transferrin
200–400 mg/dL

Progressive rise over pre-pregnancy
values – within normal range

Borderline high

Raised

Saturation
SI/TIBC: 20%–50%

Progressive fall within normal range

Normal

Low

Plasma or serum ferritin
40–200 ␮g/L

Decreases within normal range between
12th and 25th week (hemodilution)

⬍40

⬍20 (mild)–⬍10 (severe)

Soluble transferrin receptors
2.9–8.3 mg/L

First trimester –2.6–6.7 mg/L
Second and third trimester – 25%
increase (increased erythropoiesis)

Increased

Red cell indices and red cell
morphology

MCV can rise: average 4–6 fL

Normal

Mild hypochromia and
microcytosis

Erythrocyte protoporphyrin
30–70 ng/mL

Progressive rise, usually within normal
range

30–70

⬎100 – 200

Hemoglobin 12–15 g/dL

⬎11 first and third trimesters
⬎10.5 second trimester

Normal

9–12 (mild), 6–7(severe)

Other tissue changes

None

Nail/epithelial changes

Management options
Iron
Iron is available in a variety of forms – dietary, tablet,
and liquid, intravenous and intramuscular.

Dietary iron
16

In pregnancy it is recommended that iron consumption is increased by 15 mg/day to a daily recommended allowance of 30 mg/day. Women will often
find it difficult to increase dietary iron sufficiently, but

these recommended amounts are met by most prenatal vitamin formulations.
Dietary iron is predominantly in the reduced ferric form (Fe3+ ) and this is poorly soluble above a pH
of 3. It is poorly absorbed at the duodenal pH of 7–
8. The oxidized ferrous form of iron (Fe2+ ) is more
soluble at the duodenal pH and hence more easily
absorbed.
Heme dietary sources of iron – meat, fish, and
poultry have a much greater bioavailability than nonheme vegetable sources. Iron bioavailability from
heme sources is approximately 30% vs. 10% for nonheme sources.

Chapter 2. Hematinic deficiencies

Table 2.5 Iron absorption

Enhanced iron
absorption

Reduced iron iron absorption

Ascorbic acid

Phytates in bran, oats, rye, and fiber

Heme iron

Tannins in tea

Oxidized, ferrous form of
iron (Fe2+ )

Polyphenols in some vegetables
High dietary calcium content

Intraluminal factors in the gastrointestinal tract
also affect absorption (Table 2.5).

Tablet and liquid iron
Iron can be given to supplement dietary iron and
maintain iron stores at a time of marked increased
iron demand. Most studies report that this approach
decreases the prevalence of iron deficiency anemia
at delivery. This may help anemia in infancy, but
it is unclear whether iron supplementation in wellnourished non-anemic women improves birth outcome.
It can be given selectively based on a measure of
iron stores or routinely. The need for iron supplementation in Western countries is debatable, but the practice is recommended in the developing world. The
World Health Organization (WHO) recommend universal oral iron supplementation with 60 mg elemental
iron daily for 6 months in pregnancy in areas where
the prevalence of iron deficiency is less than 40%. The
supplementation is continued for 3 months postpartum in areas where the prevalence is greater than 40%.
The Center for Disease Control and prevention recommends supplementation with 30 mg elemental iron
daily as does the American College of Obstetricians
and Gynecologists.
Universal supplementation is considered practical
and cost effective by some. The debate is ongoing. A
recent Cochrane database library review demonstrated
no definite advantage to mother or fetus with routine
iron or iron and folate supplementation.1
Women with iron deficiency anemia should receive
iron supplements of 30–120 mg elemental iron until
the anemia is corrected and there has been time for
iron stores to replenish. Oral iron is an effective, cheap,
and safe way of replacing iron, provided there is compliance.
There are a large number of oral iron-containing
preparations and they often come combined with other
vitamins and minerals. As a general principle, enteric

coated or slow release formulations should be avoided
as the iron is released beyond the duodenum and proximal jejunum where it is maximally absorbed. Women
should be counseled regarding diet and the factors that
can inhibit iron absorption. Iron salts should ideally
not be given with food because the phytates, tannins,
and phosphates within the diet can bind iron preventing its absorption. Antacids should also be avoided
around the ingestion of iron and ideally ascorbic acid
should be taken to enhance absorption.
The iron preparation of choice is based on effectiveness and minimal side effects. The three ferrous salts
available are ferrous fumurate, ferrous gluconate, and
ferrous sulphate. They each contain differing quantities of elemental iron:
r ferrous fumarate – 65 mg elemental (ferrous) iron
per 200 mg tablet
r ferrous sulphate – 60 mg elemental iron per
300 mg tablet
r ferrous sulphate, dried – 65 mg elemental iron per
200 mg tablet
r ferrous gluconate – 35 mg elemental iron per
300 mg tablet.
The recommended oral dose of elemental iron for
the treatment of iron deficiency is 100–200 mg daily.
Ferrous sulphate 200 mg three times daily provides
195 mg elemental iron and, on this treatment, regimen
the hemoglobin should rise 2 g/dL over 3–4 weeks.
Once the hemoglobin has normalized, the treatment
should be continued for a further 3 months to replenish the iron stores.
Side effects are experienced in 10%–20% of patients
at treatment doses. Iron salts irritate the gastrointestinal tract and can cause nausea, vomiting, epigastric
discomfort, and altered bowel habit (constipation or
diarrhea). There appears to be a clear dose relationship
with the upper gastrointestinal symptoms, but this is
less clear with the altered bowel habit.
If side effects occur, an iron preparation containing
a smaller dose of iron can be tried. Liquid preparations
can be useful, allowing patients to titrate their dose to
a level where side effects are acceptable. Iron can be
taken with meals, but this will decrease the amount
absorbed.

Parenteral iron
Parenteral iron therapy is available as iron dextran or
sucrose. It is reserved for patients unable to tolerate

Section 1. Cellular changes

oral iron or where compliance is in doubt or in patients
where there is a level of bleeding that exceeds the ability of the GI tract to absorb iron or there is malabsorption. It should be noted that parenteral administration
does not produce a faster response than correctly taken
oral iron that is absorbed adequately. It merely ensures
compliance. First trimester administration is not recommended.
There are currently two well-established preparations approved for use in the UK:
R
r iron dextran (Cosmofer
) – a complex of ferric
hydroxide with dextran containing 50 mg of
elemental iron/ml that can be given either
intramuscularly or intravenously;
R
r iron sucrose (Venofer
) – a complex of ferric
hydroxide with sucrose containing 20 mg of
elemental iron/ml that is approved for intravenous
use.
Dose is calculated according to body weight and
R
has the advantage of being
iron deficit. Cosmofer
licensed for administration as a single total dose infusion. Anaphylactoid reactions can occur with parental
iron preparations and a test dose is recommended
prior to the first dose. Cardiopulmonary resuscitation facilities should be available with injectable
1:1000 adrenaline solution, antihistamines, and corticosteroids.
Iron dextran has safety issues related to anaphylaxis. The high molecular weight dextran moiety is
thought to share antigens with gastrointestinal organisms. Much of the reported experience with this drug
is in hemodialysis patients. The safety of intravenous
iron dextran has been reviewed in 573 hemodialysis
patients:2
r 4.7% had an adverse reaction.
r Ten patients (1.7%) had reactions classified as
anaphylactoid including cardiac arrest in 0.2%,
chest pain1%, and hypotension 0.5%.
r There were no deaths.
r Only in 4 of the 10 with anaphylactoid reactions
did these occur during the test dose
administration, emphasizing the need for
vigilance.
r Drug allergies were strong predictors for
reactions.

The iron dextran SPC report severe anaphylactoid
reactions as being very rare ⬍1/10 000.

Iron sucrose appears to be safe even amongst
those with a prior history of sensitivity to iron dextran. Again, the experience comes from hemodialysis patients. A group of 665 patients including 80 with
previous iron preparation intolerance experienced no
adverse reactions to iron sucrose.3
The next generation of parenteral iron has recently
become available:
R
r Ferric carboxymaltose (Ferinject
) – contains
50 mg of elemental iron/ml that can be given
intravenously.
Dose is also calculated according to body weight
and iron deficit. It is contraindicated in the first
trimester of pregnancy. Clinical data on pregnant
women are not currently available and the SPC
advises a careful risk/benefit evaluation prior to
use in pregnancy. The appeal of this new product
includes significantly reduced infusion times and no
requirement for a test dose. Adverse events from
pooled data from 10 multicenter trials involving 2800
patients reported no serious or life-threatening hypersensitivity (anaphylactic) events, but as with other
parenteral iron preparations the SPC warns that
facilities for cardiopulmonary resuscitation must be
available.

Intramuscular iron
Iron administration given by deep intramuscular
injection into the gluteal muscle. This route is often
painful, can stain the skin and the mobilization of iron
from intramuscular sites is slow and often incomplete.
There have been difficulties in sourcing the intramuscular preparation and its administration has become
less popular.

Erythropoietin
Recombinant human erythropoietin is widely used for
anemia associated with chronic renal failure, malignancy, and cytotoxic chemotherapy. It has been used in
difficult anemia cases in pregnancy. The widest experience is with Jehovah’s witnesses. It does not cross
the placenta, but carries a risk of hypertension and
thrombosis. Currently, its role in the treatment of
maternal anemia or to increase the yield in autologous or salvage techniques in pregnancy is not well
established. The usual dose is of 50–200 IU/kg subcutaneously two to three times weekly, usually along
with supplemental iron. Trials of intravenous iron with

Chapter 2. Hematinic deficiencies

Table 2.6 Summary of treatment options in iron deficiency

Treatment
option

Indication

Dose

Advantages

Disadvantages

Oral iron

Standard
treatment

30–120 mg elemental
iron/day until anemia
corrected and stores
replenished

Cheap
Easy to administer

Low bioavailability
Poorly tolerated
Frequent side effects
Often poor compliance

Intravenous
iron

Non-compliance
or intolerance

Calculated according to
Fast Efficient
Anaphylactoid reactions (see
body weight and iron deficit Ensures compliance
text)
Reduced need for blood transfusions

Intramuscular
iron

Non-compliance
or intolerance

Calculated according to
Ensures compliance
Pain, abscess, skin
body weight and iron deficit Reduced need for blood transfusions pigmentation at injection site
Difficult to source product

Erythropoietin Specialist
sub-groups of
patients

Blood
transfusion

50–200 IU/kg sc 2–3
times/week

Emergency
Assessment based on
treatment in acute volume lost and
hemorrhage
hemoglobin

or without recombinant erythropoietin conclude that
intravenous iron therapy is the first-line treatment
in resistant iron deficiency anemia but that erythropoietin may be considered in severe anemia requiring rapid correction in patients who do not respond
to intravenous iron alone. The hemoglobin rise with
combination therapy is quicker than with parenteral
iron alone. Median duration of therapy in the combined treatment group was 18 days vs. 25 days in
the group treated with i.v. iron alone. It has also
been used effectively in the setting of postpartum anemia.4,5 Further assessment of clinical benefit and costeffectiveness is required.

Blood transfusion
Blood transfusion should be avoided if possible. Transfusion should be reserved for acute hemorrhage. In
chronic iron deficiency, transfusion is not indicated.
There are circumstances when women with severe iron
deficiency are not detected until just prior to delivery
and there is not enough time for iron in any form to
raise the hemoglobin. Transfusion may be required in
these circumstances and this is regrettable as it reflects
lack of antenatal surveillance and action.
Transfusion has many known risks, including
transmission of viruses and bacteria, immunomodulation and increase of post-operative infections, morbidity, and mortality. The possibility of transmitting

Useful adjunct in difficult cases and
where blood transfusion prohibited
(e.g., Jehovah’s witnesses)

Hypertension
Pure red cell aplasia
Clinical benefit and
cost-effectiveness not well
established

Fast rise in hemoglobin

Risks of infection,
contamination, reaction,
antibody formation

prions is an increasing concern. Furthermore, confidential reporting systems indicate that human error
resulting in incorrect blood being transfused is still the
commonest serious hazard of transfusion.
Top-up transfusions are inappropriate especially
in a patient group that are young, essentially well
and undergoing a physiological process known to be
demanding on iron stores. Transfusion remains the
emergency treatment for acute hemorrhage.
To reduce transfusion in this group of patients,
anemia prevention strategies during pregnancy and
peripartum are required, including antepartum
screening, monitoring, and iron supplementation and
treatment when necessary, to optimize pre-delivery
hemoglobins. There should also be strategies to
minimize parturition hemorrhage.
Despite the high incidence and burden of disease
associated with iron deficiency, good-quality studies
evaluating clinical, maternal, and neonatal effects of
iron administration in pregnant women with anemia
are lacking.6 (Tables 2.6, 2.7).

Prevention strategies
Iron supplementation for all
Iron supplementation is a controversial issue in pregnancy. Despite the precarious and often depleted
iron stores during and after pregnancy, iron

Section 1. Cellular changes

Table 2.7 Summary of the iron formulations available

Oral iron – tablets
Oral iron – syrups
Oral iron – over-thecounter
preparations

Intravenous iron

Intramuscular iron

Formulations available

Elemental iron content

Prophylactic
dose

Ferrous fumarate
Ferrous sulphate
Ferrous sulphate, dried
Ferrous gluconate
Ferrous fumurate –
ferasamal
Sodium Feredetate – sytron
Spatone – iron rich water
straight from source

65 mg/200 mg tablet
60 mg /300 mg tablet
65 mg /200 mg tablet
35 mg /300 mg tablet
45 mg/5 mL

1 tablet daily
1 tablet daily
1 tablet daily
2 tablets daily
5 mL 3 x daily

27.5 mg/5 mL
5 mg/10 mL sachet
26% to 40% bioavailability
compared to an average
5%–20% from food and 3%
–10% from iron pills

5 mL 3 x daily
1 sachet daily

50 mg of elemental iron/ml
20 mg of elemental iron/ml

N/A

Calculated according to
body weight and iron
deficit: consult
product literature

50 mg of elemental iron/ml

N/A

Calculated according to
body weight and iron
deficit: consult
product literature

R
Iron dextran (Cosmofer
)–
a complex of ferric
hydroxide with dextran
R
)–a
Iron sucrose (Venofer
complex of ferric hydroxide
with sucrose
R
)–
Iron dextran (Cosmofer
a complex of ferric
hydroxide with dextran

supplementation in non-anemic women has not
been shown to improve pregnancy outcome.
The Cochrane library database has reviewed the
effects of routine oral iron supplementation with or
without folic acid during pregnancy. Forty trials with
a total of 12 706 women were reviewed. Iron supplementation with or without folic acid does reduce the
number of women with a hemoglobin less than 10 g/dL
in late pregnancy, at delivery and 6 weeks postpartum,
but there are no clear conclusions regarding clinical
outcomes for mother, fetus, or neonate.1
Routine iron and folic acid supplementation is
recommended by international organizations in areas
where there is a high prevalence of anemia.
Selective iron supplementation is the approach
adopted in most industrialized countries. Assessment
of iron stores, usually with a ferritin level in the first
trimester, identifies women with low or depleted iron
stores and these women are the ones given iron supplements. In the United States the Centers for Diseases Control and Prevention and the American College of Obstetricians and Gynecologists recommend
routine iron supplementation with lower doses of
elemental iron (30 mg/day) as a primary prevention
intervention.
There are several concerns about iron supplementation in iron-replete women. These include hemoconcentration leading to impaired placental circula-

Treatment dose
1 tablet twice daily
1 tablet 2–3 x daily
1 tablet 2–3 x daily
4–6 tablets daily in
divided doses
10–20 mL twice daily
10 mL 3 x daily
2 sachets daily

tion and fetal growth, the production of free radicals,
and oxidative damage and the risk of iron overload
in women with hemochromatosis. Iron-replete women
given iron do not increase their hemoglobin levels, and
hemoglobin concentration in these women is a reflection on the degree of plasma volume increase. Ineffective plasma volume expansion is predictive of poor
pregnancy outcome.
Claims that iron causes a variety of chronic diseases
and birth defects have not been substantiated. Accidental ingestion of iron supplements by children is a
potential hazard, however. Although most fatal cases
involve ingestion of 2–10 g of iron, as little as 1–2 g
can cause death in young children. Women should be
made aware of this and advised to keep iron tablets well
out of the reach of young children.

Screening for iron deficiency
In industrialized countries screening of women for
iron deficiency anemia is done by measuring the
hemoglobin concentration at booking, 28 weeks and
again at 36 weeks if the 28-week blood result is abnormal. Practice does vary in different countries.
For many clinicians it is difficult to accept that
in well-nourished populations the extra requirements
of pregnancy are not met by a normal mixed diet.
The hemodilution, which occurs in healthy pregnancy,

Chapter 2. Hematinic deficiencies

has encouraged the acceptance of abnormally low
hemoglobin levels as being physiological.
It is my personal opinion, based on experience in
the UK, that iron store depletion without anemia is
not well identified. Women rarely have routine first
trimester ferritin measurements, which are then acted
on to prevent the development of anemia. By the time
an anemia is established in pregnancy, it is more difficult to correct and replenish stores as the iron requirements continue to escalate. Either ferritin screening and selective supplementation needs improvement or universal supplementation should be adopted
as a practical and cost effective approach. Controlled trials do not demonstrate any obvious benefit of iron supplementation, but indirect associations
such as general maternal wellbeing, reduced fetal placental ratio, preterm deliveries, postpartum hemorrhage, and recovery from blood loss at delivery have
not been looked at in an objective manner.

Postpartum anemia
Postpartum anemia is defined as a Hb value less than
10 g/dL and an acute or severe anemia corresponds to
Hb less than 8 g/dL. The prevalence of postpartum anemia varies from 4% to 27%. In industrialized countries,
iron stores are depleted in approximately one-third to
one-half of parturients.
The physiological hemodilution that occurs in
pregnancy protects the mother from blood loss at
delivery, but the 5% of deliveries that have a blood loss
greater than 1 litre can result in a symptomatic anemia
and an increased risk of blood transfusion.
At the present time, there is no consensus on the
management of postpartum anemia, and clinical practice varies. Treatment currently consists of oral iron
therapy and blood transfusions.
The Cochrane library database has reviewed the
treatment for women with postpartum iron deficiency
and concluded that there is some limited evidence
of favorable outcomes for treatment of postpartum
anemia with erythropoietin. Some of the studies suggest improved lactation with this approach. Laboratory hematological parameters improve, but it is not
clear how this relates to clinical outcomes. Six randomized controlled trials were reviewed involving 411
women. All the trials involved erythropoietin. The
authors state that further high-quality trials assessing
the treatment of post partum anemia with iron supple-

mentation (e.g., intravenous administration of iron)
and blood transfusions are needed.4

Summary – key points
Reference ranges in pregnancy are different from the
non-pregnant adult female population, but are rarely
quoted on laboratory reports.
Hemoglobin levels fall in pregnancy as a result
of a physiological increase in plasma volume that is
greater than the pregnancy-associated increase in red
cell mass.
Iron stores are exhausted by the end of pregnancy
in the majority of women unless iron is given.
Iron deficiency accounts for over 90% of anemia
during pregnancy, therefore iron should be the mainstay of therapy.
Anemia affects quality of life and virtually all
organs.
Maternal anemia influences mortality, fetal
growth, premature death in utero, and fetal programing.
Anemia is screened for in pregnancy at booking,
28weeks and possibly 36 weeks (if the 28-week test
result is low). The frequency of testing is dependent on
the country of care – France, for example, carries out a
hemoglobin level at every pregnancy visit.
Anemia screening is primarily done by measuring
the hemoglobin. Further investigation is usually a ferritin level or a trial of iron.
All tests assessing iron status have to be assessed in
the light of gestational changes.
Oral iron is the usual first-line treatment in iron
deficiency.
Parenteral forms of iron are of use if there is nontolerance or non-compliance of oral iron.
Blood transfusion should not be given as top-ups
for iron deficiency anemia.
Peripartum transfusions are often inappropriate
(in 2004 one series put the rate at 32%).
The prevalence of postpartum anemia varies
between 4% and 27%.
Iron stores are depleted in 33%–50% of parturients
in industrialized countries (Fig. 2.1)

Folate deficiency
Epidemiology
Folate deficiency has a prevalence of less than 5% in
developed countries and a very low prevalence where

Section 1. Cellular changes

BOOKING First trimester FBC

Normal result
Abnormal result

Hb ≥ 11

Abnormalities of white cells
and/or platelets in addition to
low Hb or in isolation

↓ Hb only
Repeat Hb
At 28/40

Normal
Hb ≥ 10.5

Discuss with senior
obstetric/hematology staff

Check hemoglobinopathy
screen
If positive check ferritin
prior to treatment with iron

Abnormal
Hb <10.5

Consider referral to
hematology/obstetric clinic
Hb < 11
Hb > 9

Hb < 9
Repeat Hb
At 36/40

Normal

Abnormal

Hb ≥ 10.5

Hb < 10.5

Send ferritin
Start treatment
doses of iron
whilst awaiting
results, give
dietary advice

Start iron and
folate
supplements, e.g.
Pregaday, give
dietary advice

Women at risk of
anemia
Multiple pregnancies
Multiparity and
pregnancy recurring
after a short interval
Previous iron deficiency
anemia

Repeat approximately 2−4
weekly depending on
hemoglobin and timing in
pregnancy

NB ? require permission fig
4.1 hematology in Pregnancy
chptMidwife’s guide to
antenatal investigations A
Sullivan, L Kean, A Cryer

Response

Continue

Non-response

•
•
•

Check compliance
Investigate and treat causes
Refer to consultant for individual
management plan

Fig. 2.1 Algorithm summarizing the management of ante-natal hemoglobin level – haematology in pregnancy, chapter in Midwife’s Guide
to Antenatal Investigations, A Sullivan, L Kean, A Cryer.

Chapter 2. Hematinic deficiencies

there is food fortification with folic acid. Worldwide
folate deficiency is far more common and may complicate one-third of pregnancies. It is a reflection of nutritional status.

Red cell folate levels may show a slight downward trend
in pregnancy, but recover by 6 weeks postpartum. Normal references in pregnancy have not been established
and standard adult reference ranges quoted on laboratory reports are not applicable in pregnancy. 7

Pathogenesis
Folate is a water-soluble B vitamin. It cannot be synthesized by humans, but is found in a wide variety of food
sources, including leafy green vegetables, liver, citrus
fruits, nuts, bread, and dairy produce. Folate is heat
labile and it is often lost in the cooking process. It is
absorbed mainly in the jejunum and then taken up by
the liver. Folate stores last several months.
There are various disorders and factors that cause
or exacerbate folate deficiency including malabsorption, hemolysis (particularly congenital red cell disorders and hemoglobinopathies), myeloproliferative disorders, and anticonvulsants.

Diagnosis
Full blood count and blood film
B12 and/or folate deficiency cause a megaloblastic
anemia. This is usually suspected by the presence
of macrocytic red cells. Megaloblastic erythropoiesis
requires a bone marrow to demonstrate large developing red cells with nuclear cytoplasmic asynchrony and
giant metamyelocytes. In practice, this is rarely necessary. In pregnancy, interpretation of MCV can be more
difficult due to the physiological increase in red cell
size and the increased likelihood of an additional iron
deficiency anemia that may reduce the MCV. Blood
film examination can provide useful diagnostic clues.
Features suggestive of a megaloblastic anemia include
hyper-segmented neutrophil nuclei (more than five
segments), oval macrocytes and mild leukopenia, and
thrombocytopenia in severe cases. If iron deficiency
co-exists with a megaloblastic anemia, the blood film
will be dimorphic with a mixture of large and small red
cells (macrocytic and microcytic cells).

Hematinic assays
Red cell folate assays give an indication of overall body
tissue levels and are better than serum folate levels
that are affected by recent diet and fluctuate significantly from day to day. Even so, red cell folate does
not have good sensitivity or specificity in pregnancy.
Serum and red cell folate levels are lower in smokers.

Homocysteine levels
Homocysteine is the precursor to methionine in the
remethylation cycle and increases in B12 or folate deficiency, as both are required as cofactors. This indirect
measurement is a sensitive marker for folate deficiency.
Throughout pregnancy, plasma homocysteine levels
are lower than non-pregnant controls. The lowest levels are in the second trimester but may rise slightly
in the third trimester. Normal references in pregnancy
have not been established and standard adult reference
ranges quoted on laboratory reports are not applicable
in pregnancy. 7

Bone marrow
Megaloblastic erythropoiesis is demonstrated by the
finding of large erythroblasts and giant abnormally
shaped metamyelocytes. Although this is a rapid and
reliable method of assessment, the invasive nature of
the test means it is rarely done to diagnose folate deficiency as there are several reliable non invasive tests.
Bone marrow examination is generally reserved for
patients with pancytopenia.

Management
Prophylaxis of folate deficiency
Mild folate deficiency can be associated with neural
tube defects. Folic acid supplementation at a dose of
400 ␮g/day is recommended 3 months prior to conception and throughout the first trimester. Periconceptual folic acid reduces the incidence of neural tube
defects by 70%. If women have had a child affected by a
neural tube defect, the recommendation is for a higher
dose of folic acid periconceptually – at least 5 mg/day.
Folate deficiency is also an independent risk factor
for thrombosis.
Folate prophylaxis is required in those with an
increased red cell turnover seen in inherited and
acquired red cell disorders and also for those on anticonvulsants. Currently, doses of 5 mg/day are used for
these indications (Table 2.8).

Section 1. Cellular changes

Table 2.8 Folate requirements

Folate demand/
Folate supply requirement
Typical Western
diet

250 ␮g/day

Average daily
requirements
(non-pregnant)

100 ␮g/day

Average daily
requirements
(pregnant)

400 ␮g/day
increased folate
metabolism
transfer of folate from the
mother to the fetus
(approximately 800 ␮g at
term).

Puerperium

100 ␮g/day from 6 weeks
postpartum
25 ␮g/day in breast milk7

Treatment of folate deficiency
Proven folate deficiency should be treated with folic
acid 5 mg three times a day. Dietary history should be
taken and advice given. Deficiency of B12 should be
excluded as folic acid in these doses can improve the
anemia of B12 deficiency, masking the underlying B12
deficiency and thereby potentially exacerbating a neurological deterioration.

Treatment of hyperhomocysteinemia
This is based on increasing folate, vitamin B6, and vitamin B12 levels. Doses used in treatment are usually
1–5 mg folic acid, 10 mg vitamin B6, and 0.4–1 mg
vitamin B12.

defects by 20% from 37.8 per100 000 live births in
1998 to 30.5 per 100 000 live births currently.8 There
are concerns that, in the effort to reduce the neural
tube defects, other patient groups may have suffered
from the increased levels of folate in the diet. Folic
acid supplementation is potentially harmful in B12
deficiency, where it can mask the anemia leading to
delayed treatment and risk of neuropsychiatric symptoms such as peripheral neuropathy, mood changes,
dementia type syndromes, and posterolateral spinal
cord demyelination (subacute combined degeneration
of the cord). There are also concerns that high intakes
of folic acid may speed up the progression of certain
cancers.
A major objection against folic acid fortification
in the UK is that it requires mass supplementing the
population at large to treat a relatively small target
group of young mothers and that this group should be
targeted by other means. It is estimated that, by adding
folic acid to bread, spina bifida is prevented in 120
babies in the UK every year. For every baby saved, half
a million people, male and female, will have to take the
added folic acid.

Implementing periconceptual folic acid
supplementation
Neural tube closure is complete 4 weeks after conception, when many women are not aware that they are
pregnant and will not have initiated folic acid supplements. Women need to be made aware of the recommendation for folic acid so that they may start it early
when attempting to conceive.

Prevention strategies

Folate food fortification

Summary

The addition of folic acid to bread has been considered
in government committees for years. The fortification
of the nation’s food supply with vitamins and minerals dates back to the post-second world war era. The
British population at this time had significantly poor
nutrition and, as a consequence, diseases due to vitamin deficiencies such as rickets.
In the US, the Food and Drug Administration have
made folic acid food fortification mandatory since
1998. All enriched flour, pasta, rice, and other grain
products contain 140 ␮g of folic acid per 100 grams.
This strategy has reduced the incidence of neural tube

Periconceptual folic acid is advised to reduce the incidence of neural tube defects. Ideally, it should be
started 3 months prior to conception and continued
throughout the first trimester. The dose is 400 ␮g daily
unless there has been a previously affected child in
which case the dose should be at least 5 mg/daily.
Folate prophylaxis should be considered in at risk
groups such as those on anticonvulsants and with
chronic hereditary or acquired red cell disorders.
Folate stores can be depleted within months and
women need education on diet to ensure recommended folate consumption.

Chapter 2. Hematinic deficiencies

B12 deficiency
Epidemiology
Deficiency of B12 in pregnancy is rare. It is usually
associated with infertility. B12 plays a key role in the
development of new tissue; thus women who are deficient may not ovulate, or a fertilized egg may not
develop, resulting in miscarriage
The most common cause of B12 deficiency in the
general population is pernicious anemia and this is
rare in women of childbearing years. Pernicious anemia usually begins after the age of 40 years. Pernicious
anemia is due to lack of intrinsic factor that is required
to bind B12 in the stomach prior to absorption in the
terminal ileum. Other causes of B12 deficiency include
ileal resection, partial gastric resection, Crohn’s disease, tropical sprue.
Dietary deficiency can occur and is most often seen
in vegans who do not eat animal products. Even vegans, however, obtain B12 from bacteria synthesis in the
gastrointestinal tract or on legumes and in marmite.
Maternal cobalamin stores are around 3 mg and the
daily dietary requirement is approximately 3 ug/day.
The developing fetus requires 50 ug/day. It takes about
5 years for a deficiency of B12 to manifest itself clinically because of the stores.9

Table 2.9 Features of B12 and folate deficiency

Folate deficiency
Symptoms Anemia
Very occasionally
neuropsychiatric
symptoms

Signs

B12 deficiency
Gradually progressive
symptoms and signs
of anemia
Neuropsychiatric
symptoms inc.
dementia

Neural tube defects
Mild jaundice may be
Vascular disease
present
(associated with high
Glossitis
levels of homocysteine) Angular stomatitis
purpura due to
thrombocytopenia
Subacute combined
degeneration of the
cord
Rarely optic atrophy

a normal pregnancy outcome but a low B12 level in
the baby especially if the baby is breastfed. This usually
becomes apparent at about the age of 6 months when
the infant fails to thrive, has regression of development
and anemia. Prompt recognition and treatment with
B12 will limit neurological damage.

Diagnosis
See Table 2.9 for clinical signs and symptoms of B12
deficiency.

Pathogenesis
Vitamin B12, also known as cobalamin, is present in
animal-derived foodstuffs such as meat, milk and eggs.
It is required for methionine synthesis and the conversion of methylmalonyl CoA to succinyl CoA. It is
involved in myelin synthesis, protein and DNA synthesis, and fatty acid degradation.
Inadequate B12 levels leads to hyperhomocysteinemia and this in itself can be associated with obstetric
complications. Small subsets of women with recurrent
miscarriages have been found to have elevated homocysteine and is hoped that treatment with vitamins will
reduce levels and prevent pregnancy loss. Low levels
of vitamin B12 have also been found in women with
children with neural tube defects. It is unknown, however, whether vitamin B12 status affects the incidence
of neural tube defects. Meta-analyses have suggested
an association, but methodological differences in the
studies mean it is difficult to draw this conclusion.
There is a correlation between maternal and neonatal B12 levels. Whilst persistent deficiency can lead to
infertility, mild B12 deficiency can be compatible with

Full blood count and blood film
Megaloblastic anemia is the hallmark of B12 deficiency. Blood film examination can be useful. See
above, under diagnosis of folate deficiency.

Hematinic assays
B12 assays give an indication of overall body tissue levels. B12 levels fall in pregnancy, but this is not thought
to represent a true tissue deficiency. It is likely to be
a consequence of increasing maternal plasma volume
and transfer to the fetus. The physiological reduction
can be 30%–50% during pregnancy. Levels tend to be
lower in smokers. The levels return to normal rapidly
after delivery without supplementation. Levels greater
than 130 ng/ml may be considered normal but levels with less than 130 ng/ml with macrocytosis and/or
neurological symptoms should be considered for B12
treatment.
B12 levels can be falsely lowered by folate deficiency that resolves with folate treatment.

Section 1. Cellular changes

Homocysteine and methylmalonic
acid levels
Homocysteine is the precursor to methionine in the
remethylation cycle and increases if B12 and/or folate
are deficient as both are required as cofactors. Methylmalonic acid is the precursor for the conversion of
methylmalonyl-CoA to succinyl CoA. It increases if
there is a deficiency of cobalamin, but it is not affected
by folate stores.
This indirect measurement is a sensitive marker for
B12 deficiency in the non-pregnant setting. In pregnancy, however, there is a poor correlation between
serum B12 and no correlation between urinary methymalonic acid and serum B12. Normal references in
pregnancy have not been established and standard
adult reference ranges quoted on laboratory reports are
not applicable in pregnancy.

Auto-antibodies
Intrinsic factor antibodies can be helpful in the diagnosis of pernicious anemia if the results are positive.
Antibodies to intrinsic factor are found in 70% of
patients with pernicious anemia. These antibodies can
cross the placenta and cause intrinsic factor deficiency
in the fetus. Antiparietal antibodies are non-specific
and not very sensitive in diagnosing pernicious anemia. They are no longer recommended.

Schilling test
This test has been used classically to diagnose pernicious anemia. It is contradicted in pregnancy because
of the radiation risks.

Trial of B12

A therapeutic trial of B12 can confirm the diagnosis. A reticulocytosis occurs within 3–4 days and
peaks at day 6–7. The hemoglobin concentration rises
within 10 days and usually returns to normal within
8 weeks. Hyper-segmented neutrophils disappear at
around 10–14 days.
In patients with severe anemia, hypokalemia can
occur as potassium is used in the production of new
red cells. This requires monitoring and potassium supplementation if necessary.
Neurological abnormalities are slower to improve
and can take months.

Bone marrow
Megaloblastic erythropoiesis is demonstrated by the
finding of large erythroblasts and giant abnormally
shaped metamyelocytes. Although this is a rapid and
reliable method of assessment, the invasive nature of
the test means it rarely done to diagnose B12 deficiency
as there are several reliable non-invasive tests. Bone
marrow examination is generally reserved for patients
with pancytopenia.

Management
Most mechanisms of B12 deficiency are absorptive and
treatment is generally parenteral. Hydroxycobalamin
or cyanocobalamin 1 mg is given three times a week
for 2 weeks and then every 3 months. Neurological
involvement may require higher doses.
Oral B12 can be given if dietary deficiency is the
etiology. There is literature supporting the use of high
dose cobalamin 1–2 mg/day in patients with impaired
intrinsic factor function. There is a second less efficient cobalamin transport system that does not require
intrinsic factor. This type of treatment requires very
good patient compliance and monitoring of cobalamin
levels.

Dilemmas
B12 assays are often coupled with
folate assays
Often unrequested B12 results are generated because
of the coupling of these tests. This can lead to difficulties as the B12 level is almost always low in the pregnant population but the quoted reference range on the
laboratory report is that of a non pregnant population. This can lead to many phone calls, referrals, and
concerns. Ideally, B12 assays should not be carried out
unless a specific request, based on clinical grounds, has
been made.

Summary
B12 deficiency is rare in pregnancy and vitamin B12
levels should be interpreted with caution. B12 levels
fall in pregnancy by up to 50% in the third trimester.
The reference ranges quoted on reports are for nonpregnant populations.

Chapter 2. Hematinic deficiencies

References
1.

Pena Rossa JP, Viteri FE. Effects of routine oral iron
supplementation with or without folic acid for women
during pregnancy. Cochrane Database of Systematic
Reviews 2006; 3: CD004736.

Fishbane S, Ungureanu VD, Maesaka JK et al. The
safety of intravenous iron dextran in hemodialysis
patients. American Journal of Kidney Disease 1996; 28:
529–534.

Aronoff GR, Bennett WM, Blumenthal S et al. Iron
sucrose in hemodialysis patients: safety of replacement
and maintenance regimens. Kidney International 2004;
66: 1193–1198.

Dodd J, Dare MR, Middleton P. Treatment for women
with postpartum iron deficiency anaemia. Cochrane
Database of Systematic Reviews 2004; 4: CD004222.

Sifakis S, Angelakis E, Vardaki E et al. Erythropoietin
in the treatment of iron deficiency anemia during

pregnancy. Gynecological Obstetric Investment 2001;
51: 150–156.
6.

Reveiz L, Gyte GML, Cuervo LG. Treatments for
iron-deficiency anaemia in pregnancy. Cochrane
Database of Systematic Reviews 2001; 2:
CD003094.

Magda M Megahed, IM Taher. Folate and
homocysteine levels in pregnancy. British Journal of
Biomedical Science 2004; 62: 84–86.

Honein MA, Paulozzi LJ, Mathews TJ et al. Impact of
folic acid food fortification of the food supply on the
occurrence of neural tube defects. Journal of the
American Medical Association, 2001; 285: 2981–
2986.

Murphy MM, Molloy A, Ueland PM et al. Longitudinal
study of the effect of pregnancy on maternal and fetal
cobalamin status in healthy women and their offspring.
The Journal of Nutrition 2007; 137: 1863–1866.

Section 1
Chapter

Cellular changes

Inherited red cell disorders
Emma Welch and Josh Wright

Introduction
The hemoglobinopathies are common genetic disorders. They may result in significant morbidity and
mortality, affecting all age groups and genders. This
chapter will concentrate on sickle cell disease and thalassemia. The abnormalities of hemoglobin can be of
two kinds.
Structural: such as in sickle cell disease, where a single nucleotide change in the ␤-globin gene leads to the
substitution of valine for glutamine at position 6 on the
␤-globin chain.
Or
Disorders resulting from unbalanced globin chain production: the thalassemias, the globin chains produced
are structurally normal, but reduced in quantity.

Ante- and neonatal screening for
hemoglobin disorders
Rationale
Ante-natal screening aims to allow informed reproductive choice by identifying couples, at risk of an
affected infant, at an early stage in pregnancy. Options
include pre-natal diagnosis with either termination or
continuation of affected pregnancies.
It has long been known that morbidity and mortality in children with sickle cell disease is high in the
first 5 years of life. The protective effects of high levels of HbF in the newborn decline over the first 4–6
months of life, thereafter much of the mortality is due
to pneumococcal septicemia and acute splenic sequestration. Successful antibiotic prophylaxis, vaccination
and education programs have all but eliminated these
problems2 and are perhaps the single most important
step in the improved survival of sickle cell disease.

Since these severe complications are often the presenting features of sickle cell disease, a screening program is required to identify at-risk couples and/or
affected newborns.
In ␤ thalassemia major the failure of ␤ globin chain
production results in a severe transfusion-dependent
anemia, which is manifest as HbF levels reduce in
the first few months of life. From this point on, the
management of thalassemia is based upon regular
transfusion and iron chelation to reduce the risk of
organ damage, particularly cardiac. Care of the patient
with thalassemia involves collaboration of hematologists, endocrinologists, diabetologists, cardiologists,
with occasional input from other specialities such as
hepatology. With appropriate care and good compliance, life expectancy may be normal; however, early
cardiac death is common in those who do not comply
with iron chelation.
Neonatal screening for inherited disease is only
undertaken if:
1. it is common in a particular population;
2. there is a cost-effective reliable screening strategy;
3. detection of disease leads to improvements in
care/survival.

NHS sickle and thalassemia
screening – an example of a linked
ante-natal and neonatal program
The newborn program screens all births in England,
with samples collected by heel prick onto a Guthrie
card. The regional screening laboratories generally use
high performance liquid chromatography (HPLC) to
detect the presence of significant variant hemoglobins;
second-line confirmation is performed by iso-electric
focusing. The program has close links with Child

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

Chapter 3. Inherited red cell disorders

Fig. 3.1 Distribution of ␣ thalassemia
(taken from Barbara Bain,
Hemoglobinopathy Diagnosis).
Reproduced with permission.

MED
α 3.7 I 5−15%
αTα

5%−15%

60%
(−α 3.7)
αTα
5%−40%
(−α 3.7 1)
αTa

SEA
α 3.7 I
α 4.2
T
α α

5%−80%

α 4.2
α 3.7 III
αTα

15−80%
(−α 3.7 I)
(−α 3.7 II)
(−α 4.2)

α + thalassemia
α0 thalassemia

Table 3.1 Outcomes for neonatal screening

Hemoglobin results

Diagnostic possibilities

Sickle disorders
FS

Sickle cell disease 81%
Sickle cell ␤0 thalassemia 17%
Sickle HPFH 2%

FSC

Hemoglobin SC disease

FSA
HbS⬎ HbA

Sickle ␤+ thalassemia
?transfusion

Other significant disorders
F only

Possible ␤ thalassemia major
Prematurity
Homozygous HPFH

Homozygous hemoglobin E
Hemoglobin E ␤ thalassemia
Hemoglobin E HPFH

FA plus Hb Bart’s
Barts⬎20% A

Hemoglobin H disease
␣ thalassemia carrier

HPFH, Hereditary persistence of fetal hemoglobin.

Health, to allow appropriate referral of those requiring further follow-up, and the antenatal laboratories
to highlight mothers at risk of an affected child. The
main aim of the program is the detection of children
with sickling disorders, those at risk of thalassemia
major will be highlighted for further investigation (see
Table 3.1).

With the aim of the antenatal program being
choice, there is considerable time pressure to obtain
results of the patient and partner and to counsel and
arrange antenatal diagnostic procedure if required.
Since termination is one option, early diagnosis is crucial and a target for identification of at risk couples is
set at 10 weeks.
All couples at risk of having an affected child
should be offered pre-natal diagnosis, although many
will decline. Prenatal diagnosis is usually by chorionic
villous sampling between 10 and 12 weeks’ gestation.
The fetal loss rate is approximately 1%. Alternatively,
amniocentesis may be performed at 15 weeks or more
with a miscarriage rate 0.5%–1%.
If prenatal testing results confirm a fetus affected
with a major hemoglobin disorder, then a couple
need counseling about living with a child affected
by hemoglobinopathy. The earlier a diagnosis of a
hemoglobinopathy is made, the higher the likelihood
that termination is acceptable. In a study examining
at prenatal testing in thalassemia amongst British
Pakistanis, 70% accepted prenatal diagnosis if offered
in the first trimester, with over 90% of pregnancies
being terminated. However, if testing was offered in
the second trimester, only 40% of couples accepted
prenatal testing with fewer affected pregnancies
terminated.1

Section 1. Cellular changes

Antenatal screening for hemoglobin disorders is
universal in areas of high prevalence and, where prevalence is low, the selection for screening is on the basis
of family origin using an ethnicity questionnaire and
red cell indices (see screening algorithms: Figs. 3.2 and
3.3).

Sickling disorders in pregnancy
The sickling disorders are a group of inherited chronic
hemolytic anemias with clinical manifestations occurring as a result of the polymerization of hemoglobin S.
The disorders in which sickling occurs are:
r Homozygous sickle cell disease – HbSS. The most
common and generally the most severe.

Compound heterozygous states
r
r
r
r
r

Hemoglobin SC disease
Hemoglobin S␤ Thalassemia
Hemoglobin SD Punjab
Hemoglobin SO Arab
Hemoglobin SLepore Boston.

Carriage of hemoglobin S is not associated with
significant disease and its only significance in pregnancy is in terms of genetic counseling and the need
for partner testing.

Pathogenesis
The clinical manifestations in sickle cell disease are
as a result of many interacting pathological processes
including:
r polymerization of HbS;
r hemolysis and nitric oxide depletion;
r vaso-occlusion.

Polymerization of HbS
HbS forms insoluble polymers at low oxygen tensions.
The polymers interact with red cell membrane proteins
causing progressive damage ultimately leading to the
formation of the typical sickled blood cell.

Hemolysis

Sickle cell disease is characterized by chronic intravascular and extravascular hemolysis, red cell lifespan is
shortened from 120 days to 16–20 days. This chronic
hemolysis leads to the liberation of free hemoglobin
which mops up nitric oxide released from the vas-

cular endothelium. This, in turn, leads to endothelial
activation and vasoconstriction, providing ideal conditions for adherence of cellular blood components.

Vaso-occlusion
The combination of poorly deformable red blood
cells, increased viscosity, endothelial activation, and
vasoconstriction causes ongoing vaso-occlusion in the
microvasculature. The process is further exacerbated
by leukocytosis, platelet activation, and increased levels of pro-inflammatory cytokines. Vaso-occlusion
leads to both the acute complications of sickle cell disease such as painful crises as well as chronic organ
damage, including cardiac and renal impairment seen
in older patients.2,3

Contraception
There are few data to guide contraceptive choice for
women with sickle cell disorders. What is certain is
that the risks of pregnancy in sickle cell disease far
outweigh the risks of contraception. The condition is
listed as a relative contraindication for some combined
oral contraceptive preparations based upon the theoretically greater risk of thromboembolism in sickle cell
disease. There is little evidence to support this, particularly with the lower dose pills, which are commonly
prescribed. Progesterone-only contraceptives are also
safe, indeed limited data suggests they are associated
with a favorable change in hematological parameters
such as reduction in hemolytic rate and increased HbF.
Levonorgestrol implants and intrauterine systems are
safe and have a low failure rate. Copper-containing
intrauterine devices have been felt to be contraindicated because of infection, and possibly heavier menstrual loss.
In general, sickle patients should be offered the full
range of contraceptives available and counseled about
the risks and benefits of each method.

Maternal and fetal complications
of pregnancy
Much of the published information on pregnancy in
sickle cell disorders relates to homozygous (SS) sickle
cell disease. This, and S␤0 thalassemia, are, in general,
the most severe forms. Patients with milder sickle conditions such as SC disease and S␤+ thalassemia can
also have complicated pregnancies though the risks are
lower. All patients with sickling disorders should be

Refer to
Consultant
hematologist

Other
variant

No further
action

Refer to
Consultant
hematologist

HbA2 4.0%
or
HbF > 5%

Test baby’s
father

HbA2 > 3.5%
beta thal trait

Fig. 3.2 Testing algorithm for laboratory screening in low prevalence areas.

Test baby’s
father

HbS, HbC,
HbDPunjab, HbE ,
HbOArab,
HbLepore

No Hb variant
HbA2 < 4%
HbF < 5%

HPLC

No further
action

Hb variant

High-risk
family origin

Low-risk
family origin

Consider
family origin

MCH > 27 pg

HbA2 < 3.5%

Test baby’s
father

No further
action

Low risk of
alphao
thalassemia

Consider family
origin

MCH < 25 pg

High risk of
alphao
thalassemia

FBC

No further
action

Iron deficiency
alpha thal

MCH > 25 pg

HPLC

MCV < 27 pg

Test baby’s
father

HbF > 5%

Test baby’s
father

HbS, HbC,
HbDPunjab, HbE,
HbOArab,
HbLepore

Hb variant

Refer to
Consultant
hematologist

Other variant

HbA2 > 3.5%
beta thal trait

Test baby’s
father

Other variant

Refer to
Consultant
hematologist

Test baby’s
father

No further
action

Low risk of
alphao
thalassemia

No further
action

Iron deficiency
alpha thal

Consider family
origin

High risk of
alphao
thalassemia

MCH > 25 pg

HbA2 < 3/5%

MCH < 27 pg

No variant

MCH < 25 pg

Fig. 3.3 Testing algorithm for laboratory screening in high prevalence areas.

Test baby’s
father

HbS, HbC,
HbDPunjab, HbE,
HBOArab,
HbLepore

Hb variant

FBC
HPLC

Test baby’s
father

HbF > 5%

Refer to
Consultant
hematologist

HbA2 > 4.0%
or
HbF > 5%

MCH > 27 pg

No further
action

HbA2 < 4.0%
HbF < 5%

Chapter 3. Inherited red cell disorders

jointly managed by an obstetrician and a hematologist
with interest and experience in these diseases. Since
these pregnancies are high risk, patients will require
frequent review by the multidisciplinary team.
Twin and multiple birth pregnancies are associated
with a higher rate of serious complications.

Problem-free pregnancies
Despite the potential complications, more than onequarter of these pregnancies occur without problems.
Table 3.2

Maternal risks
Increased mortality
Painful crisis
Infection
Chest syndrome
Hypertension & pre eclampsis
Worsening anaemia
Increased cesarian rate
Thrombosis

Table 3.3

Fetal/neonatal risks
Miscarriage
Increased perinatal mortality
Intrauterine growth retardation and low birth weight
Premature delivery
Increased cesarean rate

Maternal mortality
Maternal mortality rates are known to be increased
in sickling disorders. Prior to the 1970s, 30%–40% of
women with sickle cell disease did not survive pregnancy, prompting obstetricians to question whether
the maternal risks of pregnancy were justified. Recent
decades have seen a marked improvement, currently
mortality has been shown to be 1%–2% in studies from
USA and Europe.4,5 In Africa, maternal mortality rates
are between 7% and 12%, probably as a result of a lack
of ante-natal care. In Benin, one of the least developed countries in Africa, an active pre-natal program
reduced mortality to 1.8%, comparable to the West.6

Mortality and morbidity rates have been found to be
similar in both HbSS and HbSC pregnancies.
In the triennial “Confidential Enquiries into
Maternal Deaths in the UK” there were five deaths
between 1982 to 1999 associated with sickling conditions. These were due to pneumonia, multi-organ
failure following placental abruption in SS disease,
acute chest crisis in SS disease, septicemia in S␤
thalassemia and sickle crisis with multi-organ failure
in SC disease. From 1999 to 2005 there were four
deaths in women with sickling disorders, but not all
directly associated with their hemoglobinopathy. One
woman with SC disease died of thromboembolism,
another with SS disease and myocardial fibrosis
died whilst having a fit and a painful crisis, another
woman also died during an epileptic fit, and finally
a woman with SC disease died of an amniotic fluid
embolism.
The recent NCEPOD report (“A Sickle Crisis?”
July 2008) highlights difficulties with death certification and autopsy in sickle cell disorders. Few pathologists have significant experience and non-specialist
sickle clinicians are in a similar position. It is recommended that pathologists with appropriate experience
perform such autopsies, though there are now national
guidelines for autopsy in sickle cell disease. Clinicopathological correlation is crucial, for example, in differentiating sickle chest from pneumonia or whether
thrombosis is likely to have been in situ or embolic.
Notwithstanding this proviso, the reports into maternal death illustrate the importance of multidisciplinary
management and, in several cases, suggest a lack of
awareness of the nature and difficulty of sickle cell
pregnancy.
These women may have complex co-existing medical problems which can make the management of their
pregnancy even more challenging.

Perinatal mortality
The last 30 years or so have seen marked improvements
in fetal outcomes as a consequence of joint obstetric/
hematology care. Peri-natal mortality was reported to
be as high as 50%–80% prior to the 1970s. More recent
studies in USA and Europe have reported a peri-natal
mortality rate of between 1–8%,5 even in Benin rates
are between 12% and 19%.6 Howard et al. reported a
peri-natal death rate of 60 per 1000 in the period 1991–
1993 in UK centers, five times higher than the general
obstetric population at this time.5

Section 1. Cellular changes

Miscarriage

Hypertension

There is known to be an increased risk of miscarriage in the sickling disorders. This has previously been
reported at between 19% and 24%. A recent study in
Jamaica found a miscarriage rate of 36% in sickle pregnancies and 10% in controls. This is higher than previously documented. 7

Pregnancy-induced hypertension and pre-eclampsia
complicate one-third of pregnancies in sickle cell disease. There is an association between hypertension
with proteinuria and simultaneous sickling complications.4

Thrombotic risk
Premature deliveries
Since the 1970s it has been known that women with
sickling disorders are more likely to have premature
deliveries. This has been reported at an average of
between 34.1 to 38.5 weeks’ gestation. In a recent
Jamaican study the mean gestational age was found to
be 37.0 weeks compared with 38.7 weeks in controls.
In African Americans the mean gestational age was 37
weeks. Infants born to SS mothers are twice as likely to
be preterm compared to Hb SC mothers.7

Fetal intrauterine growth retardation
Intrauterine growth retardation is a well-documented
complication of sickle cell pregnancy. This is thought
to arise as a consequence of maternal anemia
and impaired placental function resulting from
vaso-occlusion in uteroplacental circulation. Histological studies have shown placental infarction with
abruptions and villous edema.
Of infants born to mothers with sickle cell anemia,
77% have a birth weight below the 50th centile, with
21% below the 10th centile. Neonates born to mothers
with Hb SS disease are significantly smaller than babies
born to mothers with Hb SC disease.4–7

Infections

Patients with sickle cell disease have a complex
immune defect. In addition to hyposplenism, there are
data suggesting subtle changes in leukocyte function,
opsonization and complement pathways. Urinary tract
infections are increased in normal pregnancies and can
lead to pyelonephritis and premature labor. There may
be a further increase in risk in sickle pregnancy.
Other common sites of infection include chest
and bone. Common pathogens include Pneumococcus, Salmonella, E. Coli and Mycoplasma. Infection is
a common precipitant of painful crises.

Pregnancy labor and the puerperium are associated
with complex changes of the hemostatic enzyme systems. Thrombotic risk is increased in normal pregnancy. To further complicate this situation, it has long
been recognized that steady state sickle cell disease
is associated with evidence of platelet and coagulation activation. Furthermore, changes in the levels of
the naturally occurring anticoagulants and endothelial
activation also have the potential to increase the risk of
thrombosis in sickle cell pregnancy.
Despite these biochemical changes, the role of
thrombosis in sickle cell disease has been difficult to
establish. The pregnant patient with sickle cell disease
should be regarded as at high risk of venous thromboembolism. Pulmonary embolism is difficult to diagnose in this setting, but should be considered within
the differential of a patient presenting with dyspnea
and chest pain.

General management of sickle cell
pregnancy
Preconception

r Discuss maternal and fetal risks of pregnancy and
counsel about availability of pre-natal diagnosis.
r Partner screening.
r Folic acid supplements.
r Review medications, with assessment of risks vs.
benefits for individual drugs. Stop
hydroxycarbamide 3 months before conception
and discuss potential need for transfusion.

At booking

r Discussion of pregnancy and associated risks.
r Early involvement of a hematologist with
expertise in the hemoglobinopathies.
r Review by an obstetrician experienced in the care
of women with hemoglobinopathies.

Chapter 3. Inherited red cell disorders

r Early booking appointment and establishment of
a planned schedule of care between obstetrician
and hematologist.
r FBC, Hb electrophoresis/ HPLC, U&E plus full
red cell phenotype. Check ferritin and folate
status.
r Ensure partner screening.
r Discussion of pre-natal diagnosis, if appropriate
r Folic acid 5 mg daily, continued throughout
pregnancy.
r Take full history particularly frequency and
management of crises, transfusions, previous
pregnancies, evidence of chronic organ damage,
which may contribute to risk.
r Review medication – penicillin, folic acid,
hydroxycarbamide, iron chelators, analgesic usage.
r Stress the importance of early presentation if
unwell.
r Education about the signs and symptoms of
infection.
r Ante-natal screening for Hepatitis B, C and HIV,
given likely transfusion history.
r Echocardiogram to assess left ventricular function
and pulmonary pressures if evidence of iron
overload or cardiorespiratory symptoms/signs.
r Ultrasound to assess viability and confirm
gestation.

Throughout pregnancy
r
r
r
r
r
r
r
r
r
r
r
r

Continued health education.
Continue folic acid 5 mg.
Iron supplementation if ferritin low.
Regular FBC checks every 4 weeks and U&E every
8 weeks.
Serial ultrasound scans from 20 weeks to assess
fetal growth/placental function.
Monthly mid-stream urine culture.
Low threshold for admission especially if limb,
bone, abdominal, chest pain after 28 weeks.
24-hour admission policy and contact numbers.
Appropriate plan for use of analgesia in
pregnancy. Avoid non-steroidal
anti-inflammatory drugs after 34 weeks.
Involve obstetric anesthetist to discuss
management in labor.
Prompt treatment of emesis to avoid dehydration.
Transfuse only after discussion with a
hematologist.

r Watch closely for features of acute chest
syndrome. Seek advice from obstetrician,
hematologist and anesthetist. Chest crises are
most likely to occur during late third trimester
and postpartum.
r If admitted during pregnancy, use low molecular
weight heparin for thrombopropylaxis and
compression stockings.

Painful crisis in pregnancy

r The majority of severe crises occur in the third
trimester often, at the time of delivery, often the
complications of sickle cell disease precipitate
labor rather than labor precipitating sickling
complications.
r 30%–80% of women with Hb SS pregnancies have
crises.
r 30% of women with HbSC have crises in
pregnancy. SC disease is generally a milder
condition when not pregnant but patients may
present with pain and other sickle complications
in the third trimester.
r Labor and early puerperium are risk periods for
development of pain. This becomes more likely in
the presence of infection, dehydration or acidosis.
r Sickle patients have a renal concentrating defect
from early childhood and pass large volumes of
dilute urine. Attention to hydration status is
therefore crucial.
r Crises in pregnancy may present as abdominal
pain which can be difficult to distinguish from
obstetric complications.
r The risk of thromboembolism increases in
pregnancy.

Management

r Admit to obstetric or hematology ward as per
local protocol. In the final trimester with the high
risk of obstetric problems, the obstetric setting is
most appropriate.
r Inform relevant staff (hematologist/obstetrician).
r Ensure rest and warmth.
r Give oxygen if hypoxic on monitoring of O2
saturation.
r Ensure adequate hydration – oral or intravenous
fluids 3–4 liters. Strict fluid balance essential.
r Pain relief – take account of previous analgesic
history. Use paracetamol, non-steroidals if

Section 1. Cellular changes

r
r

r
r
r
r

r
r
r

pregnancy less than 34 weeks but subcutaneous
opiates are often necessary.
Pethidine is not recommended for the treatment
of sickle pain. Morpine, diamorphine or
oxycodone are appropriate but intravenous use
should be discouraged.
Use linear analog scale to assess pain control.
Patient-controlled analgesia or subcutaneous
pumps are occasionally required.
Regular assessment of sedation and conscious
level if on strong opiates. The recent NCEPOD
report highlights deficiencies in the care and
monitoring of patients on opiate analgesics.
Investigations – FBC, reticulocytes, U&Es, group
and screen, pulse oximetry and arterial blood
gases if appropriate.
Microbiology – urine culture, blood cultures and
throat swabs.
Consider chest X-ray if chest involvement.
Antibiotics are not routinely required unless
evidence of infection, low grade fever ⬍38 ◦ C is
common in painful crisis even in the absence of
infection.
Low molecular weight heparin
thromboprophylaxis and compression stockings.
Discuss indication for transfusion or exchange
transfusion with hematologist.
Chest physiotherapy including incentive
spirometry will reduce the risk of a subsequent
chest syndrome in patients with rib pain.

Acute chest syndrome (ACS)
This condition remains one of the most common
causes of death in sickle cell disease. It is characterized
by pulmonary infiltrates on the chest X-ray, chest pain,
shortness of breath and fever. Not surprisingly, those
unfamiliar with sickle cell disease frequently diagnose
a chest infection and manage with antibiotics alone.
Despite the radiological appearances (which may lag
behind clinical signs), this is predominantly a vascular
event and responds well to blood transfusion.

Management of chest crises
in pregnancy
36

r Inform consultant obstetric and hematology staff
on admission.
r Continuous monitoring of O2 saturation and
supplemental oxygen.

r Investigations – CXR, blood gases on air, pulse
oximetry, FBC, reticulocytes.
r Broad spectrum antibiotics – should include a
macrolide.
r Bronchodilators.
r iv fluids.
r Transfusion, either exchange or top up, should be
considered in hypoxemia (SaO2 ⬍5% lower than
patient’s steady-state level), deteriorating clinical
status or progressive multi-lobe involvement.
r The timing of transfusion rather than the volume
is critical (i.e. early in disease course).
The key to appropriate transfusion in ACS is the
timing rather than the volume of blood used or the
target %HbS. In most cases early top-up or partial
exchange transfusion is the optimal approach. In the
United States, The National ACS study group showed
simple top-up transfusion was performed in 68% of
patients using an average of 3.2 units of packed cells.
This appeared to be as effective as an exchange transfusion. In the absence of a randomized controlled trial
a sensible approach is to use simple top-up transfusion, aiming for a hemoglobin of no more than 9–10
g/dL, in patients with relatively mild episodes or those
with severe anemia, e.g. ⬍5 g/dL and to use exchange
transfusion in the more severe cases. Again, the timing of exchange transfusion is crucial. It is preferable
to perform a limited manual partial exchange urgently
rather than waiting for several hours or overnight
until staff are available to perform an automated
exchange.

Labor and delivery

r Aim to achieve a vaginal delivery, no need to
schedule delivery.
r Keep warm.
r Maintain good hydration – commence iv fluids at
time of admission in labor at rate 1 L/8 hours to
maintain good urine output. Strict fluid balance.
r Check full blood count, blood group, and
antibody screen.
r Continuous pulse oximetry. May need
supplemental oxygen.
r Continuous CTG monitoring throughout labor
r Epidural analgesia is pain relief of choice.
r Avoid prolonged labor, not more than 12 hours
and prolonged rupture of membranes, which
increase the risk of infection and dehydration.

Chapter 3. Inherited red cell disorders

r If operative delivery necessary, discuss with
hematologist. Regional (rather than general)
anesthesia reduces the likelihood of sickle crisis
and post-op acute chest syndrome.
r Thromboprophylaxis with low molecular weight
heparin and graduated compression stockings.
r Alert pediatricians.

Postpartum
Baby

r Monitor for signs of respiratory depression if
opiates have been used intrapartum.

Mother

r Maintain hydration and oxygenation. Watch for
signs of painful or chest crises.
r 4-hourly observations for 24 hours post-delivery.
r Low threshold for the use of antibiotics
particularly after operative delivery.
r Check FBC day 1 post-delivery.
r Mobilize early and continue thromboprophylaxis
until discharge.
r No contraindication to breastfeeding.
r Give appropriate contraceptive advice prior to
discharge.
r Ensure patient has follow-up both for post-natal
check and with hemoglobinopathy team.

Dilemmas
Operative deliveries
To section or not? The management of labor in patients
with sickling disorders varies widely from unit to unit.
There are risks and benefits of planned vs. spontaneous labor. Many units offer a planned induction at
38 weeks. There is, however, no evidence to support
this approach and in general spontaneous labor is preferred. Induction leads to a higher cesarean section
rate, with its own complications plus the implication
that future pregnancies will need a trial of scar and be
associated with a risk of subsequent operative delivery. Elective Cesarean is not usually advised in sickling disorders. They are associated with a 30% increase
in maternal morbidity, significantly higher than when
emergency section is performed in spontaneous labor
for obstetric reasons.

If operative delivery is felt necessary then
the patient’s condition should be optimized preanesthetic. Particular attention needs to be paid to
hydration and oxygenation. The procedure may be
undertaken without transfusion support. However,
if felt necessary, then simple top-up transfusion is
adequate.
Post-operative chest physiotherapy including
incentive spirometry may reduce the risk of chest
syndrome.

Hydroxycarbamide
Many patients with sickle cell disease are routinely
managed with hydroxycarbamide. This agent induces
hemoglobin F and also acts as a nitrous oxide donor.
Hydroxycarbamide has been found to reduce the
occurrence of painful and chest crises and may also
prolong life. It has been found to be teratogenic in animal studies. Males and females on hydroxycarbamide
should therefore be counseled about the importance of
using contraception whilst on the drug. They should
be asked to stop hydroxycarbamide at least 3 months
before trying to conceive. However, case series have
been published showing that hydroxycarbamide can
be taken throughout pregnancy without complication.
If conception occurs accidentally whilst on hydroxycarbamide, the drug should be stopped.

Prophylactic transfusion
The role of transfusion in sickle cell disease in pregnancy is controversial though it is generally accepted
that transfusion is not required as part of the management of uncomplicated sickle pregnancy.
The rationale is to reduce the amount of circulating hemoglobin S thereby improving oxygenation and
placental function. A single randomized control trial
in 1980s concluded that routine prophylactic transfusion from the onset of pregnancy does not alter the
outcome for the fetus; however, the numbers involved
in this study are small and it should therefore be interpreted with caution.8 A retrospective study of the use
of red cell transfusion in the UK noted a trend towards
fewer sickling complications in third trimester and
puerperium.4 There was no evidence that transfusion
improved fetal growth or outcome. A further study
compared a restricted transfusion policy (not transfusing blood unless the hemoglobin fell below 6 g/dl)
vs. a prophylactic transfusion policy (transfusing if

Section 1. Cellular changes

hemoglobin fell below 10 g/dl). They found similar
rates of crises and other complications in both groups.
The risk of alloimmunization was found to be 10%–
20%, this rate can be reduced but not completely abolished by the use of phenotypically matched blood.4
These antibodies have potential to produce hemolytic
disease of the newborn and may cause difficulty in
provision of compatible units for future transfusions.
All women should have a group and full phenotype at
booking visit to screen for antibodies present.
In conclusion, transfusion should be reserved for
high-risk pregnancies. This would include twin pregnancies, women with previous poor obstetric history,
chest crises, recurrent pain, and severe anemia.

Summary
The key to successful outcome of sickle pregnancy lies
in the close interaction between obstetric teams and
hematologists. Close monitoring, awareness of risks
and complications is essential. The majority of pregnancies have a successful outcome. Where possible,
pregnancy should be allowed to proceed with minimal intervention there being little evidence that transfusion or operative delivery are of any benefit in the
majority of cases.

Thalassemia and pregnancy
In the past, thalassemia major was associated with a
high mortality rate in the first decade of life. Over
recent years outcomes have improved, with children
surviving into adult life in good health, leading normal
lives, and able to have families of their own.
The mainstay of management is regular transfusions with concurrent iron chelation to reduce iron
overload. The most common cause of death is cardiac
failure due to siderosis, although iron overload can
also occur in the endocrine glands, pancreas, and liver.
Many patients develop growth failure, central hypogonadism, and diabetes.

Pathogenesis

The thalassemias are almost always autosomal recessive disorders caused by mutations or deletions in the
␣ or ␤ globin genes leading to diminished or absent
production of one or more globin chains. The other
globin chain is produced in relative excess and precipitates within erythroid precursors causing chronic
hemolysis and ineffective erythropoiesis.

␣ thalassemia

Four ␣ globin genes are inherited as a pair from
each parent. A normal individual is annotated thus
(␣␣/␣␣). The more ␣ genes deleted, the more severe
the condition (Table 3.4).
Alpha thalassemias are the commonest single gene
disorders worldwide. Approximate frequencies and
types of carriage are illustrated in Fig. 3.1.

␣ thalassemia carrier (␣␣/–), (␣-/␣-) or
(-␣/␣␣)
Carriers of ␣ thalassemia are asymptomatic and are
usually first detected at ante natal screening. Their
hemoglobin is in the normal range or minimally
decreased with low mean cell volume (MCV) and
mean cell hemoglobin (MCH).

Hemoglobin H disease (–/-␣)
Those affected by hemoglobin H disease have three
non-functioning alpha genes. The hemoglobin is
commonly in the range 8–9 g/dl with microcytic,
hypochromic red cell indices, and splenomegaly. HbH
disease is a mild form of thalassemia intermedia,
those affected rarely need transfusion. The anemia may
worsen in pregnancy and with infection. The condition
is diagnosed by the presence of an HbH peak on the
HPLC trace and typical “Golf ball” cells on supravital
staining.

Hemoglobin Bart’s hydrops (–/–)
A complete absence of ␣ chains is incompatible with
life and results in the unopposed ␥ chains forming
tetramers called hemoglobin Bart’s. This is a common
cause of stillbirth in areas with a high frequency of
(–/␣␣) such as SE Asia and the Eastern Mediterranean.
The fetus is stillborn at 34–40 weeks or dies soon
after birth. The Hb Bart’s binds oxygen poorly impairing tissue oxygenation. The fetus appears edematous
and jaundiced with massive hepatosplenomegaly and
ascites.
Couples at risk of a child with Bart’s Hydrops
should be picked up by ante-natal screening programs
and offered ante-natal diagnosis. If found to have an
affected infant, termination should be offered.

Chapter 3. Inherited red cell disorders

Table 3.4 Effects of alpha gene deletion

Table 3.5 Effects of iron overload

Genotpye

Outcome

Effect

␣␣/␣␣

Normal

-␣/␣␣

Heterozygous ␣+
thalassemia trait

Frequently silent or
slight decrease in
MCV/MCH

-␣/-␣

Homozygous ␣+

MCH⬍25 pg

–/␣␣

Heterozygous ␣0
thalassemia trait

MCH⬍25 pg

–/-␣

Hemoglobin H disease

Hb 8–9 g/dL

–/–

Hemoglobin Bart’s
Hydrops

Death in utero

MCV, mean cell volume, MCH, mean cell hemoglobin.

␤ Thalassemia carrier
Asymptomatic and diagnosed at ante-natal screening
or during investigation of microcytic, hypochromic
indices. The hemoglobin is rarely less than 10 g/dl.
Hemoglobin A2 is raised. Iron replacement need not
be given unless a deficiency state is proven by reduced
serum ferritin.

␤ Thalassemia intermedia
A range of interacting genetic lesions may lead to a
thalassemic phenotype of varying severity. Some will
be asymptomatic whilst others require intermittent
transfusion. The hemoglobin is usually 10–12 g/dl,
but can be as low as 5–6 g/dl in severe forms. Hepatosplenomegaly may be present.

␤ Thalassemia major
This is the inheritance of severe abnormalities in
both ␤ globin genes. Onset of symptoms of anemia
occurs as fetal hemoglobin levels decline in the first
few months of life. Patients are transfusion dependent. If not treated with transfusion, extramedullary
hematopoeisis occurs leading to characteristic skeletal
deformities and hepatosplenomegaly. Morbidity and
mortality in this condition is now caused by transfusional iron overload (Table 3.5).

Management
Carriers of ␣ and ␤ thalassemia and those with
hemoglobin H disease or other mild forms of thalassemia intermedia can be managed as a normal pregnancy. Anemia may worsen during pregnancy because
of the normal physiological changes. Oral iron supplements should be given where there is a reduced

Common problems due to iron overload with
relevance to pregnancy
r
r
r
r
r

Central hypogonadism- may require referral to assisted
conception unit
Diabetes or impaired glucose tolerance
Cardiac siderosis
Small stature
endocrine dysfunction, for example, hypothyroidism

ferritin, but not for microcytosis and hypochromia
alone.
It is important to identify couples at risk of a baby
affected by hemoglobin Bart’s. This should be picked
up by the ante-natal screening program and parents
offered counseling, education and pre-natal diagnosis. The mother may also develop “mirror syndrome” a
severe pre-eclampsia, and delivery of a hydropic fetus
and placenta can cause obstetric difficulties.
␤ thalassemia major and severe forms of intermedia are clinically significant in pregnancy and require
careful multidisciplinary management.

Fertility
Because of the effects of iron overload, transfused
patients often have hypogonadotrophic hypogonadism, many patients are on hormone replacement
therapies but this does not restore fertility. The Standards for the Clinical Care of Children and Adults
with Thalassemia in the UK 9,10 state that:
r Iron chelation should be optimized from
childhood to reduce the risk of infertility.
r Where there is clinical or biochemical evidence of
pubertal or hormone disturbance, management
by an endocrinologist is required.
r Early referral for discussion of fertility issues
should be offered. This should be to a clinic
experienced in treating patients with thalassemia.
r Couples may be infertile for a number of reasons
including those unrelated to thalassemia and a
range of investigations may be necessary.
r Induction of ovulation or spermatogenesis may be
required for patients with central hypogonadism.
This needs to be done in a center with experience
of such patients to minimize the risk of
hyperstimulation syndrome and multiple
births.
r It is imperative that a couple are given the
opportunity to discuss the risk of having a child

Section 1. Cellular changes

Risks to women with thalassemia in pregnancy

Risks to the baby

r
r
r
r
r

Pregnancy causes a 30%–50% increase in cardiac output,
thus patients with significant cardiac siderosis are at risk of
decompensation and death
Transfusion requirements increase in pregnancy
Risk of accelerating pre-existing diabetic retinopathy or
nephropathy
Worsening osteoporosis
High incidence of gestational diabetes
High incidence of operative delivery

Table 3.6 Risks to women of thalassemia in pregnancy

with thalassemia or other major hemoglobin
disorder, e.g. sickle cell conditions if partner is a
sickle carrier. The partner must be tested and if
they carry thalassemia or variant hemoglobin
counseled about options and offered pre-natal
diagnosis.

Preconception
Careful preassessment of a woman with thalassemia
considering pregnancy is required.
r Full cardiology assessment including
echocardiogram, T2∗ MRI quantification of
cardiac iron (where available) as well as
assessment by a cardiologist
r Endocrinological assessment including glucose
tolerance test. Optimize diabetic control if known
to be diabetic
r Iron chelation should be optimized before
pregnancy considered. For well-controlled
patients with evidence of normal pituitary
function, it may be reasonable to stop chelation
for natural conception.
r Folic acid should be started prior to conception
until the end of pregnancy.
r Review rubella status, HIV, Hepatitis C status
prior to pregnancy.
r Discuss smoking and alcohol consumption.
r Partner screening and risk assessment for
thalassemia
r Review medication – ACE inhibitors should be
changed (Tables 3.6, 3.7).

Management of pregnancy

r Early booking appointment.
r FBC, group and save and full antibody screen at
booking.
r U&Es and LFTS at booking.

r
r
r
r

Possibility of a major hemoglobin disorder (depending on
partner carrier status)
Diabetes is associated with a four fold increased risk of fetal
anomaly and threefold increased risk of peri-natal mortality
Increased risk of chromosomal non-dysjunction, related to
maternal iron overload
Increased risks of multiple pregnancies secondary to
fertility procedures
Sudden maternal death in late pregnancy

Table 3.7 Risks to the baby of thalassemia

r Regular FBCs throughout pregnancy. Transfusion
requirements are likely to increase.
r Close involvement by obstetrician (experienced in
hemoglobinopathy), consultant hematologist and
cardiologist.
r Review all medications.
r Start folic acid before pregnancy and continue
throughout.
r Continue penicillin prophylaxis (if
splenectomized) throughout pregnancy.
r Calcium and vitamin D supplements are advisable
if bone density already reduced prior to
pregnancy.
r Stop ACE inhibitors and bisphosphonates.
r Stop iron chelators prior to ovarian stimulation
and pregnancy. Rate of iron accumulation during
pregnancy is surprisingly low.
r Increased risk of thrombosis in splenectomized
patients.
r Thromboprophylaxis whilst an inpatient and
during labor and puerperium.
r Discuss mode of delivery in advance – consider
cardiac problems and possible bony abnormalities
of pelvis to assess suitability for vaginal delivery.
r Discuss contraception post-delivery.

Medical problems in pregnancy
Bone problems

r Transfusion-dependent thalassemics show very
high rates of osteoporosis and osteopenia which
may be exacerbated by pregnancy.
r During pregnancy bisphosphonates need to be
stopped but vitamin D and calcium supplements
may be continued.
r Patients should be advised against smoking and
alcohol and encouraged to take regular exercise.

Chapter 3. Inherited red cell disorders

r Patients with back pain should be told this may
worsen in pregnancy and appropriate analgesia
discussed.

r Desferrioxamine is safe to use whilst breast
feeding. Deferiprone and deferasirox should not
be used until breastfeeding ceases.

Liver complications

Transfusion

r Common problem in thalassemia due to viral
infections, iron overload, biliary problems
secondary to gallstones, and drug toxicity.
r In North America 14% of the thalassemic
population are hepatitis C RNA positive.
r Vertical transmission of hepatitis C does occur but
is rare – upper estimates are 6%, but this increases
to 14%–17% where there is co-infection with HIV.

Endocrine problems

r The incidence of Type 1 Diabetes Mellitus in
thalassemia major is 6%–8% – these patients need
to be managed as per standard recommendations
for diabetes in pregnancy.
r Glucose tolerance should be assessed throughout
pregnancy.
r Treated hypothyroidism is present in 9% but up to
75% have evidence of thyroid dysfunction.
r Any patient with endocrine dysfunction should be
regularly assessed by a consultant endocrinologist.

Dilemmas
Iron chelation during pregnancy

r Iron chelation should be maximized prior to
pregnancy. Where possible, a low cardiac iron
load should be shown by T2∗ MR.
r It is advised that chelation agents are withheld
during pregnancy.
r There are case reports of women receiving iron
chelators throughout pregnancy without
teratogenic effects. Recommencement of chelation
could be considered for patients felt to be at high
risk of cardiac death.
r Vitamin C should also be stopped due to a risk of
precipitating cardiac damage.
r Serum ferritin levels may remain stable in
pregnancy, with no more than a 10% increase after
delivery despite cessation of iron chelation. This
may be due to the hemodilution effect or fetal
consumption of iron.
r Women should be encouraged to resume iron
chelation after delivery.

r Transfusion requirements will increase in
pregnancy.
r Patients who are not normally transfusion
dependent, e.g. ␤ thalassemia intermedia or
hemoglobin H disease may require transfusion in
pregnancy or post-delivery.
r Maintain hemoglobin over 10 g/dl in thalassemia
major.
r It is reasonable to observe patients with
thalassemia intermedia, provided there is no
cardiac dysfunction and serial ultrasound shows
normal fetal growth, transfusion may be avoided.
r Alloimmunization to minor blood antigens,
which may lead to increased difficulties in
cross-matching blood and risk of hemolytic
disease of the newborn in the fetus.
r Risk of transmission of blood-borne viral
infections via transfusion.

Delivery

r Mode of delivery needs to take account of
pre-existing cardiac problems.
r There is a high rate of Cesarean section in
thalassemic patients. In the majority of patients
this is due to cephalo-pelvic disproportion
resulting from the small stature of thalassemic
patients and normal growth of the fetus.
r In the absence of contraindications, labor may
proceed normally.

Cardiac problems

r The most common cause of death in thalassemic
patients is cardiac failure secondary to iron
deposition in the myocardium.
r Patients with poor compliance with iron chelators
and a ferritin above 2500 ␮g/l are more likely to
develop cardiac problems, pregnancy should be
delayed in such patients until chelation status is
acceptable.
r Cardiac arrythmias, cardiac failure and sudden
death can occur in a previously well patient – and
those without grossly elevated ferritins.
r Cardiac T2∗ MRI is the investigation of choice to
quantify cardiac iron and assess myocardial

Section 1. Cellular changes

function, though is only available in a few UK
centers. Cardiac T2∗ levels less than 20
milliseconds correlate with left ventricular
dysfunction. Further aggressive chelation prior to
pregnancy should be undertaken in such cases.
r Cardiovascular changes in pregnancy, anemia,
increase in plasma volume and increased cardiac
output can aggravate or precipitate cardiac failure.
r Severely impaired left ventricular function during
periods of stress maybe evident long before the
onset of cardiac failure and is a contraindication
to pregnancy.

Summary
Multidisciplinary care is essential to the management
of this complex group of patients.
Prior to conception efforts need to be made to maximize chelation and assess organ and endocrine function so that the patient can be counseled accurately.
From assessment of risk (e.g. cardiac, endocrine)
through to induction of ovulation and management of
established pregnancy it is vital to maintain good communication between the various specialist teams.

Red cell membrane disorders

Hereditary spherocytosis refers to a group of disorders
characterized by spherical erythrocytes of increased
osmotic fragility. There are a variety of molecular
lesions which are typically inherited in an autosomal
dominant manner and result in defects in the protein structure and interaction between various red
cell membrane components, leading to loss of membrane surface area and reduced deformability. These
cells have a reduced lifespan, resulting in a hemolytic
anemia. Hereditary spherocytosis occurs in all ethnic
and racial groups and there is considerable heterogeneity reflecting the wide range of molecular lesions.
Diagnosis is made by the typical blood film appearances, most patients have anemia, with hemoglobin
between 9–12 g/dL associated with a reticulocytosis
and other biochemical evidence of hemolysis, such
as reduced haptoglobin, raised LDH and bilirubin.
Approximately 10% of patients may have a more severe
anemia (6–8 g/dL). The diagnosis can be confirmed by
an incubated osmotic fragility test or flow cytometry.
Many patients lead normal lives and indeed the diagnosis may be an incidental finding.
For the most part, there are few implications for
pregnancy and the outcome is good. Some experience

anemia greater would be expected from the expanded
plasma volume due to higher hemolytic rate. Folate
requirements are increased in any hemolytic anemia
and patients known to have HS should be encouraged to take pre-conception folic acid supplements and
continue these through their pregnancy. In the more
severe cases transfusion may be required on an intermittent basis.
A cord sample should be taken for hemoglobin
and bilirubin levels. Neonates who have inherited HS
themselves may require transfusion, but it is worthy of
note that the degree of anemia at this stage does not
correlate with the hemoglobin level in later life.
Elliptocytosis has no significant implications for
pregnancy, though folate supplementation throughout
is prudent. Hereditary pyropoikilocytosis is a related
condition and is associated with typical blood film
appearances and a more severe degree of anemia. In
addition to folate supplementation, such patients may
require transfusion. The need for intervention with
transfusion in all red cell membrane disorders should
be judged individually and based upon hemoglobin
level, symptoms and assessments of fetal wellbeing.

Glucose-6-phosphate dehydrogenase
deficiency
Deficiencies in red cell enzymes often lead to shortened red cell lifespan. G6PD deficiency was the first
of such abnormalities to be discovered and is the most
common. The presence of G6PD is crucial to protect
the red cell from oxidative damage. The deficiency is
X linked. Despite the mode of inheritance, females
may have clinical manifestations and be susceptible
to hemolysis. Because of X chromosome inactivation,
heterozygotes have two populations of red cells, one
normal and one G6PD deficient.
The prevalence of G6PD deficiency varies considerably being rare in Northern European populations
to frequencies of 20% in parts of Southern Europe,
Africa, and Asia. A large number of mutations within
the gene for G6PD may result in a deficient phenotype. The majority cause mild deficiency and only
result in significant hemolysis in “stress” situations
such as infection and as a complication of certain
drugs. Rarely individuals have a more severe chronic
non-spherocytic hemolytic anemia. Hemolysis is characterized by the presence of denatured hemoglobin
within the red cell, which can be seen on supravital
staining (Heinz bodies). Diagnosis may be made using

Chapter 3. Inherited red cell disorders

G6PD deficiency screening tests available in the majority of hematology laboratories or by direct quantification, which is available in certain centers and used to
confirm positive screens. For the most part, mild deficiency has little effect on the pregnancy.

Antenatal management

r Determine history of hemolytic episodes and
precipitating factors.
r FBC, blood film for characteristic red cell changes,
serum folate, G6PD assay if not previously tested.
Reticulocyte count, LDH, and bilirubin. Heinz
body preparation is helpful during active
hemolysis.
r Advise against oxidant drugs (see BNF) and
consumption of fresh or lightly cooked broad
(fava) beans. If a drug is felt to be indicated and
there is no alternative, then the risks and benefits
must be taken into account. G6PD deficiency is
heterogenous, patients with a significant history of
hemolytic crises or chronic hemolysis are more
likely to react adversely than those with a milder
phenotype.
r Check folate status and prescribe folic acid 5 mg
daily for all patients with chronic hemolysis.
r Patients should be made aware of the symptoms
and signs of acute hemolytic anemia. Hemolysis is
usually self-limiting, as reticulocytes have higher
enzyme activity. However, red cell transfusion
may be required in severe cases. Occasionally,
renal failure can complicate acute severe
intravascular hemolysis and should be treated as
required.
r Caution with all drugs prescribed to the mother to
ensure there is no associated risk of hemolysis.

Management of the neonate
Neonatal erythrocytes have an increased susceptibility to oxidative hemolysis. Immaturity of hepatic
enzyme systems may enhance the risk of jaundice,
G6PD deficiency has rarely been described as a cause
of Kernicterus. Hemolysis is usually self limiting but
exchange transfusion may be required for those cases
with severe jaundice.
r A cord sample should be taken at birth for
hemoglobin and bilirubin. G6PD assays should
also be performed, although this may be difficult
to interpret.
r Phytomenadione (a fat soluble preparation of
vitamin K) can be administered to the baby in
accordance with normal procedures. (Water
soluble preparations of the vitamin K should be
avoided in view of the possible risk of hemolysis
in newborns, though the evidence for this is
conflicting.)
r Observe over the first 4 days of life for jaundice.
Hemolysis is usually self limiting but exchange
transfusion using G6PD screened blood may be
required in selected cases.

Breast feeding

r The mother should be advised that certain drugs
may be excreted in breast milk and may trigger
hemolysis in a G6PD deficient baby.

Acknowledgment
The authors are grateful for the review and constructive comments of Dr. D Fothergill Consultant Obstetrician, Jessops Hospital for Women, Sheffield.

Section 1. Cellular changes

References/suggested reading
1. Modell B, Harris R, Lane B et al. Informed consent in
genetic screening for thalassemia during pregnancy:
audit from a national confidental inquiry. British
Medical Journal 2000; 320: 337–341.
2. Powars DR. Natural history of sickle cell disease: the
first 10 years. Seminars in Haematology 1975; 12:
267–285.
3. Serjeant GR. Sickle Cell Disease. Oxford: Oxford
University Press, 1992.
4. Howard RJ, Tuck SM, Pearson TC. Pregnancy in sickle
cell disease in the UK: results of a multicentre survey
of the effect of prophylactic blood transfusion on
maternal and fetal outcome. British Journal of
Obstetrics and Gynaecology 1995; 102: 947–951.
5. Smith JA, Espeland M, Bellevue R et al. Pregnancy in
sickle cell disease: experience of the cooperative study
of sickle cell disease. Obstetrics and Gynaecology 1996;
87: 199–204.

6. Rahimy MC, Gango A, Adjou R et al. Effect of active
prenatal management on pregnancy outcome in sickle
cell disease in African setting. Blood 2000; 96:
1685–1689.
7. Serjeant GR, Loy LL, Crowther M et al. Outcome of
pregnancy in homozygous sickle cell disease.
Obstetrics and Gynaecology 2004; 103: 1278–1285.
8. Koshy M, Burd L, Wallace D et al. Prophylactic red cell
transfusion in pregnant patients with sickle cell
disease. A randomized cooperative study. New
England Journal of Medicine 1988; 319: 1447–1452.
9. United Kingdom Thalassemia Society Standards for
the Clinical Care of Children and Adults with
Thalassemia in the UK, 2005.
10. Jensen CE, Tuck SM, Wonke B. Fertility in ␤
Thalassemia major: a report of 16 pregnancies,
preconceptual evaluation and a review of the literature.
British Journal of Obstetrics and Gynaecology 1995;
102: 625–629.

Section 1
Chapter

Cellular changes

Maternal autoimmune cytopenias
Hamish Lyall and Bethan Myers

Introduction
Autoimmune conditions are characterized by the production of antibodies against self-antigens (autoantibodies). Since these conditions often occur during the
second and third decades of life, they may occur during, or predating pregnancy. In these circumstances,
the additional considerations of both the effect of
pregnancy on the disease, and the disease (and its
treatment) on the pregnancy need to be taken into
account.
It is recognized that pregnancy may influence
the course of maternal autoimmune diseases. This
can result in remissions, relapses, or new presentations of these disorders. The pathogenesis of this phenomenon is likely to be related to the hormonal and
complex immunological changes that occur during
pregnancy. Immunological changes in pregnancy are
necessary to prevent rejection of the fetus, which
expresses both paternal as well as maternal antigens. Placental immunology, and modulation of the
systemic immune response, have been identified as
important mechanisms of this immune tolerance. It is
probable that these features have a significant influence
on autoimmune hematological disorders that occur
during pregnancy.
In this chapter, three auto-immune hematological conditions that may complicate pregnancy:
immune/idiopathic thrombocytopenic purpura (ITP),
autoimmune hemolytic anemia (AIHA), and autoimmune neutropenia (AIN) are discussed. They are characterized by the development of an autoantibody specific for a surface antigen on the platelet, erythrocyte,
or neutrophil. Premature cellular destruction occurs,
by reticuloendothelial phagocytosis, T lymphocyte
cytotoxicity, or complement mediated cell lysis.

To date, the relationship between these immune
mechanisms and the immunological changes in
pregnancy is not fully understood.
Cytopenias occur when the enhanced clearance of
the platelet, erythrocyte or neutrophil from the peripheral blood is greater than the bone marrow’s ability to
produce new cells. ITP is by far the most frequently
seen condition. AIN and AIHA rarely occur in pregnancy and few cases are reported in published literature. The three conditions usually occur in isolation
but occasionally may be seen together, e.g. Evans syndrome (ITP and AIHA).
About two-thirds of cases present prior to pregnancy with the diagnosis already established, but the
remaining third present during pregnancy, either as
an incidental finding or less commonly in the symptomatic state. For many women, pregnancy is the first
time that a full blood count (FBC) is performed. Careful evaluation of any abnormal result is required before
an immune cytopenia can be diagnosed.
The majority of autoantibodies implicated in these
disorders are of the IgG subtype, and hence are able
to cross the placenta. Consideration therefore, needs
to be given not just to the implications for the mother,
but also for the developing fetus, and after delivery, the
neonate.
Management is difficult because, where treatment
is required, there are no agents which are universally efficacious and all carry the potential for adverse
effects. As with all therapies in pregnancy, the benefits of treatment compared with the relative risks to
mother and baby have to be considered. A multidisciplinary approach, combining expertise from obstetricians, hematologists, anesthetists, and neonatologists
is required for optimal care.

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

Section 1. Cellular changes

Idiopathic/immune thrombocytopenic
purpura (ITP)
Introduction
ITP is usually a chronic condition in adults, often
occurring in young women, and can be challenging
to diagnose and manage in pregnancy. Although it
is principally mediated by autoantibodies, the development of specific assays as a diagnostic tool has, to
date, proved unsuccessful. Therefore, the diagnosis is
predominantly one of exclusion with frequent difficulty in excluding alternative causes of thrombocytopenia. Fortuitously, the risk of major hemorrhagic
complications is low. Successful management requires
maintaining adequate platelet counts for pregnancy
and delivery whilst minimizing the risks of treatmentrelated side effects for mother and baby. Potential risks
of fetal thrombocytopenia need to be appreciated and
measures taken to prevent hemorrhagic complications
at delivery.

Epidemiology
The annual incidence, of acute and chronic ITP in
adults, from population-based studies is estimated as
2–4 per 100 000, when defined using a platelet count of
less than 100 × 109 /L. These incidence figures are similar for Europe and the USA.1 In keeping with other
immune disorders, it is more common in women than
men (F:M 1.7–1.9:1), and frequently occurs during
the reproductive years, occurring in all ethnic groups.
The incidence in pregnancy has been estimated at 0.1–
1 per 1000 pregnancies,1,2 accounting for about 3%3
of cases of thrombocytopenia in pregnancy. Approximately two-thirds of cases of ITP already have an
established diagnosis prior to pregnancy, allowing the
opportunity for pre-pregnancy counseling and planning for a future pregnancy.

Pathogenesis

Thrombocytopenia is predominantly caused by
autoantibodies specific for platelet glycoproteins
binding to platelets in the maternal circulation. This
results in immune mediated platelet destruction. The
immune dysregulation which permits autoantibody
formation is still the subject of much research. More
recently it has been found that, in addition to increased
destruction of platelets, there is also suppression of
megakaryopoiesis in the bone marrow. Therapeutic

agents targeting this phenomenon are now licensed
for use in the non-pregnant setting. There is usually no
apparent stimulus for the autoantibody production;
however, occasionally a history of recent viral illness
or drug exposure can be implicated. ITP usually
occurs in isolation but may occur with other immune
cytopenias or be secondary to a systemic autoimmune
condition, e.g. SLE. The spleen has an important
role in ITP, being both a major source of antibody
production and the predominant site for destruction
of antibody-bound platelets. The antibodies are of
the IgG subtype and therefore able to cross the placenta and potentially cause thrombocytopenia in the
fetus/neonate.

Diagnosis
Thrombocytopenia in pregnancy
The reference range for platelet counts outwith pregnancy is 150–400 × 109 /L. During pregnancy there
is a general trend downwards in platelet count, especially in the last trimester, resulting in a fall of around
10% from the pre-pregnancy level.4,5 This is thought
to be due to accelerated destruction of platelets and
normal physiological dilutional effects. For the majority of women this will not result in the platelet count
falling below the normal laboratory range. However,
if the pre-pregnancy platelet count lies at the lower
end of the normal range, or if there is a more severe
drop in counts, thrombocytopenia occurs. The finding of mild thrombocytopenia in pregnancy is common, with approximately 8%–10% of women having a
platelet count below the laboratory normal range.4
Since the diagnosis of ITP is one of exclusion (when
presenting during pregnancy), alternative diagnoses
must be considered and excluded where possible. The
principal differential diagnoses of thrombocytopenia
in pregnancy are discussed below and are summarized
in Table 4.1.
Gestational thrombocytopenia
The majority of cases of thrombocytopenia in pregnancy (74%) are attributable to gestational thrombocytopenia (incidental thrombocytopenia) of pregnancy.5 This is a benign condition and represents
no bleeding risk to mother or fetus. It probably reflects the extreme end of the normal physiological effect described above. It typically occurs in
the third trimester and usually results in a mild
thrombocytopenia. Platelet counts below 70 × 109 /L

Chapter 4. Maternal autoimmune cytopenias

Table 4.1 Causes of thrombocytopenia in pregnancy

Thrombocytopenic condition

Pathogenesis of thrombocytopenia

Diagnostic characteristics

Gestational thrombocytopenia

Physiological dilution
Accelerated destruction

Third trimester, plts ⬎70 × 109 /L
Incidental finding,
no features of other disease

HIP/ Pre-eclampsia/eclampsia

Peripheral consumption

Unwell patient
clinical features – hypertension, proteinuria,
neurological signs/symptoms

Micoangiopathic hemolytic anemias
(MAHA) – TTP, HUS, HELLP syndrome

Mechanical destruction and peripheral
consumption (accumulation of
micro-thrombi in small vessels)

Unwell patient.
Clinical features – neurological signs, fever,
renal impairment, deranged LFTs, hemolysis

ITP

Immune mediated peripheral consumption
and occasional bone marrow suppression

Absence of other causes of thrombocytopenia
Diagnosis of exclusion

Hereditary thrombocytopenia

Bone marrow underproduction

Family history
Somatic abnormalities
Abnormal blood film

Leukemia/lymphoma

Bone marrow infiltration

Lymphadenopathy, hepatosplenomegaly,
Other FBC abnormalities

Pseudothrombocytopenia

EDTA artefact

Platelet clumping seen on blood film

Viral infection

Multifactorial

Recent viral illness. Risk factors

Drugs

Multifactorial

Timing of drug exposure

HIP: Hypertension in pregnancy, TTP: Thrombotic Thrombocytopenic purpura, HUS: Hemolytic uremic syndrome, HELLP: Hemolysis with
elevated liver enzymes and low platelets, ITP: Immune thrombocytopenic purpura

should alert the physician to consider alternative diagnoses, although in rare cases the diagnosis has been
subsequently confirmed in women with counts as
low as 50 × 109 /L.6 Gestational thrombocytopenia
is not immune mediated and therefore poses no risk
to the fetus. A platelet count that has been normal
before pregnancy, and normal in the first and second
trimesters is useful in helping make the diagnosis. The
FBC returns to normal within a few weeks of delivery.
It may cause diagnostic difficulty with ITP when there
are no pre-pregnancy counts.
Hypertensive disorders
Hypertensive disorders of pregnancy complicate
between 12%–22% of pregnancies, and are a common
cause of thrombocytopenia in pregnancy, accounting
for approximately 20% of cases. “Gestational hypertension,” which includes “hypertension in pregnancy”
(HIP), pre-eclampsia and eclampsia, is responsible for
the vast majority of these. Thrombotic thrombocytopenic purpura (TTP), hemolytic uremic syndrome
(HUS) and Hemolysis, Elevated Liver Enzymes and
Low Platelets (HELLP) syndrome can share similar
features with pre-eclampsia, and distinguishing
between these conditions is sometimes problematic.
Together, these rarer microangiopathic hemolytic

anemia (MAHA) conditions cause less than 1% of
pregnancy-related thrombocytopenia. Management
of these conditions is described in Chapters 17 and
18. Hypertensive disorders may be associated with the
disseminated intravascular coagulation (DIC), which
will contribute to further platelet reduction.
Constitutional thrombocytopenia
Hereditary causes of thrombocytopenia are rare,
accounting for less than 1% of cases. This includes
MYH-9 disorders characterized by giant platelets on
the blood film and Dohle body inclusions in neutrophils. Of these, the May Hegglin anomaly is the
most likely to be encountered in pregnancy. Rarely,
hereditary bone marrow failure syndromes such as
Fanconi’s anemia may present with an isolated thrombocytopenia in pregnancy. These diagnoses may be
suspected if there is a family history of thrombocytopenia, unexplained thrombocytopenia in more than
two first-degree relatives, or physical abnormalities
suggestive of the disorder.
Drugs and infections
Thrombocytopenia is a frequently occurring side
effect of many medications. Heparin induced thrombocytopenia, a potentially life-threatening condition,

Section 1. Cellular changes

may occur rarely with unfractionated heparin use in
pregnancy, but to date has not been described with
low molecular weight heparin therapy in pregnancy.
As with the non-pregnant setting, viral infection is
an important cause of thrombocytopenia. Whilst this
occurs as a transient phenomenon with many viruses,
specific consideration should be given to hepatitis and
HIV infection, particularly if risk factors are present.
Thrombocytopenia in this setting is likely to be multifactorial with both an immune and non-immune
pathogenesis. Diagnosing these infections early in
pregnancy may allow treatment to be initiated, reducing the risk of related complications and vertical transmission.
Others
Hematological malignancies can present in pregnancy
and the initial feature may be isolated thrombocytopenia. Occasionally, a bone marrow examination may be
required to exclude these.
Laboratory artifact from EDTA present in the sample tubes may account for some cases of apparent
thrombocytopenia. Examination of the blood film is
essential to exclude this possibility.
There are no specific diagnostic tests for ITP.
Although platelet glycoprotein specific antibodies can
be detected in the majority of cases, this test lacks the
sensitivity and specificity to be of clinical use. Diagnostic parameters for ITP in pregnancy are: thrombocytopenia with a past history of ITP, or a platelet count
during pregnancy of less than 70 × 109 /L with other
causes excluded. Mild thrombocytopenia presenting
in the first or second trimesters may also represent
ITP, but this is not clinically significant for the mother
since no treatment is required. In all cases, a careful
history and examination of the blood film are critical
to the evaluation of thrombocytopenia and diagnosing
ITP in pregnancy. The important clinical and laboratory points for diagnosis are discussed below and listed
inTable 4.2.

History
r Where there is a preceding history of ITP, check
diagnosis for accuracy.
r Documented response to corticosteroids or
intravenous immunoglobulin (IVIG) is usually
diagnostic of ITP. In addition, this information is
valuable for deciding on future treatment.

r Previous pregnancy experience and any
documented blood counts both during and
outside of pregnancy are very useful.
r Neonatal platelet counts from previous successful
pregnancies should be noted.
r Note any illnesses associated with ITP (e.g. SLE),
or the occurrence of other autoimmune disorders
in the patient.
r A family history of thrombocytopenia may
suggest a hereditary disorder.
r Identify any risk factors for HIV and viral
hepatitis, and include the relevant tests.
r Any current medications should be considered for
the possibility of drug-induced
thrombocytopenia.
Clinical examination
r Clinical examination is occasionally of value.
r The presence of purpura or mucosal bleeding
should be sought.
r Splenomegaly and/or lymphadenopathy are not
characteristic of ITP.
r Physical abnormalities may suggest a hereditary
disorder.
Laboratory assessment
r FBC: Incidental finding of thrombocytopenia
should prompt a recheck of the FBC.
r Blood film is essential to exclude alternative
diagnoses:
(a) Spurious thrombocytopenia is caused either by
EDTA artifact (causing platelet clumps – if
present, repeat count in citrate sample) or platelet
satellitism. Both are readily seen on the film.
(b) Check erythroid and leukocyte morphology and
confirm within normal limits. Abnormal red cell
(e.g. fragmentation) or white cell morphology
suggests alternative diagnosis.
(c) Check platelet morphology. Giant platelets may be
seen in ITP but, if this is the dominant finding,
consider a MYH-9 disorder and examine the
neutrophils for Döhle bodies. Giant platelets and
thrombocytopenia may also be seen with Bernard
Soulier disease, but a lifelong history of abnormal
bleeding would be expected. Abnormally small
platelets may be seen with hereditary
thrombocytopenia and bone marrow failure

Chapter 4. Maternal autoimmune cytopenias

Table 4.2 Evaluation of suspected ITP

Specific point to elicit

Relevance

Current history

Is patient hemorrhagic?
Viral illness/ risk factors for HIV or hepatitis

Thrombocytopenia genuine. Indication for treatment
Viral cause for thrombocytopenia;
Check serology

Family history

Family history of unexplained
thrombocytopenia

Consider hereditary causes

Past medical history

Known ITP

ITP likely cause thrombocytopenia.
? previous response to steroids or immunoglobulin

SLE, thyroid or other autoimmune
disorders

ITP likely cause thrombocytopenia. Possibility of SLE related
complications

History of pre-eclampsia or previous
thrombocytopenia in pregnancy
Previous baby with neonatal
thrombocytopenia

Increased likelihood of recurrence
Establish cause

Past obstetric history

ITP likely. If not, consider possibility of neonatal alloimmune
thrombocytopenia (NAIT)(see chapter 5A)
May predict risk future neonatal thrombocytopenia
Clinical examination

Mucocutaneous bleeding
Lymphadenopathy, hepatosplenomegaly

Thrombocytopenia genuine; indication for treatment
Possible leukemia/lymphoma
Not consistent with ITP

Laboratory assessment

Platelets: clumping present?
Giant platelets
Normal red cell and white cell numbers
and normal morphology
Schistocytes/red cell fragments
LFTs abnormal
coagulation screen abnormal

Pseudothrombocytopenia; repeat FBC in citrate
Check platelet count by alternative method. Consider MYH-9
disorders
Consistent with gestational thrombocytopenia or ITP
Consider MAHA
Consistent with HELLP syndrome
Consider DIC, hypertensive disorders of pregnancy

syndromes. Automated platelet counts may be
erroneous if they are performed on an analyzer
that relies on impedence counting and if the
platelets are very large. This should be suspected if
the film appearances differ significantly from the
analyzer result. An alternative method of platelet
measurement available on some analyzers (e.g.
flow cytometry platelet count) may give a more
accurate measurement.
Some analyzers are able to measure reticulated
platelets and this percentage increases significantly in
ITP.
r Other routine investigations which should be
performed are listed in Table 4.2.
Bone marrow examination
r A bone marrow examination can confirm the
presence of normal megakaryocytes, normal
hematopoiesis, and the absence of bone marrow
infiltration.
r This is not necessary for younger patients where
there are no other clinical or laboratory features to

suggest bone marrow failure or infiltration (BSCH
guidelines 2003).
r Consider performing a bone marrow examination
in cases that do not respond to standard
treatments.
r NB: bone marrow examination will not
differentiate between ITP and gestational
thrombocytopenia or other consumptive causes,
which constitute the main differential diagnoses,
only confirming that thrombocytopenia is due to
peripheral consumption.

Management
The aim of management of ITP in pregnancy is not
to achieve a sustained normal platelet count but simply to maintain a platelet count which is adequate to
avoid hemorrhagic complications during pregnancy,
delivery and immediately postpartum. This conservative approach minimizes the risks of maternal and
fetal exposure to therapeutic agents. There are no
universally accepted criteria for “safe” platelet counts
in pregnancy. It is advisable that members of the
team involved in managing these cases (obstetricians,

Section 1. Cellular changes

Table 4.3 Suggested platelet thresholds for intervention

Table 4.4a Corticosteroids

Intervention

Platelet count

Advantage

Disadvantage

Ante-natal, no invasive procedure planned

⬎20 × 109 /L

Vaginal delivery

r
r
r

⬎40 × 109 /L

Operative or instrumental delivery

⬎50 × 109 /L

Epidural anesthesia

⬎80 × 109 /L
r

hematologists, anesthetists) agree a consensus for minimum accepted platelet thresholds. Generally, these
can be low in the antenatal period if the patient is not
hemorrhagic. Thresholds typically need to be higher
for delivery. Suggested platelet thresholds for ITP are
stipulated in Table 4.3.
Monitoring during pregnancy
Platelet counts in women with ITP need to be closely
monitored through pregnancy: in general, monthly in
the first and second trimesters, 2-weekly in the third,
and weekly near term, although the frequency of monitoring will depend on the rate of change as well as
absolute values.

Treatment
There are two decisions to be made in treating ITP
in pregnancy: when to treat and what treatment to
give. The majority of women will not require therapy throughout the whole duration of the ante-natal
period.7,8 Only women with very low platelet counts
(⬍20 × 109 /L) or who are hemorrhagic will require
treatment at this stage. By contrast, treatment is
often required to raise the platelet count prior to
delivery. The two treatment options for the initial
management of ITP usually considered are corticosteroids and intraveous immunoglobulin (IVIG); antiD immunoglobulin appears to have equivalent efficacy
to IVIG, and could be considered as an alternative
in non-splenectomized Rhesus positive patients.9 The
characteristics of these agents are summarized in Table
4.4a, 4.4b and 4.4c. The choice of which agent to use
requires discussion with the individual about the relative risks and benefits of each treatment. Some authorities advocate first-line therapy with IVIG rather than
corticosteroids. Currently, there is little experience of
using Anti D in pregnancy for ITP; however, it is
widely used in Rhesus D negative women for the prevention of hemolytic disease of the newborn (HDN).
It should be noted that the dose for ITP is substantially higher than for HDN and this may result in an

r
r

Oral therapy
Most experience
Can be used for extended
periods if prolonged
platelet count rise is
required
Dose can be tapered to
minimum required for
desired effect.
Not a blood product
Inexpensive

r
r
r
r
r

Risk of gestational
diabetes mellitus
Immunosuppressive
Slow response 3–7 days
for first response, maximal
response 2–3 weeks
Risk of osteoporosis with
prolonged therapy
Risk of hypertension
Possible adverse effects
on fetus at high doses
(but 90% metabolized)

Table 4.4b Intravenous immunoglobulin (IVIG)

Advantage

Disadvantage

r
r

Established therapy
Response to
treatment is rapid
(6–72 hours)
No corticosteroid
side effects
Low risk to fetus

r
r
r
r
r

Intravenous therapy with long
duration of administration
Pooled plasma product
therefore potential risk of
pathogen transmission for
mother and fetus
Transient response (⬍1 month)
Risk of infusional reactions
Risk of aseptic meningitis
Headache common
Expensive

Table 4.4c Anti D immunoglobulin

Advantage

Disadvantage

r
r
r

Short administration
period (3–15 min)
Good reported efficacy
Non-immunosuppressive
No corticosteroid side
effects

r
r
r

r
r

Limited experience in
pregnancy
Transient response ⬍ 1
month
Occasionally may induce
significant hemolysis
Pooled plasma product
therefore potential risk of
pathogen transmission for
mother and fetus
Crosses the placenta.
Fetus may be at risk of
hemolysis
Only available to patients
who are Rh positive
(approx. 90% of
individuals)
No efficacy if prior
splenectomy

increased risk of neonatal hemolysis. Currently, there
are no anti D preparations licensed in the UK for
the treatment of ITP. Patients with contraindications
to corticosteroids (diabetes mellitus, concurrent infections, history of steroid psychosis) should be managed
with IVIG or Anti D alone.

Chapter 4. Maternal autoimmune cytopenias

Fig. 4.1 Algorithm for initial ITP therapy
∗ IVIG 0.4 g/kg for 5 days or 1 g/kg for 2
days
Anti D 50–70 mcg/kg single dose
Consider methylprednisolone 1 g IV in
addition to IVIG or Anti D.
∗∗ Reassess earlier if obstetric indications
that might need early delivery

Assessment at 36
weeks **

<36 weeks

Platelets >20 and
asymptomatic?

Yes

Monitor FBC until
treatment required

Platelets > 50
and stable

Yes

Treatment required

Prednisolone 20 mg
OD for 1 week

Yes

Response to
treatment?

No
Taper
prednisolone to
lowest effective
dose

Consider alternative
therapies

The choice of therapy depends on the following factors:
r the speed with which a platelet increment is
required;
r the length of time for which a rise needs to be
sustained;
r which therapy carries the least potential risk for a
given individual.
A suggested algorithm for initial therapy is shown in
Fig. 4.1. This algorithm is only suitable for uncomplicated cases. Suggested management for various scenarios are listed below.
Patients with moderate/severe thrombocytopenia (⬍20 ×109 /L)
r Prednisolone 20 mg/day10 (it is common practice
to use less than the 1 mg/kg to avoid adverse
effects).

Prednisolone 60 mg
OD for 1 week

Yes

Response to
treatment?

Yes

Response to
treatment?

Continue Prednisolone.
Add IVIG or Anti D*

Patients with very severe thrombocytopenia (≤10× 109 /L)
significant major bleeding:
r requires treatment to raise the platelet count
urgently;
r IVIG +/− high dose corticosteroids (usually
60 mg daily);
r consider platelet transfusions if significant
bleeding.

Patients with life-threatening bleeding
r platelet transfusion + ;
r IVIG + IV methyl-prednisolone;

Where possible, the dose of prednisolone should be
promptly reduced to the minimum effective dose.
Unless hemorrhage is a major feature, prolonged
therapy (⬎6 weeks) with high doses of prednisolone
is considered to carry too high a risk of adverse
events for the mother. The response to IVIG or anti

Section 1. Cellular changes

D is often transient. Patients receiving these treatments may require repeat infusions. The addition of
methylprednisolone 1 g intravenously is used to speed
up/improve the response as compared to standard
prednisolone in difficult or refractory cases.
It is not always possible to achieve the desired
platelet count in individuals with ITP. Many patients
(35% in one series) diagnosed with ITP in pregnancy
will not respond to corticosteroids or IVIG. In addition, response to platelet transfusions are transient
with poor increments as circulating antibody rapidly
clears transfused platelets.

r Rituximab: this agent is an anti-CD20
monoclonal antibody, which is increasingly used
to treat non-pregnancy related ITP. However,
there is insufficient evidence regarding safety and
efficacy to advocate its use during pregnancy. The
manufacturer currently recommends avoiding
pregnancy for 1 year following treatment.
Other agents, which are useful outside pregnancy,
such as androgen analogs (e.g. danazol), and cytotoxic
agents such as cyclophosphamide or vinca alkaloids,
are contraindicated in pregnancy.

General measures – the following should be avoided:
Management of refractory cases

r In considering other therapeutic options the
balance of risks need to be considered between
treatment-related toxic effects vs. risk of major
bleeding with prolonged severe
thrombocytopenia. In many circumstances it may
be preferable or necessary to accept the increased
hemorrhagic risk of significant thrombocytopenia
rather than use more aggressive therapies.
r Splenectomy: this procedure has a well
established, though diminishing, role in ITP. It
can generally be performed safely in pregnancy
but carries the risks of general surgery and of fetal
loss. Where possible, it should be performed in
the second trimester. This avoids the risks of
teratogenicity associated with drugs in the first
trimester. In the third trimester the gravid uterus
may make splenectomy technically more
demanding, although laparoscopic splenectomy
may make the procedure more feasible.
r Tranexamic acid: this is an antifibrinolytic,
normally avoided in pregnancy because of
concerns that it may increase thrombotic risk.
Reproductive animal studies do not indicate risk
to the fetus, but there are no adequate and
well-controlled studies done on pregnant women
(category B). It could be considered in the
refractory patient with ongoing symptoms, after
the first trimester.
r Azathioprine is used as a second-line agent, and
has been given safely in pregnancy, but there is
insufficient evidence to currently advocate its
routine use in this setting. It has a slow onset of
action (about 8 weeks), which also reduces its
utility.

r aspirin and non-steroidal medication;
r intramuscular injections;
r strenuous activity.

Planning for delivery
Consideration of potential maternal and neonatal
thrombocytopenia is required in addition to any
obstetric factors that may be present when planning
delivery.

Maternal considerations
The principal concern is hemorrhage. This may be during delivery or postpartum. Postpartum hemorrhage is
of particular concern due to the sharp fall in procoagulant factors that occurs at this time. As discussed above,
there is no universally agreed safe platelet count; however, hemorrhage caused by thrombocytopenia occurring at a platelet count ⬎50 × 109 /L would be considered unusual.
Epidural analgesia is of particular concern, as even
a small increase in venous hemorrhage could have the
potential for spinal cord compression. The risk is considered to be greatest at the time of insertion and withdrawal of the catheter. There is controversy over the
safe threshold for epidural anesthesia; there is some
evidence to suggest that a platelet count of 50 × 109 /L
is adequate (based on British Society of Haematology guidelines), however anesthetic practice is to use a
threshold of at least 80 × 109 /L, in experienced hands
(based on BCSH and anesthetic guidelines). A predelivery anesthetic consultation is helpful to discuss
alternative analgesia during labor. The role of spinal
anesthetic is more difficult. This procedure may allow a
cesarean section to be performed without the need for

Chapter 4. Maternal autoimmune cytopenias

a general anesthetic. A decision may be taken that the
risks of a single pass spinal needle could be less than
those of a general anesthetic in some situations and, if
an experienced obstetric anesthetist is available, a cutoff of 50 × 109 /L is suggested.
Chronic immunosuppression antenatally for ITP
may increase the risks of postpartum sepsis.

Neonatal considerations
The principal neonatal risk is intracranial hemorrhage
due to severe thrombocytopenia and birth trauma.
This is rare (⬍1% of ITP cases), although potentially
devastating when it occurs. The overall incidence of
thrombocytopenia in neonates born to mothers with
ITP is reported in various studies as 14%–37.5%.8,11,12
However, only approximately 5% of babies born to
mothers with ITP will have platelet counts ⬍20 ×
109 /L, with a further 5% having counts between 20 ×
109 –50 × 109 /L.8,13
Unfortunately, predicting which babies may be
affected or directly assessing the fetal platelet count is
difficult. No correlation has been established with the
severity of maternal ITP or levels of circulating antibody. Although there are no reliable predictors of its
occurrence or severity, neonatal thrombocytopenia is
more likely if:
r there is a previous sibling with
thrombocytopenia.13
r the mother has had a splenectomy prior to this
pregnancy (although not all studies confirmed
this finding).
r severe maternal ITP.13,14
Where babies have been born previously with severe
thrombocytopenia, testing for paternal platelet antigen incompatibility to exclude Neonatal alloimmune
thrombocytopenia (NAIT) is required.
There is currently little role for the routine measuring of fetal platelet counts by percutaneous umbilical
blood sampling (PUBS) in ITP. Studies evaluating this
technique have estimated the procedure-related risk to
be greater than the risk of preventing neonatal hemorrhage. Platelet counts taken from fetal scalp samples are prone to erroneously low results, and carry
the risk of scalp hematoma, and are therefore best
avoided.

Mode of delivery
Concerns regarding potential neonatal thrombocytopenia and birth trauma have previously led some

clinicians to recommend cesarean section. There is
currently no evidence that cesarean section reduces
the incidence of intracranial hemorrhage in susceptible babies compared with an uncomplicated vaginal delivery. This is true for congenital bleeding disorders as well as ITP. For this reason it is recommended
that the mode of delivery is determined by obstetric
indications rather than ITP. However, vaginal delivery that is augmented by ventouse or rotational forceps does carry an increased risk of head trauma to
the neonate and where possible should be avoided.
Induction of labor at the time of maximal platelet
count may be required if platelet count rises are very
transient with therapy. The exact mode and timing of
delivery has many patient-specific variables and therefore an individualized plan with multidisciplinary
input is advised.

Management of labor when platelet count
has not been corrected
In these cases a pragmatic approach needs to be taken.
Experiences suggest that normal delivery can occur
without excess hemorrhage, reassuringly, even at very
low platelet counts. It is advisable to have platelet transfusions available on standby and to proceed with delivery. If time allows, high dose IVIG (1 g/kg) may be
used. Epidural anesthesia should be avoided as should
non-steroidal anti-inflammatory (NSAIDs) drugs for
postpartum pain relief.

Postpartum – neonatal care
The neonatal team should be alerted prior to delivery. A cord platelet count should be measured at
birth. If the platelet count is normal, further neonatal platelet counts are not required. If thrombocytopenia is present, this should be confirmed on a capillary
or venous sample. Intramuscular injections are best
avoided, if severe thrombocytopenia is present, and
vitamin K given orally.
Further alternate-daily FBC measurements over
the next week are required to ensure that the neonate
is not at risk of hemorrhage. The nadir platelet count
is usually between days 2 and 5.
Babies with severe thrombocytopenia of ⬍20 ×
109 /L or clinical hemorrhage require treatment
with IVIG. Life-threatening complications should
be treated with immediate platelet transfusions and
IVIG. Consideration should be given to using HPA

Section 1. Cellular changes

1a 5b negative platelets if available until NAIT is
excluded.
Babies with severe thrombocytopenia should have
a cranial ultrasound to assess for evidence of intracranial hemorrhage.

Prenatal counseling
Women who have an established diagnosis of ITP may
request pre-natal counseling before deciding whether
to embark on a pregnancy. There are few predictors of
outcome that can be used to assess risk. While pregnancy should not be discouraged, it is suggested that
the following points should be discussed:
r Circulating antiplatelet antibodies may still be

r
r
r
r
r

r
r

present in the maternal blood. This is particularly
relevant for women who have had a splenectomy.
In these circumstances the ITP may appear in
remission with normal platelet counts. However,
this is primarily due to an inability to clear
platelet–antibody complexes rather than a
cessation of antibody production. These women
will still be at risk of neonatal thrombocytopenia
or hemorrhagic complications in utero.
ITP may relapse or worsen during pregnancy.
If treatment of ITP is required it will carry both
maternal and fetal risks.
There is an increased risk of hemorrhage at
delivery, but the risk is small even if the platelet
count is low.
Epidural anesthesia may not be possible.
Although it is not possible to accurately predict if
a neonate will be affected, the risk is high if a
sibling had thrombocytopenia, or mother had
undergone splenectomy.
Maternal death or serious adverse outcomes for
mothers with ITP are rare.
The risk of intracranial hemorrhage for the
fetus/neonate is very low.

Autoimmune neutropenia (AIN)
Introduction

Neutropenia is a common finding in routine FBC testing, and is defined as an absolute neutrophil count
(ANC) of ⬍1.5 × 109 /L (or ⬍ 1.2 for some ethnic
groups – see below). The majority of cases are mild,
transient, and no specific etiology is determined. By
contrast, AIN is a rare disorder and can cause severe

neutropenia associated with recurrent infection.15 It
may occur in isolation or in conjunction with ITP
or AIHA. Many cases in adults are secondary, associated with collagen vascular disorders, rheumatoid
conditions, and SLE. Primary AIN is predominantly
a disease of childhood. The main complication of this
condition is recurrent infection, which occurs if the
neutropenia is severe (ANC ⬍ 0.5 × 109 /L). Diagnosis can be problematic as laboratory investigation of
neutropenia is limited, and usually restricted to specialist centers. Pregnancy poses an additional problem, as autoantibodies may cross the placenta resulting in neonatal neutropenia after delivery. Currently,
published evidence on management of these cases is
lacking.

Incidence and pathogeneisis
The true incidence of AIN is not known. Persistent
neutropenia in adults is a common finding and is
frequently not investigated if asymptomatic and mild
(ANC 1.0 × 109 –2.0 × 109 /L). Cases are often labeled
as chronic idiopathic neutropenia (CIN). It is probable that some cases with a presumptive diagnosis
of CIN are immune mediated. The benign nature of
asymptomatic CIN means that specialist investigation is often of little value and immunological studies are therefore not pursued. This may not be the
case for women of childbearing age, as identification
of immune-mediated cases may help with neonatal
assessment.
The pathogenesis of AIN is similar to that of other
immune cytopenias. It is an acquired disorder in which
autoantibodies specific for neutrophil surface glycoproteins result in reduced neutrophil survival and
neutropenia.

Diagnosis
Patients with symptomatic neutropenia (recurrent
infections, severe neutropenia) are likely to present
outside of pregnancy and have an established diagnosis. Difficulty occurs in the asymptomatic patient
if an incidental finding of neutropenia is made following FBC testing during routine ante-natal care.
Assessment involves a careful history, examination of
the other FBC indices and inspection of the blood
film.
The differential diagnosis includes: drugs, viral
infections, immune mediated disorders, large granular lymphocyte (LGL) disease (often associated with

Chapter 4. Maternal autoimmune cytopenias

Table 4.5 Severity of neutropenia according to the ANC

ANC

Severity

Clinical effect

⬎1.0 × 109 /L

Mild

Usually asymptomatic

0.5 × 109 –1.0 × 109 /L

Moderate

Usually asymptomatic

0.2 ×109 –0.5 × 109 /L

Severe

Infections possible

⬍0.2 × 109 /L

Very Severe

High risk of infection

rheumatoid arthritis), benign ethnic neutropenia, and
CIN. Important clinical and laboratory aids to diagnosis are listed below.

History

r History of SLE, rheumatoid arthritis or other
autoimmune disease suggests secondary immune
neutropenia. (more common than primary AIN
in adults.)
r Ethnic origin (ANC ⬎ 1.2 × 109 /L may be
considered within normal limits for some African,
Middle Eastern and Yemenite Jew populations).
r Ask about any recent viral illness.
r Assess risk factors for HIV.
r Assess for evidence of recurrent infections,
particularly unusual infections or mouth ulcers (if
there is a temporal pattern – consider cyclical
neutropenia).
r Take a careful drug history (especially antithyroid
drugs, phenothiazines, and NSAIDs), which are
known to cause neutropenia.
r Is there a known family history of neutropenia?
r Is there a history of ITP or AIHA?

Laboratory assessment

r A blood film should be examined to confirm
neutropenia. Severity may be graded using the
criteria in Table 4.5.
r Increase in LGLs on the blood film should be
noted.
r The presence of abnormalities other than
neutropenia suggests an alternative diagnosis to
AIN.
r Asymptomatic cases with ANC ⬎0.5 × 109 /L and
where there is no apparent cause are best managed
by repeating the test after 4 weeks. Further
investigation during pregnancy is warranted if the
neutropenia persists, or if the patient is
symptomatic.
r Anti-neutrophil antibody results also produce
frequent false negatives and positives, similar to

anti-platelet antibodies, making the test of little
use. Repeat samples may help diagnosis in some
cases.
r A bone marrow examination is of value in cases of
severe neutropenia. The bone marrow
appearances in AIN may show normal
hematopoiesis or an apparent arrest at the
metamyelocyte stage with a reduction in the
number of mature neutrophils and band forms.

Management
There are two main risks during pregnancy – the
maternal risk of sepsis and the risk of neonatal neutropenia. Sepsis in pregnancy may provoke miscarriage or premature labor and is the main concern,
for example, a normally benign urinary infection may
progress to pyelonephritis and septicemia.
Information on neonatal outcomes in women
with AIN is limited. Neutropenia from all causes
in neonates is common. Information from neonates
affected by neonatal alloimmune neutropenia (NAIN)
suggests that infections are, in the main, mild and
death or serious morbidity from sepsis is very rare.
As with ITP, steroids are the usual first line of treatment, if required. IVIG may be given if no response.

Sepsis
Sepsis in individuals with severe neutropenia is an
emergency. Untreated sepsis in the setting carries a
significant mortality for both mother and baby. Blood
cultures should be taken and broad-spectrum intravenous antibiotics commenced promptly according to
local protocols. Fetuses tolerate pyrexia poorly, and
neurological damage may occur if the baby suffers prolonged fever.

Granulocyte colony stimulating factor (GCSF)
GCSF has significantly changed the management of
severe chronic neutropenia. For many individuals the
administration of low doses of GCSF 2–3 times per
week substantially reduces the incidence of infection.
GCSF has replaced traditional therapies such as IVIG,
corticosteroids or splenectomy as first line therapy
outside of pregnancy. Long-term follow-up to date
has suggested that this is a safe treatment and therefore patients with symptomatic neutropenia are often
on regular therapy.15 It is not yet clear that GCSF
is safe for use in pregnancy. Studies investigating
prematurity have noted a potential association with

Section 1. Cellular changes

spontaneous preterm birth and elevated cytokines
including endogenous GCSF. In addition, GCSF carries a small risk of venous thrombo-embolism, which
may constitute a significant risk factor for some pregnancies. It is known that GCSF crosses the placenta.
Despite these reservations, it is likely that GCSF is relatively safe in pregnancy. Several cases of successful
pregnancy with continuation of GCSF in pregnancy
are documented in the published international severe
chronic neutropenia registry and this is supported by
individual case reports.

Postpartum
Women with proven AIN or where AIN is strongly
suspected are at risk of delivering a neutropenic baby.
The neutrophil count at birth should be measured and
subsequent measurements performed according to the
degree of neutropenia and the infection risk. Immune
neutropenia may take several weeks to resolve.

Practical approach to pregnant patients
with diagnosed AIN

r Individuals who are asymptomatic are unlikely to
benefit from specific therapy.
r If the ANC is ⬍ 0.5 × 109 /L, advice on treating
sepsis promptly with intravenous antibiotics is
required.
r Individuals who are symptomatic and already on
GCSF may benefit from continuing therapy but a
careful discussion of the risks of therapy is
necessary. Consideration can be given to stopping
GCSF, particularly for the first trimester.
r Monitoring FBC to tailor GCSF dose may be
required.
r The neonatal team should be alerted prior to
delivery.
r A cord blood sample should be taken.
r A postpartum FBC should be sent.

Management of a newly presenting case of neutropenia
in pregnancy

r Exclude other causes of neutropenia.
r Check hematinics – (ferritin, B12 and folate – see
Chapter 2).
r Assess for evidence of associated auto-immune
conditions.
r If severe neutropenia, warn patient of risk of
life-threatening infection – ensure they

understand that prompt treatment is necessary,
and have clear, efficient self-referral route.
r Treatment options should be discussed: steroids
are first-line choice in pregnancy, with IVIG and
GCSF as second- and third-line options if no
response.

Autoimmune hemolytic anemia (AIHA)
Introduction
Hemolysis is defined as shortened red cell survival,
the average lifespan of an erythrocyte being 120 days.
Mild hemolysis is compensated for by an increase in
bone marrow erythropoeisis and may not affect the
hemoglobin concentration. Anemia occurs when red
cell survival is sufficiently shortened to exceed this
increase in erythropoetic activity. Causes of hemolysis are listed in Table 4.6. AIHA is a common cause
of hemolysis but rarely complicates pregnancy. Nonimmune hemolysis occurs more frequently in pregnancy and is mostly associated with pre-eclampsia or
other hypertension-related disorders. It is essential to
distinguish between these types of hemolysis as the
management is very different. AIHA may be further
divided into “warm” and “cold” types. Warm AIHA is
usually IgG mediated. Cold AIHA is mostly IgM and
complement mediated. The blood film appearances
and direct antiglobulin test (DAT) are characteristic.
Treatment of AIHA in pregnancy is similar to outside
pregnancy. Transplacental passage of IgG antibodies
may occur, but neonatal hemolysis is rarely severe.

Epidemiology and pathogenesis
AIHA in pregnancy is a rare disorder with an estimated incidence of 1:50 000 pregnancies.16 Pregnancy
appears to be a stimulus for AIHA with a 4 × higher
incidence than outside pregnancy. Cases of AIHA may
predate conception and relapse in pregnancy or occur
as a new presentation. Secondary causes include lymphoproliferative disorders, infections (mycoplasma,
Epstein–Barr virus) and connective tissue disorders.
AIHA is caused by the production of autoantibodies
directed against a red cell surface antigen, which on
binding results in premature destruction of the erythrocyte. This is usually extravascular in the spleen or
liver but occasionally may be intravascular. The antibodies are most frequently IgG followed by the IgM
subtype. A spectrum of severity exists. In mild cases
a positive direct antiglobulin test (DAT) is the only

Chapter 4. Maternal autoimmune cytopenias

Table 4.6 Causes of hemolysis

Immune

Autoimmune warm type

IgG mediated

• Autoimmune cold type
• Autoimmune mixed type
• Alloimmune

IgM mediated
IgG and IgM mediated
Reaction to blood transfusion,

Hereditary

• Disorder of hemoglobin synthesis
• Disorder of red cell enzymes
• Disorder of red cell membrane

e.g. sickle cell anemia
e.g. G6PD deficiency
e.g. hereditary spherocytosis

Mechanical

• Red cell fragmentation

• Mechanical heart valve
• MAHA (TTP, HUS, HELLP syndrome, pre-eclampsia)

Paroxysmal nocturnal
Hemoglobinuria

• Clonal stem cell disorder

Increased susceptibility to complement lysis

Drugs

• Oxidative stress, immune

e.g. Dapsone

Infections

• Bacterial enzymes

e.g. Clostridium perfringens

abnormality found. More severe cases have evidence of
compensated hemolysis with the most severe resulting
in significant anemia.

Diagnosis
Anemia during pregnancy is a common finding. For
patients presenting during pregnancy, the diagnosis of
AIHA requires careful exclusion of other causes of anemia, biochemical evidence of hemolysis and serological evidence that the hemolysis is immune mediated.
Important clinical and laboratory features for diagnosis are summarized below.

History

r Is the patient symptomatic of anemia?
r Is there a history of cardiovascular or pulmonary
problems which may impair ability to cope with
anemia?
r Is there evidence of a secondary cause, e.g. recent
chest infection (mycoplasma) or autoimmune
disorders?
r Is the patient on any drugs known to cause
hemolysis (especially penicillins, methyldopa,
NSAIDs)?
r Identify other potential causes of anemia
(hematinic deficiency, hereditary disorders, etc).

Examination

r Clinical examination may demonstrate evidence
of a secondary disorder.
r Cases of chronic hemolysis (e.g. hereditary
spherocytosis) can have mild splenomegaly
present.

Laboratory
Hemolysis is characterized by:
r ↑bilirubin, ↑LDH, ↑reticulocytes, ↓haptoglobins;
r blood film – polychromasia, spherocytes, red cell
agglutination (Cold AIHA);
r immune-mediated hemolysis – characterized by
positive DAT (Coombs test);
r Intravascular hemolysis – characterized by
urinary hemosiderin, hemoglobinuria.

Management
The principal risk is of a sudden fall in hemoglobin
resulting in symptomatic anemia and spontaneous
abortion. Successful management requires maintaining an adequate hemoglobin level with red cell transfusion and giving specific therapy (usually prednisolone)
to arrest the hemolysis. Although transplacental passage of antibodies occurs, the risk of developing anemia in utero or significant neonatal anemia with associated hyperbilirubinemia is small. Published experience of AIHA in pregnancy is limited, but the majority
of reports are favorable using this approach.

Blood transfusion
The presence of autoantibodies can cause difficulty in
identifying suitable units for transfusion. Autoantibodies may mask alloantibodies present in the
maternal serum, with the possibility of causing a
hemolytic transfusion reaction. Specialist investigation is required to exclude an alloantibody or identify
the specificity of an alloantibody if present. This may

Section 1. Cellular changes

delay the provision of suitable units. Many hospital
transfusion laboratories refer this work to specialist
transfusion centers, and the following points should
be considered.
r Ensure close liaison with the transfusion
laboratory to ensure that adequate samples have
been provided for testing.
r Ensure that the time within which blood is
required is clearly agreed with the transfusion
laboratory.
r In cases requiring emergency transfusion, the
risks of issuing blood without compatibility being
fully determined should be discussed between the
hematologist and obstetrician.
r Patients with cold hemaglutinin disease (CHAD)
may benefit from receiving transfusions via a
blood warmer.

increased erythropoiesis. Increased dosage may
occasionally be necessary.
r Thromboprophylaxis should be considered.
Hemolysis is a prothrombotic condition and there
is an increased risk of venous thromboembolism
(VTE). Individual assessment of the degree of risk
is necessary, and should include assessment of
other risk factors for VTE. General measures
should be emphasized, such as ensuring adequate
hydration, and re-evaluation of degree of risk
should continue through the pregnancy. The
puerperium is a peak time for thrombotic events,
and pharmacological thromboprophylaxis
during the first 6 weeks postpartum is
recommended.

Treatment of hemolysis

Considerations for fetus and at birth

Corticosteroids may be effective in reducing hemolysis. The risks of corticosteroid use are listed in Table
4.4a on ITP. Patients with warm AIHA are more likely
to respond than those with cold AIHA. A similar treatment pattern to that for ITP may be used, and as with
ITP the minimum dose possible to control hemolysis
should be used.
Experience with other agents in pregnancy is limited. IVIG can be effective and its use may be justified in pregnancy if corticosteroids are ineffective or
contraindicated. Rituximab is increasingly used outside of pregnancy but there are insufficient data currently available in pregnancy to advise its use.

There is the potential for in utero hemolysis if transplacental passage of antibodies occurs. This applies only
to cases of IgG mediated hemolysis. Unlike hemolytic
disease of the newborn, the role of monitoring maternal antibody titers has not been established. Noninvasive monitoring for anemia using ultrasonography
may be of value.
The neonatal team should be alerted prior to delivery and neonates should have a hemoglobin and bilirubin measured at birth. Neonates born to mothers
with AIHA frequently have a positive DAT; however,
hemolysis is usually mild if present. Significant anemia or elevated bilirubin levels requiring treatment
is unusual. This is in contrast to hemolytic disease of
the newborn (HDN), which may result in very severe
hemolysis requiring in utero transfusion or neonatal
exchange transfusion.

Additional measures

r Folic acid 5 mg daily should be given. This
prevents folate deficiency occurring as a result of

Chapter 4. Maternal autoimmune cytopenias

References
1. Segal JB, Powe NR. Prevalence of immune
thrombocytopenia: analyses of administrative data.
Journal of Thrombosis and Haemostasis 2006; 4:
2377–2383.
2. Sainio S, Kekomaki R, Riikonen S, Teramo K. Maternal
thrombocytopenia at term: a population-based study.
Acta Obstetrica Gynecologica Scandinavica 2000;
79:744–749.
3. Gill KK, Kelton JG. Management of idiopathic
thrombocytopenic purpura in pregnancy. Seminars in
Hematology 2000; 37: 275–289.
4. Boehlen F, Hohlfeld P, Extermann P et al. Platelet
count at term pregnancy: a reappraisal of the
threshold. Obstetrics and Gynecology 2000; 95: 29–33.
5. Verdy E, Bessous V, Dreyfus M et al. Longitudinal
analysis of platelet count and volume in normal
pregnancy. Thrombosis and Haemostasis 1997; 77:
806–807.
6. Win N, Rowley M, Pollard C et al. Severe gestational
(incidental) thrombocytopenia: to treat or not to treat.
Haematology 2005; 10: 69–72.
7. British Committee for Standards in Haematology
General Haematology task force. Guidelines for
investigation and management of idiopathic
thrombocytopenic purpura in adults, children and in
pregnancy. British Journal of Haematology 2003; 120:
574–596.
8. Webert KE, Mittal R, Sigouin C et al. A retrospective
11-year analysis of obstetric patients with idiopathic
thrombocytopenic purpura. Blood 2003; 102:
4306–4311.

9. Michel M, Novoa MV Bussel JB. Intravenous anti-D as
a treatment for immune thrombocytopenic purpura
(ITP) during pregnancy. British Journal of
Haematology 2003; 123: 142–146.
10. Provan D, Stasi R, Newland AC. International
consensus report on the investigation and
management of primary immune thrombocytopenia.
Blood 2010, 115: 168–186.
11. Veneri D, Franchini M, Raffaelli R et al. Idiopathic
thrombocytopenic purpura in pregnancy: analysis of
43 consecutive cases followed at a single Italian
institution. Annals of Hematology 2006; 85: 552–554.
12. Yamada H, Kato E, Kobashi G et al. Passive immune
thrombocytopenia in neonates of mothers with
idiopathic thrombocytopenic purpura: incidence and
risk factors. Seminars in Thrombosis Hemostasis. 1999;
25: 491–496.
13. Christiaens GC, Niewenhuis HK, Bussel JB.
Comparison of platelet counts in first and second
newborns of mothers with immune thrombocytopenic
purpura. Obstetrics and Gynecology 1997; 90: 546–552.
14. Burrows, R, Kelton J. Pregnancy in patients with
idiopathic thrombocytopenic purpura: assessing the
risks for the infant at delivery. Obstetrical and
Gynecological Survey 1993; 48: 781–788.
15. Dale DC, Cottle TE, Fier CJ et al. Severe chronic
neutropenia: treatment and follow-up of patients in
the severe chronic neutropenia international registry.
American Journal of Hematology 2003; 72: 82–93.
16. Sokol RJ, Hewitt S, Stamps BK. Erythrocyte
autoantibodies, autoimmune haemolysis and
pregnancy. Vox Sanguinis 1982; 43: 169–176.

Section

Feto-maternal alloimmune
syndromes

Section 2
Chapter

Feto-maternal alloimmune syndromes

Fetal/neonatal alloimmune
thrombocytopenia
Michael F. Murphy

Introduction
Fetal and neonatal alloimmune thrombocytopenia (FNAIT) is the commonest cause of severe
neonatal thrombocytopenia, and is analogous to
the fetal/neonatal anemia caused by hemolytic disease of the fetus and newborn (HDFN).1,2 Fetal
platelet antigens are expressed on platelets in normal
amounts from as early as the 16th week of pregnancy. Feto-maternal incompatibility for human
platelet alloantigens (HPAs) may cause maternal
alloimmunization, and fetal and neonatal thrombocytopenia may result from placental transfer of IgG
antibodies. Many HPA systems have been described3 .
The majority of HPA antigens such as HPA-1a are
located on the ␤3 subunit of the ␣IIb␤3 integrin
(GPIIb/IIIa,CD41/CD61) which is present at high
density on the platelet membrane. Others such as
HPA-5b are on ␣2␤1 (GPIa/IIa, CD49b). However,
the antigen incompatibility HPA-1a is found in
about 80% of cases of FNAIT in Caucasians and, in
contrast to HDFN, FNAIT frequently occurs in first
pregnancies.
Considerable progress has been made in the laboratory investigation of FNAIT since it was first recognized in the 1950s.1 There have also been improvements in its management, particularly in the ante-natal
management of women with a history of one or more
pregnancies affected by FNAIT, resulting from a better understanding of the risk of severe hemorrhage and
advances in fetal and transfusion medicine.

Epidemiology
The normal platelet count in the fetus and the neonate
is the same as in adults. Neonatal thrombocytopenia
has many causes, and is the commonest hematological problem in the newborn infant. A platelet count

of ⬍ 150 × 109 /L occurs in about 1% of unselected
neonates, and is ⬍ 50 × 109 /L in about 0.2%. FNAIT is
the most important cause of severe fetal and neonatal thrombocytopenia, both because of its frequency
and the severity of the bleeding associated with it.
For example, FNAIT is associated with more severe
fetal/neonatal bleeding than with maternal autoimmune thrombocytopenic purpura for reasons which
are not entirely clear but could be due to associated
platelet and/or endothelial dysfunction.
A fetal or neonatal platelet count of ⬍ 20 × 109 /L
is usually caused by FNAIT due to anti-HPA-1a as are
approximately half of the cases in which the neonatal
platelet count is ⬍ 50 × 109 /L.
The most common entities in the differential diagnosis of severe fetal and neonatal thrombocytopenia
are:
r congenital infections such as toxoplasmosis,
rubella, and cytomegalovirus;
r maternal autoimmune thrombocytopenic
purpura;
r chromosomal abnormalities;
r congenital heart disease;
r disseminated intravascular coagulation (DIC).

Incidence
Prospective studies in Caucasian populations for
FNAIT due to anti-HPA-1a indicate that about 2% of
women are HPA-1a negative, and that about 10% of
HPA-1a negative women develop anti-HPA-1a.4
Alloimmunization to HPA-1a is HLA class
II restricted. There is a strong association with
HLADRB3∗ 0101 (HLADRw52a), which is present in
1 in 3 of Caucasian women, and HPA-1a alloimmunization is rare in HPA-1a negative women who lack
this antigen.

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

Section 2. Feto-maternal alloimmune syndromes

Using data from prospective studies, the overall
incidence of FNAIT due to anti-HPA-1a is estimated to
be 1 in 1163 live births (86 per 100 000), and the incidence of severe thrombocytopenia (platelet count ⬍ 50
× 109 /L) to be 1 in 1695 (or 59 per 100 000).4
FNAIT is under-diagnosed in routine clinical practice. The evidence for this is the mismatch in the incidence of FNAIT between prospective studies involving
laboratory screening for HPA antibodies and the identification of clinically diagnosed cases. It is estimated
that only 7%–23% of cases of FNAIT, and only 37% of
severe cases, are detected clinically.

Clinical diagnosis
FNAIT is usually suspected in neonates with bleeding or severe, unexplained, and/or isolated postnatal thrombocytopenia. The clinical diagnosis is one
of exclusion.
r The infant has no signs of DIC, infection or
congenital anomalies known to be associated with
thrombocytopenia.
r The mother has had a normal pregnancy with no
history of autoimmune disease,
thrombocytopenia, or drugs that may cause
thrombocytopenia.
Specific criteria which distinguish cases of FNAIT
from other causes of unexplained thrombocytopenia
include:
r severe thrombocytopenia (platelet count ⬍ 50 ×
109 /L);
r no additional, non-hemorrhagic neonatal medical
problems;
r intracranial hemorrhage (ICH) associated with
one or more of:
Apgar score at 1 minute ⬎ 5;
birthweight ⬎2.2 kg;
documented ante-natal or post-natal bleeding.

Laboratory diagnosis
Detailed laboratory investigations are required for
confirmation of a provisional clinical diagnosis, and
should be performed by an experienced reference laboratory. The diagnosis is based on:
r detection and identification of the maternal HPA

antibody;
r determination of the HPA genotype of mother,
father and, if needed, the child (or fetus).

In the past, it was difficult to differentiate between
HLA and HPA antibodies in standard serological
assays. The description of the monoclonal antibodyspecific immobilization of platelet antigens (MAIPA)
assay overcame this problem. Rather than working
with intact platelets, the assay involves capture of specific GPs using monoclonal antibodies enabling analysis of complex mixtures of platelet antibodies. However, it requires considerable operator expertise in
order to ensure maximum sensivity and specificity, and
the selection of appropriate screening cells is critical.
Immunization against HPA-1a and HPA-5b are
responsible for up to 95% of cases of FNAIT. Antibodies against other HPAs are more frequently detected
in recent large series of FNAIT. In some of these
cases, testing against standard donor platelet panels may be negative. To pursue further investigation
requires strong clinical suspicion of FNAIT. Possible
approaches include:
r cross-match of maternal serum and paternal
platelets using MAIPA;
r identification of a mismatch between maternal
and paternal (or neonatal) genotypes for low
frequency HPA antigens, and then screen
maternal serum for the corresponding HPA
antibodies.

Clinical significance of FNAIT
ICH is the major cause of mortality and long-term
morbidity in FNAIT. The long-term outcome may be
devastating with blindness and major physical and
mental disability (Fig. 5.1). ICH was reported in a large
review of the literature to occur in 74/281 (26%) of
cases of FNAIT due to anti-HPA-1a with a mortality
of 7%.5
Although there is a risk of hemorrhage due to
severe thrombocytopenia at the time of delivery,
80% of ICH associated with FNAIT occur in utero,
with 14% occurring before 20 weeks and a further 28% occurring before 30 weeks.5 There may
also be unusual presentations such as isolated fetal
hydrocephalus, unexplained fetal anemia, or recurrent
miscarriages.
Bleeding is more severe with FNAIT due to antiHPA-1a than for example anti-HPA-5b, possibly due to
the higher density of HPA-1a antigen sites on platelets.

Chapter 5. Fetal/neonatal alloimmune thrombocytopenia

with a history of FNAIT with ICH was 72% (confidence interval 46%–98%) without the inclusion of
fetal deaths, and 79% (confidence interval 61%–97%)
with their inclusion.6 The risk of ICH following a
previous history of FNAIT without ICH was estimated
to be 7% (confidence interval 0.5%–13%).
These data provide the justification for ante-natal
intervention in women with a past history of pregnancies affected with FNAIT, particularly where there has
been fetal or neonatal ICH in a previous pregnancy, to
reduce the risk of morbidity and mortality from severe
hemorrhage.
If there is paternal heterozygosity for the relevant HPA, fetal platelet genotyping should be considered, for example, by obtaining a sample using
amniocentesis.

Fig. 5.1 Intracranial hemorrhage in FNAIT: MRI scan showing
subacute hematoma (black arrow) and chronic hematoma (open
arrow). Reproduced from De Vries et al. Br J Obstet Gynaecol; 95:
299–302.

Prediction of the severity of FNAIT in
subsequent pregnancies
Laboratory testing
Unfortunately, there is no reliable laboratory method
to predict severe clinical disease, which might be used
to identify pregnancies at risk of severe thrombocytopenia and ICH. Some studies have observed an association between high levels of maternal anti-HPA-1a
and the severity of neonatal thrombocytopenia, but
this is not a sufficiently reliable association to be clinically useful. Reliable methods for quantifying the other
antibodies are not yet available. The lack of laboratory
parameters predictive of severe disease remains one of
the major barriers to optimizing ante-natal management for FNAIT, and is an important area for future
research.

Consideration of ante-natal
screening for FNAIT
Advances in the laboratory diagnosis and antenatal management of FNAIT have drawn attention to
the fact that the first affected fetus/neonate is usually only recognized after bleeding has occurred or
severe thrombocytopenia detected by chance. This
raises the question of whether routine screening for
FNAIT should be considered. It is recognized that
there are significant shortcomings in the knowledge
about FNAIT necessary for the introduction of an
antenatal screening program.7
More research is required, for example, on the clinical outcome of first affected pregnancies, the identification of laboratory measures predictive of severe disease where ante-natal intervention might be justified,
and the optimal approach for the ante-natal management of pregnant women with HPA antibodies, but
with no previous history of affected pregnancies, as
ante-natal treatment carries significant risks and costs.

Management of FNAIT

History of FNAIT in previous pregnancies

Post-natal

Subsequent pregnancies of HPA-1a alloimunized
women with a history of a previously affected infant
with FNAIT are well recognized to be associated with
a high risk of recurrence of FNAIT and poor outcome.
A detailed literature search found that the recurrence
rate of ICH in the subsequent pregnancies of women

The thrombocytopenia in FNAIT usually resolves
within 2 weeks, although it may last as long as 6 weeks.
A cerebral ultrasound should be carried out to determine if ICH has occurred because of the changes in
management that would occur if there had been a
hemorrhage.

Section 2. Feto-maternal alloimmune syndromes

The optimal post-natal management of FNAIT
depends on its rapid recognition, and prompt correction by transfusion of platelet concentrates to neonates
who are severely thrombocytopenic (platelet count
⬍30 × 109 /L) or bleeding. It is not appropriate to wait
for the laboratory confirmation of the diagnosis in suspected cases.
While there has been debate about the value of
random donor platelets in the immediate post-natal
management of FNAIT, two recent studies reported
that random donor (i.e. not HPA-matched) platelets
were often effective in increasing the platelet count
in FNAIT. However, in some of the cases, spontaneous recovery of the neonatal platelet count may have
been the reason for the apparent response to random
donor platelet transfusions. Compatible platelet concentrates were shown in another study to produce a
larger increase in platelet count and twice the length
of survival of the transfused platelets compared to random donor platelets.8
Compatible platelet concentrates, for example,
from HPA-1a and 5b negative donors, should be used
initially, if they are available, on the basis of the certainty of their effectiveness in the more than 90% of
cases of FNAIT which are due to anti-HPA-1a or antiHPA-5b. Unfortunately, the routine availability of such
HPA-1a and 5b-negative platelets for immediate use in
suspected cases of FNAIT is limited to only a minority
of countries, including England.
Although intravenous immunoglobulin (IVIG) is
effective in at least 75% of cases, the platelet count does
not increase in responders for 24–72 hours so it should
not be used for the initial therapy of FNAIT. Its role in
the management of post-natal FNAIT should be limited to those few cases with very prolonged and severe
thrombocytopenia.

Provision of information to the mother
The parents should be provided with information
about FNAIT once the platelet antigen typing and antibody results are complete, specifically to provide:

1. an explanation of the cause of FNAIT;
2. the risk of recurrence in subsequent pregnancies;
3. the options for ante-natal management as well as
the fact that this is an evolving field;
4. a request that the mother should notify the fetal
medicine center as soon as she becomes pregnant;
5. her risk for the future of transfusion reactions,
and potentially post-transfusion purpura (PTP),

although it appears that the risk of PTP is very low
with leukocyte-reduced blood components which
are now standard in the UK;
6. Testing of female relatives of the mother should be
suggested.

Ante-natal
The traditional management of subsequent pregnancies in women with a previous history of FNAIT
consisted of performing early elective Cesarean section, and then transfusing compatible platelets after
birth. Major advances in the ante-natal management
of FNAIT have been made in the last 25 years.1,2,9

Early ante-natal treatment strategies
In 1984, the use of ultrasound-guided fetal blood sampling (FBS) was described to obtain the fetal platelet
count at 32 weeks’ gestation in the second pregnancy of
a woman whose first child had ICH due to FNAIT; the
fetal platelet count was 15 × 109 /L. There was no ultrasound evidence of ICH by 37 weeks, and an in utero
transfusion of maternal platelets was given 6 hours
prior to delivery by Cesarean section. As a result, the
cord platelet count was 95 × 109 /L and there were no
signs of bleeding.
The use of in utero platelet transfusion (see Fig. 5.2)
immediately before delivery was described in greater
detail in a series of 9 cases, where FBS was carried
out at 21 weeks’ gestation to confirm the diagnosis
of FNAIT.1 FBS was repeated at 37 weeks with an in
utero platelet transfusion if the fetal platelet count was
⬍ 50 × 109 /L followed by delivery 6–36 hours later.
However, over the next 10 years, it became clearer that
an affected fetus is at risk of ICH in utero, even before
20 weeks’ gestation, indicating that earlier ante-natal
intervention is required in cases likely to be severely
affected. During this period, different groups began
to explore alternative approaches to ante-natal management, one based around serial weekly fetal platelet
transfusion, and the other around medical treatment
of the mother with IVIG and/or steroids.

Serial fetal platelet transfusions
Early studies with fetal platelet transfusions highlighted the short survival of transfused platelets, and
the difficulty of maintaining the fetal platelet count
at a “safe” level. Further experience indicated that it
was possible to maintain the count above 30 × 109 /L
using transfusions at weekly intervals (Fig. 5.3). This

Chapter 5. Fetal/neonatal alloimmune thrombocytopenia

Technical aspects and complications of FBS

Donor
platelet
transfusion

The technique employed for trans-abdominal
ultrasound-guided FBS and intravascular transfusion
is the same as for red cell alloimmunization. Unlike
HDFN, where the needle may be removed while
the hematocrit is estimated before transfusion is
commenced, removal of the needle from the umbilical
cord in the presence of a very low platelet count can
result in rapid exsanguination of the fetus. Very few
operators check the platelet count during the procedure and it is standard practice to transfuse platelets
to the fetus following FBS even if the procedure is
undertaken for diagnosis or monitoring of FNAIT
rather than part of serial fetal transfusions.
The main risks of FBS are severe cord bleeding,
cardiac arrhythmias, and miscarriage. Pooling data
from several studies indicates a fetal loss rate of 3/223
(1.3%)/procedure and 3/55 (5.5%)/pregnancy.
From 26 weeks’ gestation, FBS and platelet transfusion should be performed in the operating theater where facilities are available to perform an emergency Cesarean section, should there be signs of fetal
distress or bleeding from the sampling site. Unpublished data from the Oxford Rhesus Therapy Unit
indicate that there is approximately a 4% chance of
rapid delivery being required at the time of each
transfusion.
The volume of platelet hyperconcentrate to be
transfused is calculated from a formula:
Volume of concentrate = desired platelet increment × feto-placental blood volume for gestational
age × R ÷ platelet count of the concentrate

3-way tap

Fig. 5.2 Schematic diagram of ultrasound-guided fetal blood
sampling and platelet transfusion.

was achieved by increasing the dose of platelets, whilst
avoiding an unacceptable increase in the transfused
volume, by concentrating the platelet collection by
centrifugation and removal of plasma. Later improvements in apheresis technology allowed the preparation of leukocyte-depleted concentrated platelets suitable for fetal transfusion without the need for further
processing.
10 000

1 000

Platelets × 109/L

Fig. 5.3 Pre- and post-transfusion
platelet counts following serial FBS and
platelet transfusions. The fetal platelet
count was ⬍10 × 109 /L at 26 weeks. The
aim was to maintain the fetal platelet
count above 30 × 109 /L by raising the
immediate post-transfusion platelet count
to above 300 × 109 /L after each
transfusion. The fetal platelet count fell
below 10 × 109 /L on one occasion when
there were problems in preparing the fetal
platelet concentrate and the dose of
platelets was inadequate. CS = Cesarean
section. Reproduced from Practical
Transfusion Medicine, 3rd edn. Murphy MF
& Pamphilon D. Wiley-Blackwell
Publishing, 2009.

300
100
Platelet
transfusion

1
25

Weeks’ gestation

Days’ post-natal
CS

Section 2. Feto-maternal alloimmune syndromes

Table 5.1 Specification of the platelet product for
intra-uterine transfusion

Donor
r HPA type compatible with maternal
antibodies, usually HPA-1a negative
r Group O RhD negative for the first transfusion
(for subsequent transfusions, the ABO and RhD
group of the donor should be compatible with
the fetal blood group which should be
determined from a sample taken at the first
FBS)
r No HPA or HLA antibodies
r No high titer ABO antibodies
Platelet concentrates
r High concentration of platelets (usually in the
range 2.5 ×109 –3.0 × 1012 /L compared to
1.4 × 1012 /L for standard platelet concentrates
for use in neonates or adults) to reduce the
volume of the transfusion. The
hyperconcentrates can be prepared using a
modification of the procedure for collection of
platelet concentrates by apheresis.
r Gamma-irradiated to prevent
transfusion-associated graft-vs.-host disease
r CMV-seronegative
r Leukocyte-reduced
r Transfuse within 24 hours after collection
The feto-placental volume for gestational age is calculated from standard charts. In early fetal platelet
transfusion studies, the immediate post-transfusion
platelet increment was found to be 50% of that
expected, i.e. 50% platelet recovery, probably because
of pooling in the feto-placental circulation. The volume calculation takes account of this by introducing
the factor R = 2, thus doubling the volume of platelets
transfused.
The specification of the platelet product for intrauterine transfusion is provided in Table 5.1.

Maternal treatment
One of the main drivers for the development of maternally directed ante-natal treatment for FNAIT was
concern about the risks of FBS and platelet transfusion.

Steroids
There is considerable experience from North America with the combined use of steroids and IVIG.10

Although low dose steroids did not add significantly
to the effect of IVIG, high dose steroids (prednisolone
60 mg and later 1 mg/kg) added substantially to the
effect of IVIG. The use of 0.5 mg/kg prednisolone in
the lowest risk cases (no previous sibling ICH, initial
fetal count ⬎ 20 × 109 /L) demonstrated efficacy comparable to that of IVIG in this group of patients.
Intravenous immunogloblin (IVIG)
The first protocol involving maternal administration
of IVIG was described in 1988. Initial FBS was carried out at 20–22 weeks’ gestation to confirm the diagnosis of FNAIT and its severity. IVIG (dose 1 g/kg
body weight/week) was administered to the mother,
and FBS was repeated 4–6 weeks later to assess the
effect of IVIG. None had ICH in contrast to three of
their respective untreated siblings, two of whom had
antenatal ICH, and there were no serious complications of treatment.Overall, there was an increase of
36 × 109 /L between the first and second FBS, and an
increase of 69 × 109 /L between the first FBS and birth.
Of fetuses 62%–85% responded to therapy depending on the definition of response used, and there were
no cases with ICH. However, other reports described
cases in which IVIG was ineffective in raising the fetal
platelet count, and ante-natal ICH was reported during
maternal treatment with IVIG.
Complications of maternal teatment
The use of IVIG is expensive, and both IVIG and
prednisolone can cause adverse maternal effects. IVIG
appears to be a safe blood product when administered to otherwise healthy young women. The risks
of renal disease, hemolysis, fluid overload, and transmission of infection are extremely low, and none of
these have been reported in a patient undergoing antenatal treatment for FNAIT. Headaches occur but usually lessen with time. Prednisolone has been widely
used in pregnancy, and is known to cause fluid overload, high blood pressure, diabetes mellitus, irritability, and osteoporosis.
Recent studies of maternal treatment
A collaborative study in European centers reported
in 2003 on the ante-natal management of FNAIT
in 56 fetuses managed with either maternal treatment or platelet transfusions. Maternal therapy,
predominantly IVIG, resulted in a platelet count
exceeding 50 × 109 /L in 67%. The most serious
complications encountered were associated with FBS

Chapter 5. Fetal/neonatal alloimmune thrombocytopenia

and platelet transfusion, and the results support the
use of maternal therapy as first-line treatment for the
ante-natal management of FNAIT. The association of
lower pre-treatment platelet counts in cases with a
sibling history of ante-natal ICH or severe thrombocytopenia favors stratification of ante-natal management
on the basis of the history of FNAIT in previous
pregnancies.
In 2006, the North American team reported
two randomized controlled trials of maternal treatment stratified according to the previous history of
FNAIT.11
(1) “High risk” patients had either a sibling with
peripartum ICH or one with an initial fetal
platelet count ⬍ 20 × 109 /L. Patients underwent
FBS at 20 weeks or later, and were randomized to
receive IVIG alone (1 g/kg/week) or in
combination with prednisolone 1 mg/kg/day.
There was a satisfactory increase in the fetal
platelet count in 89% of pregnancies receiving
combination treatment compared to 35%
receiving IVIG alone (P = ⬍ 0.05). In those
with initial fetal platelet counts ⬍ 10 × 109 /L,
82% had a satisfactory response to IVIG and
prednisolone compared to only 18% treated with
IVIG alone (P = ⬍ 0.03). There was one ICH; this
occurred in a pregnancy managed with IVIG
alone.
(2) “Standard” risk patients were those with a sibling
who had not had an ICH and a fetal platelet count
between 20 and 100 × 109 /L. These patients
underwent FBS near to 20 weeks, and were
randomized to receive IVIG (1 g/kg/week) or
prednisolone 0.5 mg/kg/day. Subsequent FBS was
carried out in all patients at 3–8 weekly intervals.
There were no significant differences in the
responses to the two treatments. There were two
ICHs; one in a fetus born at 38 weeks’ gestation
with a platelet count of 172 × 109 /L, and one in an
infant with a birth platelet count of 68 × 109 /L
delivered at 28 weeks because of bradycardia
following FBS.
There were 11 serious complications out of a total
of 175 (6%) FBS confirming the dangers of FBS and
platelet transfusion in FNAIT. This study demonstrates
that effective ante-natal treatment can be stratified
according to the previous history of FNAIT.

The search for less invasive strategies for the ante-natal
management of FNAIT
Concern regarding the safety of FBS and platelet transfusion has led to a search to develop less invasive
treatment strategies involving maternal administration of IVIG while reducing or even avoiding FBS for
monitoring the fetal platelet count and administering
platelet transfusions.
Some studies suggested that the pre-treatment
platelet count had predictive value for the response
to maternal treatment. A review of patients treated in
North America found that the response rate in fetuses
with a pre-treatment platelet count of ⬎ 20 × 109 /L
was 89%, but was only 51% in those with an initial fetal
platelet count ⬍ 20 × 109 /L. The authors suggested
that additional FBS might not be warranted in those
cases with an initial fetal platelet count ⬎ 20 × 109 /L;
any gain from identifying and intensifying treatment
in “poor responders” would be offset by the complications of additional FBS.
The Leiden group have evaluated less intensive
ante-natal treatment strategies over a number of years
and found that that a non-invasive strategy based on
treatment with IVIG without FBS appears to be effective when there is no history of ICH in a previous pregnancy.12 The same group extended this approach to
the management of seven high risk pregnancies where
there had been a previous sibling history of ICH. IVIG
was administered from 16–19 weeks’ gestation in the
six pregnancies where there had been previous antenatal ICH, and from 28–29 weeks in the case where
ICH was post-natal. The total number of weekly IVIG
infusions ranged from 8 to 21. The platelet count at
birth ranged from 10 × 109 49 × 109 /L. No ICH was
seen on ante-natal or post-natal ultrasound examinations, and all infants were doing well at follow-up at
3 months.
These recent studies indicating success with less
invasive strategies suggest that further work is necessary to determine the optimal ante-natal management for FNAIT. An alternative to the “empirical” (no
FBS) and “invasive” (FBS before and during treatment) approaches is to initiate maternal treatment
(type and timing determined by consideration of the
previous history of FNAIT) without performing FBS,
and then to carry out FBS 4–8 weeks after the initiation of treatment to identify the non-responding cases
which may benefit from a change in treatment. This
is the approach being followed by some UK referral

Section 2. Feto-maternal alloimmune syndromes

Table 5.2 Suggested ante-natal management depending on
previous history of FNAIT ∗

(1) Ante-natal ICH in previous sibling:
ICH in second trimester
r At 12 weeks, IVIG 2 g/kg/week (given as 1
g/kg/twice a week)
r FBS at week 20–22
ICH in third trimester
r At 16 weeks, IVIG 1 g/kg of IVIG
r FBS at week 20–22
If fetal platelet count at first FBS ⬎ 30 × 109 /l:
r Continue current treatment
r Further FBS at 28 weeks at 34–36 weeks and/or
pre-delivery
If fetal platelet count at first FBS ⬍ 30 × 109 /l:
r Add prednisolone 1 mg/kg/day
r Repeat FBS 2 weeks later. If no response, where
relevant increase IVIG to 2 g/kg/week (given as
1 g/kg/twice a week) and repeat FBS at 2 weeks
r If no response to maximal combination
therapy, proceed to weekly IUT and discontinue
medical treatment
r If response to maximal combination therapy
repeat FBS at 2–4 weekly intervals
(2) Neonatal ICH or platelet count ≤ 50 × 109 /l
in previous sibling:
r IVIG 1 g/kg/week at 20 weeks
r FBS at 28–32 weeks
If fetal platelet count ⬎ 30 × 109 /l:
r Continue current treatment
r Further FBS at 34–36 weeks
If fetal platelet count at first FBS ⬍ 30 × 109 /l:
r Add prednisolone 1 mg/kg/day
r Repeat FBS 2 weeks later. If no response, where
relevant increase IVIG to 2 g/kg/week (given as
1 g/kg/twice a week) and repeat FBS at 2 weeks
r If no response to maximal combination
therapy, proceed to weekly IUT and discontinue
medical treatment
r If response to maximal combination therapy,
repeat FBS at 2–4 weekly intervals

Table 5.2 (cont.)

Mode of delivery:
Based on FBS at 30–32 weeks:
r If fetal platelet count ≥ 100 × 109 /L, proceed to
spontaneous vaginal delivery with no further
fetal blood sampling
r If fetal platelet count ≤ 100 × 109 /L, continue
with treatment and perform repeat sampling at
35–37 weeks, with transfusion of platelets
r If fetal platelet count ≥ 50 at 35–37 weeks
(prior to platelet transfusion), allow
spontaneous vaginal delivery
r If platelet count ⬍ 50 × 109 /L at 35–37 weeks,
discuss options:
– Induction of labor within 5 days of IUT
– Weekly IUT until either spontaneous labor,
induction of labour or planned Cesarean
section
There is no evidence to suggest that elective Cesarean section
is safer than vaginal delivery, if the platelet count is above 50 ×
109 /L.
∗ developed by Rachel Rayment, Mike Murphy and Jim Bussel
(unpublished data).

centers including our own (Table 5.2). Recommendations about the mode of delivery are also provided in
Table 5.2.

How to manage the ‘non-responders’ to initial
maternal therapy
The options are to increase the dose of IVIG, add prednisolone, switch to serial platelet transfusions and/or
consider early delivery. The North American group
have developed this concept of “salvage” or “intensification” therapy. Only about 25% of “high risk” or
“standard risk” patients required more intensive treatment because of a lack of response to their initial
therapy. “Intensification” therapy comprised adding
IVIG or prednisolone if not being used already, or
increasing the dose of IVIG, and all but six had platelet
counts at birth ⬎ 50 × 109 /L.
The ability to modify ante-natal treatment in an
individual case does depend on the use of FBS to monitor the fetal platelet count. Although empirical treatment without knowledge of the fetal platelet count
before or during treatment avoids the risks of FBS, it
has the drawbacks of the administration of potentially
unnecessary or inadequate treatment.

Chapter 5. Fetal/neonatal alloimmune thrombocytopenia

Optimal approach for the modern ante-natal
management of FNAIT
There has been huge progress in the ante-natal management of FNAIT over the last 20 years. However, the
ideal effective treatment without significant side effects
to the mother or fetus has yet to be determined.
There are some basic principles to consider in the
management of an individual case.2
1. Obtain as much information as possible about the
clinical history of previously affected pregnancies
with FNAIT focusing on the neonatal
thrombocytopenia to exclude other causes of
thrombocytopenia. It is important to determine as
conclusively as possible if an ICH has occurred
and if so, when.
2. Ensure that comprehensive laboratory
investigations have been carried out in a reference
laboratory, including testing for HPA antibodies
and the identification of their specificity, and HPA
genotyping of the mother and her partner. If the
partner is heterozygous for the relevant HPA, the
fetal HPA genotype should be established.
3. Affected fetuses should be managed in referral
centers with experience in the ante-natal
management of FNAIT. Close collaboration is
required between specialists in fetal medicine,
obstetrics, hematology/transfusion medicine, and
pediatrics.
4. The mother and her partner should be provided
with detailed information about FNAIT and its
potential clinical consequences, and the benefits
and risks of different approaches to ante-natal
management.
5. Maternally administered therapy should be the
first-line approach in all cases. This is based on
data describing the effectiveness and safety of
maternal treatment in contrast to the toxicity of
serial FBS to deliver weekly fetal platelet
transfusions.
6. An important goal is to minimize the number of
FBS. However, the debate between empirical
treatment and treatment guided by measurement
of the fetal platelet count using FBS is not yet
resolved. Either approach is acceptable until the

issue is resolved by further clinical trials. It is to be
hoped that there will be developments in
laboratory testing allowing non-invasive
assessment of the likely severity of FNAIT in
individual cases.
7. Different centers currently have different
strategies based on their own experience and
those of published studies. Stratification of
ante-natal treatment based on the history of
FNAIT in previous pregnancies is common (and
appropriate) to both empirical and “invasive”
approaches to treatment.
8. Further progress is only likely to be achieved by
conducting randomized controlled trials to
resolve outstanding management issues. Patients
should be entered into trials, wherever possible.
Even referral centers see relatively small numbers
of patients, and to obtain sufficient patient
numbers for adequately powered trials,
collaboration will be required between referral
centers.

Summary
There have been considerable advances in the clinical and laboratory diagnosis of FNAIT, and its postnatal and ante-natal management. The ante-natal management of FNAIT has been particularly problematic,
because severe hemorrhage occurs as early as 16 weeks’
gestation and there is no non-invasive investigation
which reliably predicts the severity of FNAIT in utero.
The strategies for ante-natal treatment have included
the use of serial platelet transfusions, which while
effective are invasive and associated with significant
morbidity and mortality. Maternal therapy involving
the administration of intravenous immunoglobulin
and/or steroids is also effective and associated with
fewer risks to the fetus. Significant recent progress has
involved refinement of maternal treatment, stratifying it according to the likely severity of FNAIT based
on the history in previous pregnancies. However, the
ideal ante-natal treatment, which is effective without
causing significant side-effects to the mother or fetus,
has yet to be determined, and further clinical trials are
needed.

Section 2. Feto-maternal alloimmune syndromes

References
1. Kaplan C. Neonatal alloimmune thrombocytopenia: a
50 year story. Immunohematology 2007; 23: 9–13.

7. Murphy MF, Williamson LM, Urbaniak SJ. Antenatal
screening for fetomaternal alloimmune
thrombocytopenia: should we be doing it? Vox
Sanguinis 2002; 83: 409–16.

2. Murphy MF, Bussel JB. Advances in the management
of alloimmune thrombocytopenia. British Journal of
Haematology 2007; 136: 366–378.

8. Allen D, Verjee S, Rees S et al. Platelet transfusion in
neonatal alloimmune thrombocytopenia. Blood 2007;
109: 388–389.

3. Ouwehand WH, Stafford P, Ghevaert C et al. Platelet
immunology, present and future. ISBT Science Series
2006; 1: 96–102.

9. Rayment R, Brunskill SJ, Stanworth S et al. Antenatal
interventions for fetomaternal alloimmune
thrombocytopenia. The Cochrane Library, Issue 1,
2005. Chichester, UK: John Wiley & Sons, Ltd.

4. Turner ML, Bessos H, Fagge T et al. Prospective
epidemiologic study of the outcome and
cost-effectiveness of antenatal screening to detect
neonatal alloimmune thrombocytopenia due to
anti-HPA-1a. Transfusion 2005; 45: 1945–
1956.
5. Spencer JA, Burrows RF. Feto-maternal alloimmune
thrombocytopenia: a literature review and statistical
analysis. Australia and New Zealand Journal of
Obstetrics and Gynaecology 2001; 41: 45–55.
6. Radder CM, Brand A, Kanhai HH. Will it ever be
possible to balance the risk of intracranial
haemorrhage in fetal or neonatal alloimmune
thrombocytopenia against the risk of treatment
strategies to prevent it? Vox Sanguinis 2003; 84:
318–325.

10. Bussel JB, Berkowitz RL, Lynch L et al. Antenatal
management of alloimmune thrombocytopenia with
intravenous gammaglobulin: a randomized trial of the
addition of low dose steroid to IVIg in fifty-five
maternal–fetal pairs. American Journal of Obstetrics
and Gynecology 1996; 174: 1414–1423.
11. Berkowitz RL, Kolb EA, McFarland JG et al. Parallel
randomized trials of risk-based therapy for fetal
alloimmune thrombocytopenia. Obstetrics and
Gynecology 2006; 107: 91–96.
12. Radder CM, Brand A, Kanhai HHH. A less invasive
treatment strategy to prevent intracranial hemorrhage
in fetal and neonatal alloimmune thrombocytopenia.
American Journal of Obstetrics and Gynecology 2001;
185: 683–688.

Section 2
Chapter

Feto-maternal alloimmune syndromes

Red cell alloimmunization
Alec McEwan

Introduction
Hemolytic disease of the newborn (HDN) describes a
process of rapid red blood cell breakdown, which puts
the baby at risk of anemia and kernicterus (bilirubin
induced cerebral damage) within the first few days of
life. A variety of etiologies are recognized; however,
this chapter focuses on red cell alloimmunization, i.e.
the immune-mediated destruction of erythrocytes initiated by maternal red cell antibodies which reach the
fetal circulation by transportation across the placenta,
onwards from approximately 12 weeks’ gestation.

Pathogenesis
Antibodies recognizing red cell surface antigens usually arise secondary to a blood transfusion, or following the birth of a baby with a different blood
group to the mother. Fetal red blood cells “traffick”
into the maternal circulation throughout pregnancy,
but “isoimmunization” against foreign antigens occurs
most frequently around the time of delivery when
the size of feto-maternal hemorrhage (FMH) tends
to be greatest. Other events associated with FMH are
listed in Table 6.1. These red cell antibodies can, in a
subsequent pregnancy, reach the fetal circulation and
cause immune mediated destruction of fetal red blood
cells. This transplacental transportation of maternal immunoglobulin G begins in the early second
trimester and red cell antibodies recognizing certain
erythrocyte antigens may bind and bring about premature destruction of the fetal red cells by the reticuloendothelial system. One of the breakdown products
of heme is bilirubin, and levels rise within the fetus
and amniotic fluid, although placental transfer limits
this accumulation. Progressive anemia initially stimulates the bone marrow first but, as its capacity to maintain the hemoglobin levels is exceeded, extramedullary

hematopoiesis becomes increasingly important. This
hyperactivity of the reticuloendothelial system results
in fetal hepatosplenomegaly. A degree of portal hypertension and hypoalbuminaemia secondary to liver
dysfunction may contribute to extracellular fluid accumulation within the fetus (hydrops fetalis); however,
cardiac dysfunction is more likely to be the main
explanation for hydropic change. Fetal anemia induces
a high-output cardiac state and a degree of hypoxia
may directly impair myocardial contractility. Hydrops
is characterized by skin edema, pleural and pericardial
effusions, cardiomegaly, atrioventicular valve dysfunction, ascites, polyhydramnios, and placentomegaly, all
of which can be detected by ultrasound scanning
(Fig. 6.1–6.3). These changes are seen only when fetal
hemoglobin levels decline well below the normal range
and are a late feature of erythroblastosis fetalis. Intrauterine death will ensue in severe cases if the problem
is not treated, or the baby delivered.
HDN describes the consequences of this antenatal pathogenic process which continues on into the
newborn period. Maternal immunoglobulin G (IgG)
remains with the baby for 4–6 months after birth
and top-up blood transfusions may be needed by the
infant whilst hemolysis continues. Far more concerning than this semi-chronic post-natal anemia, however, is the risk of kernicterus which occurs within
the first few days of life. The immature fetal liver is
unable to conjugate the excessive circulating bilirubin and, as serum levels rise, it permeates the blood–
brain barrier. The globus pallidus of the basal ganglia
and the brain stem nuclei are the structures most at
risk of damage from the unconjugated bilirubin, which
is thought to uncouple phosphorylation from oxidation, resulting in reduced ATP synthesis and impairment of energy-dependent metabolism. Athetoid cerebral palsy, other movement disorders, deafness and

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

Section 2. Feto-maternal alloimmune Syndromes

Table 6.1 Clinical scenarios associated with FMH and risk of
isoimmunization (adapted from RCOG Green top guideline
(No. 22))
Any birth (including by cesarean section)
Manual removal of retained placenta
Stillbirths and intrauterine deaths
Abdominal trauma in the third trimester
Delivery of twins
Unexplained hydrops fetalis
Invasive pre-natal diagnostic procedures such as amniocentesis
or CVS
Antepartum hemorrhage
External cephalic version
Hydatidiform mole
Termination of pregnancy (prophylaxis is recommended at all
gestations and with all methods)
Ectopic pregnancy (regardless of mode of treatment)
Spontaneous miscarriage ≥ 12 weeks (see below)

Fig. 6.1 Ante-natal ultrasound showing a transverse section
through the upper fetal abdomen at the level of the stomach and
liver. The calipers are measuring a 10 mm rim of ascites. There are
numerous etiologies for fetal ascites, but fetal anemia (from any
cause) is one of the more common explanations.

impaired eye movements may all be long-term sequelae of kernicterus.
Repeated exposure of an isoimmunized woman
to the same red cell antigen, as occurs in successive
pregnancies, will further stimulate antibody production. Subsequent pregnancies, which express the blood
group in question, have a tendency to show more
severe hemolysis, and at earlier gestations.

Fig. 6.2 Ante-natal ultrasound showing a transverse section
through the fetal cranium. The calipers are measuring 9 mm of scalp
edema. Edema can collect throughout the skin of the fetus in severe
anemia. This results from a combination of high output cardiac
failure and also possible hepatic dysfunction and hypoproteinemia.

Fig. 6.3 Ante-natal ultrasound showing a transverse section
through the fetal chest. A slender fetal pericardial effusion and a
small left sided pleural effusion behind the heart can be seen. The
heart is also subjectively enlarged. These features are all consistent
with, but are non-specific signs of, fetal anemia.

Genotype and phenotype
There are almost 30 different blood grouping systems,
but the ABO and Rhesus groups are arguably the most
important clinically. The Rhesus D (RhD) antigen was
discovered in 1939, but the full complexity of this

Chapter 6. Red cell alloimmunization

blood group system has only become evident much
more recently with the advent of molecular biology.
Of white Europeans, 16% are RhD negative, 5% of
West Africans, and virtually no Chinese. Of all deliveries in the UK, 10% are of RhD positive babies born
to RhD negative women. In the absence of preventive
measures, 1 in 6 RhD negative women will isoimmunize if they deliver a term RhD positive baby, and in the
1950s 1 in 2000 babies died of HDN, principally due to
RhD isoimmunization.
The Rhesus proteins are coded for by two genes,
which share a major degree of homology. RHD and
RHCE lie very close to one another on chromosome
1, back-to-back, and are thought to have arisen from
a duplication event involving the original ancestral
Rhesus gene, which can still be found in rodents and
most other mammals. The Rhesus proteins are characterized by 12 intramembranous segments and 6
extra cellular “surface” loops. Their function remains
unclear, although ammonium ion transportation and
gas exchange across the erythrocyte cell membrane
have been postulated.
The RhD negative phenotype is recognized in the
laboratory by failure of red cells to agglutinate with
standard anti-D reagents (antibodies). The underlying
genetic explanation for this phenotype is more complex. In Europeans, 90% of RhD negative individuals
have a complete deletion of RHD, with the remaining cases being explained by nonsense and frameshift
mutations which truncate the protein. However, in the
majority of African individuals typed as RhD negative the genotype is very different. The two common
RHD variants resulting in the D negative phenotype
are the RHD pseudogene, RHD␺ , which codes for a
non-functional protein, and the Cdes allele which contains segments from both the RHD and the RHCE
genes.
The situation is confused even further by alleles
of RHD, which cause subtle qualitative changes in
the extracellular surface loops of the RhD protein,
meaning that serological tests are only weakly positive
with standard anti-D reagents. Furthermore, missense
mutations causing single amino acid substitutions in
the intramembranous or cytoplasmic portions of the
RhD protein may impair integration of the protein
into the membrane, so bringing about a quantitative
reduction in the number of cell surface antigen sites
per red blood cell. This too may reduce the agglutination response of these cells to standard laboratory antiD antibodies. These “partial D” and “weak D” pheno-

Table 6.2 Key events in the history of prevention of RhD
isoimmunization
1938 Darrow concludes that “erythroblastosis fetalis” results
from the formation of a maternal antibody against some
component of fetal blood
1939 Levine and Stetson postulate that maternal
immunization is caused by a fetal antigen inherited from
the father which is lacking in the mother
1940 Landsteiner and Wiener discover the Rhesus antigen
1948 Wiener suggests that the initiating process is occult
placental hemorrhage
1957 Kleihauer devises a test able to detect fetal cells in the
maternal circulation
1961 Stern gives RhD positive red blood cells to RhD negative
volunteers, both with and without anti-D, and shows
that alloimmunization can be prevented
1966 Freda demonstrates that isoimmunization can be
prevented by giving anti-D to recently delivered RhD
negative women
1969 Widespread introduction of routine post-natal
prophylaxis with anti-D following multicenter trials

types, as they are respectively known, can be important from a clinical perspective and will be discussed in
greater detail later.
The DNA sequence of the RHCE gene shows far less
variation, and differences at just five amino acid positions result in the four different antigens C, c, E, and e.
Each allele expresses only C or c, in combination with
E or e, and, amongst Europeans, the Ce haplotype is
most common.

Prevention of RhD isoimmunization
Antibodies against all the Rhesus proteins, and other
red cell antigens, can cause erythroblastosis and HDN;
however, anti-D has historically been of greatest significance. Prevention of RhD isoimmunization, and
improvements in the ante-natal and neonatal care of
isoimmunized women and their babies, has all but
eradicated serious morbidity and mortality associated
with this condition. Some of the key landmarks in the
evolution of this success story are listed in Table 6.2. By
the early 1960s Stern had demonstrated that exogenous anti-D given to RhD negative individuals could
prevent immunization occurring when RhD positive
blood was transfused into them.
Exogenous anti-D is produced by exposing RhD
negative volunteers to the RhD antigen. These individuals are either male, or are women who have completed
their families. They regularly donate their blood, and
cold-ethanol precipitation is used to separate the

Section 2. Feto-maternal alloimmune Syndromes

Table 6.3 Tests used to quantify the size of a FMH
Kleihauer : Fetal hemoglobin (HbF) is more resistant to acid or
alkaline elution than adult hemoglobin. After treatment, any
erythrocytes containing HbF retain their hemoglobin and can
be stained and recognized. Unfortunately, some adults have
persistent HbF production, and this can confuse matters.
Furthermore, quantification is less precise with bigger bleeds.
Flow cytometry: This uses immunofluorescently stained
antibodies to recognize fetal erythrocytes, which can then be
flow-sorted and quantified. This method is often preferred for
larger bleeds.

immunoglobulins from their hyperimmune plasma.
Following the emergence of variant Creutzfeldt–Jakob
disease in the UK, only plasma from US volunteers
has been used more recently, although it is not known
for certain if prions can be transmitted via transfused immunoglobulins. A solvent/detergent treatment inactivates HIV, hepatitis B and hepatitis C. BPL,
one of the major manufacturers of anti-D, estimates
a risk of viral infection of 1 in 10 000 billion doses of
their product and, to date, there have been no recorded
cases.
There were theoretical concerns that passive antiD might itself cause haemolysis within the fetus. There
is certainly no doubt that it can cross the placenta.
Although a small number of babies were born in the
anti-D trials with a weakly positive direct antiglobulin test (DAT), the reaction was insufficiently strong to
cause significant hemolysis or anemia.
Delivery was recognized to be the time of greatest risk for FMH and by the end of the 1960s
widespread post-natal prophylaxis had been introduced. A Cochrane review of six eligible trials of routine postpartum anti-D prophylaxis gives a relative risk
of 0.12 for RhD alloimmunization in the subsequent
pregnancy, i.e. a tenfold reduction in the incidence of
isoimmunization.1 Various doses of anti-D have been
tried, and indeed protocols still vary around the world
today. Doses of less than 500 iu are associated with a
greater risk of isoimmunization; however, higher doses
do not seem to confer any obvious benefit. 125 iu antiD, is able to neutralize 1 ml of fetal red blood cells.
Feto-maternal hemorrhage (FMH) of ≥30 ml occurs
in only 0.6% of all deliveries. A dose of 1500 iu has
been adopted in the USA to cover the possibility of
larger hemorrhages. In the UK and France, a smaller
dose of 500 iu is routinely used; however, a test is also
performed to quantify the size of the FMH (Table 6.3).
Occasional bleeds exceeding 4 ml are recognized and a
higher dose of anti-D is administered.

The anti-D is usually given by intramuscular injection (although intravenous preparations are available)
and ideally should be given within 72 hours of delivery
(or any other possible sensitizing event). There may,
however, be benefit in giving anti-D as much as 9–10
days following potential isoimmunizing events.
Later came the recognition that a variety of events
during pregnancy might cause or be associated with
FMH, other than delivery, and that these might subsequently also lead to isoimmunization (Table 6.1). The
RCOG Green-Top Guideline (No. 22) lists these situations and recommends the use of anti-D prophylaxis
in these scenarios also.2 The RhD antigen is thought to
be expressed as early as 7–8 weeks gestation and there
is no doubt that FMH can be demonstrated during the
first trimester. As little as 0.25 ml of fetal RhD positive blood may be sufficient to cause isoimmunization
and older studies have shown that this value is often
exceeded with FMH occurring after 8 weeks. The studies examining the risk of first trimester isoimmunization are old, and few in number.3 The risk probably lies
between 0 and 3%, but does seem to be higher when the
uterus is instrumented.
The RCOG have recommended anti-D only for
miscarriages prior to 12 weeks if the uterus is instrumented. After 12 weeks, and before 20 weeks, 250 iu
of anti-D should be given for all threatened and actual
miscarriages. Miscarriages and other potential sensitizing events after 20 weeks should be covered by 500 iu
of anti-D and a Kleihauer should be taken to identify
those cases where the size of the FMH exceeds 4 ml.2

Routine antenatal prophylaxis
Even in the absence of defined events known to
be associated with FMH, leakage of fetal red blood
cells into the maternal circulation is known to occur
throughout pregnancy. Beyond 28 weeks’ gestation the
quantity of trafficked cells can be great enough to
bring about alloimmunization. Indeed, “silent” FMH
will cause RhD isoimmunization in 1%–2% of all RhD
negative women with RhD positive pregnancies.
There is good-quality evidence supporting the use
of routine ante-natal anti-D prophylaxis (RAADP) to
prevent these isoimmunizations. A consensus conference hosted by the RCOG and RCP in 1997 came
out strongly in favor of routine ante-natal prophylaxis.
Crowther subsequently published a systematic review
in the Cochrane database, although only two trials
were deemed of high enough quality to be included.

Chapter 6. Red cell alloimmunization

This review reported a relative risk of isoimmunization
of 0.4 in the women receiving RAADP. More recently,
a Technology Appraisal Guidance (No. 41), produced
by NICE,4 has reviewed the wider evidence from nine
trials. Although the trials varied in design and
methodology, they gave remarkably consistent results.
Without RAADP the isoimmunization rate ranged
from 0.9%–1.6%. This fell to approximately 0.3% in the
groups receiving RAADP.
A number of attempts at estimating the cost effectiveness of this intervention have been made. The
number of HDN related deaths would be reduced
from approximately 30 to 10 per year in the UK if
all women received RAADP. The cost–benefit seems
clear for women in their first pregnancy, but less so for
parous women. Ultimately, however, both the RCOG
and NICE have recommended RAADP for all RhD
negative women, irrespective of parity.
The following dosage schedules are currently in use
in the UK:
A 500 iu at 28 weeks and 34 weeks’ gestation
B A single dose of 1500 iu at 28 weeks’ gestation
Schedule A was used in the only randomized controlled trial of RAADP (Hutchet) and was most widely
adopted in the UK.4 The half-life of anti-D is 24
days and theoretically there is less circulating antiD left at 40 weeks’ gestation with schedule B than
with A. The trials using this regime however did
not show significantly poorer results. Commercially
available preparations of 1500 iu anti-D have recently
become available in the UK and there is a move
toward schedule B, mostly for reasons of convenience
and patient preference (one injection rather than
two).

Refusal of anti-D prophylaxis
A small minority of women will refuse anti-D, either
as part of RAADP or following potentially sensitizing events (including delivery), perhaps due to safety
fears or “needle phobia.” The woman should be provided with good-quality information to ensure that
this choice is truly informed; however, the final decision of course must lie with her. Declining anti-D prophylaxis carries no risk when;
1. the woman is confident she is not going to have
further children (e.g. requesting sterilization); or
2. when the father of the baby, or the fetus itself, is
known with certainty to be RhD negative.

Widespread non-invasive pre-natal fetal RhD testing is possible (see later) and, if adopted, will mean
that RAADP and the use of anti-D following sensitizing events will be reserved for women carrying a fetus which is RhD positive, or of unknown
status.5

Traditional management of
isoimmunization
Despite effective prophylaxis programs, new cases of
RhD isoimmunization do arise, either because guidelines are not followed appropriately, women fail to seek
medical advice around the time of potentially sensitizing events, or because of “silent” isoimmunizations,
perhaps occurring prior to 28 weeks’ gestation. Management of these pregnancies has become limited to a
relatively small number of centers. Preventing morbidity and mortality in these cases necessitates the identification of pregnancies at risk, subsequent monitoring of disease severity, and timely intervention in the
form of intrauterine transfusion and/or delivery of the
baby. Modern management is quite different to that of
even just 10 years ago and, to best appreciate the recent
advances made, a rapid review of traditional methods
is included here.

Historical perspectives
Routine maternal blood typing and serological testing was introduced in the 1950s. RhD negative women
with anti-D antibodies were recognized as being at risk
of having their pregnancies complicated by hydrops,
stillbirth, and hemolytic disease of the newborn.
Approximately 85% of the white European and North
American population is RhD positive, and just over
half are heterozygous. The offspring of RhD heterozygous males and RhD negative women are at 50% risk
of being RhD positive themselves, and 50% will be
RhD negative. The RhD negative fetus is at no risk
of hemolysis, whatever the levels of maternal anti-D.
Although RhD negativity in male partners could be
determined with certainty, predicting whether a RhD
positive man was homo- or heterozygous was imprecise prior to the advent of molecular biology and relied
on the results of serological testing with anti-sera to the
D, C, c, E and e antigens and racially specific incidence
charts. However, this prediction was inexact and, when
a male partner was thought to be heterozygous, the

Section 2. Feto-maternal alloimmune Syndromes

status of the fetus remained unclear. A RhD negative
pregnancy could be exposed to serial invasive testing
when there was no actual risk.
With the development of molecular genetic techniques, and improved understanding of the Rhesus
gene cluster, it became possible to determine RhD
status precisely using DNA amplification techniques.6
These are able to sensitively distinguish between
homozygotes and heterozygotes and can be applied to
DNA from amniocytes to precisely assign RhD positive or negative status to the fetus of a couple where
the male partner is heterozygous. A single amniocentesis meant that further testing could be avoided in 50%
of cases (those found to be RhD negative). Surveillance and invasive testing could then be appropriately
focused on the RhD positive pregnancies.
A number of different factors have been used to
time interventions in Rhesus disease. The simplest and
least sensitive of these is previous obstetric history.
Walker showed how, in a RhD isoimmunized pregnancy, the risk of stillbirth was 8% if there was no
previous history of HDN. This rose to 18% if a previous child had been moderately affected and to 58%
if there was a previous history of stillbirth caused
by hemolytic disease. The tendency for the disease
to become more severe, and at progressively earlier
gestations, was well recognized. Recent retrospective
reviews of isoimmunized pregnancies have confirmed
these historical conclusions. However, relying on previous history to guide intervention was imprecise and
hazardous.
Coombs demonstrated that the strength of anti-D
isoimmunization could be measured by serially diluting maternal serum until agglutination of RhD positive red blood cells no longer occurred. The more
doubling dilutions were required to lose this reaction,
the more anti-D must have been there to begin with.
Serial dilutions of 1/2, 1/4, 1/8, 1/16, 1/32, 1/64, 1/128
indicated progressively higher starting levels of antiD. More recently, levels of anti-D have been quantified more precisely in “international units per ml”
(IU/ml) using different techniques. Significant hemolysis is unlikely at levels below 4 IU/ml and is unlikely
to be severe at levels below 10–15 IU/ml. However,
this threshold too is insensitive and the relationship
between absolute anti-D levels and disease severity
weakens in pregnancies beyond the first where antibodies are detected.
Nevertheless, these factors have been used (and
still are to some degree) to decide when to investigate

further with amniocentesis, perform fetal blood sampling, or indeed deliver.

Amniocentesis
Immune-mediated hemolysis within the fetus generates bilirubin, which is excreted by the fetal kidneys into the amniotic fluid. Ballantyne, at the end of
the nineteenth century, recognized that yellow staining of amniotic fluid was associated with the subsequent development of severe jaundice in the newborn.
Bevis recognized that the degree of yellow pigmentation of amniotic fluid samples taken during pregnancy offered a guide to the final outcome; however, reliable measurement of bilirubin concentrations
proved difficult and the alternative technique of measuring the optical density shift caused by the bilirubin was adopted. Using a spectrophotometer, the optical density of amniotic fluid is assessed across a wide
spectrum of wavelengths. Bilirubin causes a shift in
absorption at the 450 nm wavelength and the degree
of this shift (⌬ OD450) is proportional to the concentration of bilirubin. In the early 1960s, Liley published a chart which could be used to estimate the
risk of severe anemia in an isoimmunized pregnancy
based on the ⌬ OD450 of amniotic fluid collected by
amniocentesis after 27 weeks’ gestation; the higher the
⌬ OD450, the greater the chance of severe fetal anemia. Results falling above a certain threshold (“Zone
3”) would prompt intrauterine blood sampling and a
subsequent transfusion if the fetus was found to be significantly anemic. When managing a RhD isoimmunized pregnancy, the timing of the first amniocentesis
was decided by a number of factors, including previous history and anti-D titer (or concentration). If the
⌬ OD450 fell below these thresholds, repeated amniocenteses were subsequently required at intervals of 1
to 4 weeks, depending on the initial result, the rate of
rise between successive samplings, the Rhesus history
and the anti-D level. In a group of pregnancies with a
high incidence of fetal anemia, the sensitivity for detection of severe anemia (Hb of less than 5 SD below the
mean) was found to be approximately 80%,7 meaning
that 1 in 5 severely anemic fetuses would be missed
by the screening test. Reducing the ⌬ OD450 threshold
above which fetal blood sampling would be performed
did improve the sensitivity to nearly 100% but went
hand-in-hand with a drop in the specificity (below
50%) and positive predictive value, meaning that a significant number of fetal blood samplings were being

Chapter 6. Red cell alloimmunization

Table 6.4 The disadvantages of amniocentesis and ⌬ OD450
measurements in the assessment of immune-mediated fetal
anemia
r
r
r
r
r

0.5–1.0% risk of miscarriage/preterm
delivery/chorioamnionitis/peri-natal loss with each
procedure
Limited performance as a screening test, particularly at
gestations <28 weeks
Cause of fetomaternal hemorrhage and subsequent rise in
antibody levels in 50% of cases (Bowell et al.)
A surrogate marker for fetal anemia. Particularly
problematic in Kell isoimmunized pregnancies (see later)
Unpleasant for the woman

prompted by the ⌬ OD450 when, in fact, the fetus was
not severely affected (false positives). Other studies
quote somewhat different sensitivities and specificities
but the overall message remains the same. Later, the
Liley charts were extrapolated backwards to 20 weeks’
gestation but this too was associated with a reduction in the sensitivity, as demonstrated by Nicolaides.
These were not the only weaknesses of amniocentesis
used in this way (Table 6.4).
Nevertheless, when the procedure-related risks
associated with amniocentesis (0.5%–1.0%) were compared with those of fetal blood sampling (1%–4%)
the benefit in “screening” by ⌬ OD450 prior to fetal
blood sampling seemed clear. The optimum timing of
first, and repeated, amniocenteses required significant
experience but it soon became adopted as standard
practice in most centers. Some questioned whether,
with improvements in fetal blood transfusion techniques, it should be abandoned; however, the practice
continued until newer non-invasive methods became
more widespread at the beginning of the new millennium. It is rare now for amniocentesis to be performed
in the management of RhD isoimmunization.

Fetal blood transfusion
Pre-natal treatment options for RhD hemolytic disease, other than preterm delivery, really began in the
1960s when Liley showed that the fetus could be transfused in utero by injection of blood into the fetal peritoneal cavity under X-ray guidance. A radio-opaque
dye was injected into the amniotic cavity and taken
up by the fetus. On reaching the bowel, it outlined the
peritoneal cavity, into which the blood was injected.
Erythrocytes were then absorbed directly across the
bowel wall into the fetal intravascular compartment.
This hazardous procedure was later superseded by
ultrasound guided transfusions into the peritoneal

Table 6.5 Historical landmarks in the development of fetal
blood transfusion
1963 (Liley)

X-ray guided intraperitoneal IUT

1964 (Freda and
Adamsons)

Direct IUT through hysterotomy
incision

1981 (Rodeck)

Fetoscopically guided intravascular IUT

1981 (Berkowitz and
Hobbins)

Ultrasound guided intraperitoneal IUT

1986 (Nicolaides and
Rodeck)

Ultrasound guided intravascular IUT

cavity and then directly into the fetal circulation. The
timeline of the evolution of this remarkable fetal therapy is detailed in Table 6.5.
Intraperitoneal transfusion (IPT) had a number of drawbacks. No pre- or post-transfusion fetal
hemoglobin level or hematocrit was available, so the
volume to be transfused was, at best, an educated
guess. Absorption of the donated red cells across the
bowel wall was slow and this was of particular concern for the very anemic fetus where immediate and
rapid correction was needed. The presence of edema
in the bowel wall of a hydropic fetus could interfere
with absorption completely. Furthermore, overdistension of the fetal abdomen by the donated blood possibly endangered the fetus by interfering with venous
return and cardiac output. As intravascular transfusion (IVT) became more popular, comparative studies
suggested a sixfold greater risk of fetal death following
IPT.
Ultrasound guided direct intravascular transfusions are performed via a number of different routes.
Direct intracardiac transfusion is possible but, for
understandable reasons, is not ideal. The majority of
fetal IVT are performed into the umbilical vein at the
insertion of the umbilical cord into the placenta or
percutaneously into the intrahepatic portion of the
umbilical vein. 8 A free loop of cord can be used if the
placental insertion cannot be accessed. There is little evidence to suggest any one method is superior
to the others. Whichever route is favored, the uterine wall, amniotic membrane, and sometimes the placenta, must be breached by the needle. The risk of fetal
loss, membrane rupture, bradycardia, or fetal bleeding requiring delivery is quoted between 2% and 4%
per procedure, but this depends on gestation, operator experience, and fetal condition prior to the transfusion. Transfusions below 20 weeks gestation are a particular challenge and fetal loss rate is at least 10% if
there is hydrops.

Section 2. Feto-maternal alloimmune Syndromes

The blood used for a fetal transfusion is crossmatched against the maternal blood and should
be fresh, CMV negative, and irradiated. The crossmatched unit ideally has a hematocrit of 75%–80%, so
as to minimize the volume load of the transfusion on
the fetus. A degree of “over transfusion” to a hematocrit
of 40%–50% will prolong the interval between subsequent transfusions. However, the transfused blood is
acidotic and this and the volume load can be hazardous
to the very compromised hydropic fetus. In these cases,
a hematocrit of 25% should be aimed for with the first
transfusion which can be followed 2 or 3 days later
with a second “top-up.” The total volume (in ml) to
be transfused (VT ) is determined by the fetoplacental blood volume (VFet ), which increases with gestation, the hematocrit of the fetal and donated blood
(HctFet and HctDon ), and the target “desired” hematocrit (HctDes ) according to the equation:
HctDes − HctFet
VT (ml) = VFet (ml) ×
HctDon − HctDes

There is a steady increase in the total blood volume of
the fetoplacental circuit (VFet ) with advancing gestation, from approximately 25 ml at 20 weeks’ gestation
to 100 ml at 28 weeks’, and 210 ml at 34 weeks’ gestation.
Maternal sedation is usually employed because the
procedure can be uncomfortable and may last more
than 30 minutes. These sedative drugs may also help
to reduce fetal movement. Antibiotics and oral tocolytics are used by some operators, but there is no evidence to support or refute this practice.The maternal
abdomen is sterilized and draped and strict aseptic
technique is followed. Local anesthetic is injected into
the maternal abdominal wall. Under ultrasound guidance, a 20 Gauge needle is inserted through the uterine wall into the umbilical cord, or the vessels within
the fetal liver. A small sample of blood is taken to
confirm correct positioning and the hematocrit measured immediately, so that the total volume required
can be calculated rapidly. Transfusion of blood at 5 ml
per minute is usually tolerated well by the fetus. A
post-transfusion hematocrit is taken to help guide the
interval between transfusions. The hematocrit drops
by approximately 1% per day, but this rate may decline
as subsequent transfusions replace the fetal RhD positive blood with donated RhD negative blood, which
survives longer. The interval is usually 2–3 weeks and
newer scanning techniques may help to fine-tune this
(see later). Serial transfusions are usually performed

until 34–36 weeks’ gestation, after which delivery is
organized. Severe, early onset hemolytic disease may
necessitate more than five transfusions in a single pregnancy. A vaginal birth is aimed for unless there are
other obstetric factors, or the fetus is sick.

Recent advances in management
Despite the declining incidence of RhD isoimmunization, the management of affected pregnancies has not
stood still. The need for invasive assessment of the
pregnancy at risk can be directed now with greater
precision, and newer treatment options have shown
promise in helping to avoid the need for transfusions
at very early gestations.

Non-invasive testing for fetal
RhD status
The plasma of pregnant women contains free (i.e. noncell associated) fetal DNA (ffDNA) in significant quantities, from the early first trimester. The main source of
this DNA is debated, but it probably originates from
trophoblastic cells at the maternal–fetal interface. This
DNA is fragmented and is degraded rapidly. Maternal free DNA is present in much larger quantities.
Lo, in 1997, demonstrated how ffDNA could be used
for the non-invasive pre-natal diagnosis of fetal RhD
status9 and, since then, the same principles have been
applied to non-invasive fetal sexing and the inheritance of paternally derived disease causing mutations.
This technology is strongly promoted by “SAFE” (The
Special Non-invasive Advances in Fetal and Neonatal
Evaluation Network), a multinational European group
established in 2004 and funded by monies from the
European Union.10
A RhD negative woman should have no RhD DNA
sequences in her plasma because her negative status
is usually caused by a deletion of the RhD gene (but
see later). If DNA probes designed to recognize RhD
gene exons are added to her plasma, along with DNA
polymerase, then no product should form if the fetus
is also RhD negative, because there are no binding
sites for the probes. If the fetus is RhD positive, then
the probes will bind to the ffDNA and a PCR product
will be produced, which can be detected easily using
standard molecular techniques. Three separate exons
from RHD are amplified, and if only one or two of the
PCR products is generated then the result is considered

Chapter 6. Red cell alloimmunization

equivocal and further investigation is required before
a result can be given.
Use of this technology has become almost routine practice in the UK, and in parts of Europe, where
the father of the baby is a RhD positive heterozygote,
where his status is unknown, or if paternity is uncertain. The team performing this work in the UK is based
at the International Blood Group Reference Laboratory in Bristol, and their results are impressive.11 Over
a thousand samples have so far been tested and there
have only been two false positive and two false negative results. The accuracy of the test is independent
of gestation, however the rate of inconclusive results is
higher at gestations below 16 weeks and the test must
be repeated in a fifth of cases (personal communication March 2008). A simple maternal blood test effectively avoids the need for amniocentesis to determine
fetal RhD status when the father is heterozygous, or
his blood group is unknown. This, in turn, bypasses
the 1% risk of miscarriage associated with the amniocentesis, and the likely fetomaternal transfusion of red
cells which may cause a subsequent rise in anti-D levels in already isoimmunized women. A RhD negative
result provides welcome relief and reassurance without
having put the pregnancy under any risk at all.

Middle cerebral artery blood flow
Since the late 1980s the relationship between fetal anemia and the velocity of blood flow in the middle cerebral artery (MCA) has been clearly documented. This
can be measured using ultrasound, and a collaborative
group, led by Mari,12 are usually credited with enhancing the profile of this technique so that today it has
effectively replaced amniocentesis in the monitoring of
pregnancies complicated by red cell antibodies.
A decrease in total red cell mass results in a reduction in blood viscosity and an increase in cardiac output. The effect on viscosity is thought to be the principal mechanism causing an increase in fetal peak systolic blood flow velocities (PSV). The vessel where this
can be measured most easily and reliably is the middle
cerebral artery.
The technique is not difficult; however, a number
of factors influence the MCA PSV (see Table 6.6) and
guidelines must be adhered to strictly. The angle of
insonation of the pulse wave Doppler must be as close
to 0 degrees as possible, or angle correction must be
used. The vessel must be insonated within the first
2 mm of the proximal portion of the vessel as it arises

Table 6.6 Factors influencing the MCA PSV value
Fetal
Gestational age
Fetal activity
Cardiac status
Gender
Uterine contractions
Technical
Angle of insonation of Doppler wave
Positioning of the Doppler gate along the MCA

Fig. 6.4 Power Doppler study showing the fetal Circle-of-Willis and
the near-field middle cerebral artery (MCA)

from the Circle of Willis (see Figs. 6.4 and 6.5). Measurements taken at a distal point in the vessel may be
6–10 cm s−1 less than values taken proximally. Usually
the MCA closest to the transducer is chosen, for ease,
but the far-field vessel gives similar results. The fetus
must be quiescent as fetal movement and breathing can
have a significant impact on the values obtained.
Normal ranges and charts (Fig. 6.6) are available
to plot the values onto and these show a normal gradual increase in the MCA PSV as gestation advances.
Although the trend in values is important in any particular case, the most valuable indicator of significant
fetal anemia has proven to be a threshold of 1.5 MoM
(multiples of the median). Below this level, it is highly
unlikely that the fetus will be more than just mildly
anemic. The higher the value lies above this line, the
more likely moderate or severe anemia becomes. If the
MCA PSV in an “at-risk” pregnancy is found to fall
above this threshold, the study will usually be repeated
within 24 hours. If the finding is persistent, then fetal
blood sampling (with or without transfusion) will usually be performed soon after. MCA PSV studies are
usually performed at intervals varying from 3 days to

Section 2. Feto-maternal alloimmune Syndromes

Fig. 6.5 Color Doppler measurement of
middle cerebral artery peak systolic
velocity (MCA PSV). This study was
performed at 22 weeks’ gestation. A PSV
of almost 50 cm s−1 in the MCA at this
gestation is well above 1.5 multiples of
the median and did indeed indicate fetal
anemia in this case. Although the
circulation is hyperdynamic in fetal
anemia, the actual cause of the rise in
blood velocity is thought to be a
reduction in blood viscosity secondary to
a falling fetal hematocrit. The fetal
hemoglobin was found to be 4 g/dl on
subsequent testing.

1.5 MoM

MCA PSV (cm s−1)

60
Median
40

Gestational age (weeks)

4 weeks, depending on the perceived degree of risk,
and previous values.
The collaborative group found a 100% sensitivity
for the prediction of moderate to severe anemia, with a
false positive rate of 28%.13 This compares very favorably with amniocentesis. Other studies have failed to
achieve quite such impressive results, although those
of Zimmermann are typical and are at least as good
as amniocentesis. They found a sensitivity of 88% and
a positive predictive value of 53%, i.e. approximately
1 in 10 cases of moderate-severe anemia were missed
and half of all cases with a PSV greater than 1.5
MoM required transfusion. They recommended that

Fig. 6.6 A chart showing median fetal
middle cerebral artery peak systolic
velocities (MCA PSV) throughout the
second and third trimesters. An imaginary
line has been drawn at 1.5 multiples of
this median value and acts as an action
line to prompt fetal blood sampling +/−
transfusion. Superimposed is a fictitious
case of severe RhD isoimmunization with
a RhD positive fetus. The blue arrows
show where maternal IVIG was given, and
the red arrows represent fetal blood
transfusions. The MCA PSV can be seen to
fall immediately following a fetal blood
transfusion, but gradually increases again
as hemolysis continues. As the fetal RhD
positive blood becomes replaced by
successive donations of RhD negative
blood, the interval between transfusions
increases. However, this donated blood
also has a limited lifespan, even though it
is not subject to the antibody mediated
hemolysis.

this technique should not be used after 35 weeks’ gestation, when the false positive rate is higher. Other
studies have suggested that the MCA PSV operates
well in the second trimester, in contrast to ⌬ OD450.13
Even if MCA PSV only matches the predictive abilities of liquor ⌬ OD450 measurements, the advantage
remains clear; the technique is non-invasive. Indeed,
by adopting MCA PSV measurements as the method
of determining when fetal blood sampling is required,
the number of invasive procedures can be reduced by
two-thirds. Furthermore, reducing the time interval
between MCA studies should improve the sensitivity of the test. Recent studies show that it remains a

Chapter 6. Red cell alloimmunization

Table 6.7 Possible mechanism of action of IVIG in Rhesus D
isoimmunization

Table 6.8 Management strategies where moderate to severe
HDN is expected

Increased catabolism of maternal IgG

Close communication with neonatal team

Competitive blockade of placental IgG transport mechanisms

Planned delivery (induction or Cesarean section)

Fc receptor blockade within the fetal reticuloendothelial system

Cord blood analysis for bilirubin, Hb, group and DCT

Precipitation of immune complexes by excess antibody

Drainage of effusions and ascites in the hydropic infant

Antigen neutralization

Phototherapy

Binding of exogenous anti-idiotypic Ab to endogenous Ab

Intravenous immunoglobulin
Exchange transfusion

useful tool for timing second, third, and subsequent
fetal blood transfusions.

Adjunctive ante-natal treatments
Targeting the maternal immune response is a tempting strategy for tackling severe cases of isoimmunization. The knowledge that disease severity is, at least in
part, related to absolute anti-D levels led to the proposal that plasmapheresis might help ameliorate the
disease. Anti-D levels can be kept under control with
this technique, but it is not without maternal risk,
causes a rebound of antibody levels when treatment
comes to an end, and was never convincingly shown
to make a difference in erythroblastosis fetalis. For
these reasons, its use as an adjunct to well-established
management techniques had fallen out of favor until
more recently when it has been used in combination
with a second form of immunomodulation which itself
has shown greater promise. The use of intravenous
immunoglobulin (IVIG) to prevent/treat fetal and
neonatal alloimmune thrombocytopenia is described
in Chapter 5 and there is more evidence of its value
in this condition than there is for Rhesus isoimmunization. Nevertheless, non-randomized studies provide support for its use in cases of severe RhD
isoimmunization. 1 g/kg is administered on a weekly
or fortnightly basis from 13 to 20 weeks’ gestation,
the aim being to delay the onset of moderate-severe
hemolysis to a point in the pregnancy where IVT can
be more readily and reliably performed. The number of cases of hydrops can be reduced, as can the
number of fetal blood transfusions needing to be performed. Table 6.7 lists some of the possible mechanisms by which IVIG might work. A more recent
study has reported on a combination of IVIG with
serial plasmapheresis from 12 weeks’ gestation for
women with the most severe histories, with impressive outcomes.14 The contribution of the two different methods of immunomodulation is not possible to

Recombinant erythropoietin
Top-up blood transfusions

assess. IVIG is expensive, and prepared from multiple
donors. Rarely, it may cause unpleasant and potentially
serious side effects, including;
r pyrexia and rigors;
r headache, backache, and myalgia;
r hypotension and tachycardia;
r tachypnea and chest tightness;
r alopecia;
r hemolytic anemia;
r renal impairment.
Figure 6.6 illustrates how serial MCA PSV monitoring, early IVIG administration and multiple fetal
blood transfusions can support a pregnancy at risk of
early and severe erythroblastosis fetalis through to a
gestation where induction can be expected to bring
about the normal birth of a non-hydropic baby with
adequate hemoglobin levels.

Pediatric management
It is rare now for a baby to be born at risk of HDN without the prior knowledge of maternity and pediatric
staff. A multidisciplinary approach is vital for optimizing outcomes. Infants born with only a low risk of significant hemolysis should, at the very least, have cord
blood sent for Coombs test (DCT), blood group,
hemoglobin and bilirubin levels. Close observation
over the next 2–3 days is necessary and repeat bilirubin
estimations may be required, as may phototherapy.
Management strategies for more significant cases
of HDN are listed in Table 6.8.
It is far preferable to treat a hydropic fetus in
utero than it is to deliver the baby in such poor condition. However, complications from an intrauterine
transfusion may precipitate the unplanned delivery
of such a baby in which case intubation, ventilation,

Section 2. Feto-maternal alloimmune Syndromes

and drainage of pleural effusions and ascites will be
required. These babies are volume overloaded, making
transfusion hazardous (although still necessary). They
are at risk of hypoglycemia, hypocalcemia, hyponatremia, hyperkalemia, hyperbilirubinemia, acidosis,
and renal failure. Mortality rates are high.
With modern ante-natal and fetal management,
this situation is fortunately rare. Nevertheless, planning delivery is important for the cases of moderate
or severe erythroblastosis fetalis, even if intrauterine
transfusions have minimized the risk. Bilirubin levels will rise sharply after birth and phototherapy must
begin immediately.
Light from the blue–green region of the spectrum (425–490 nm) is most effective at converting non-polar bilirubin to water-soluble photoisomers and fluorescent tubes producing irradiance of
⬎30 ␮W/cm2 /nm are optimal. The surface area of the
baby exposed to the light is crucial, and fiber-optic
pads placed under the neonate, or the use of specifically designed “bili-beds,” ensure that this is maximized. Bilirubin levels must be measured regularly and
phototherapy may need to continue for a number of
days. Gestation specific charts are available for bilirubin levels and thresholds for exchange transfusion are
recognised. A rise in serum bilirubin beyond these levels puts the newborn at increasing risk of kernicterus.
Severe anemia, high absolute bilirubin levels,
excessive rise in bilirubin concentration, and unsafe
bilirubin-to-albumin ratios are all indicators for
exchange transfusion. An intravenous catheter is
placed into the inferior vena cava via the umbilical vein
through the cord stump and the entire blood volume of
the neonate is usually replaced twice (“double-volume”
exchange) by removing neonatal blood and replacing it
with RhD negative blood in 5–10 ml aliquots. This process removes bilirubin and antibody-coated red blood
cells, and at the same time provides new albumin with
unoccupied bilirubin binding sites and RhD negative
erythrocytes. Between 70% and 90% of all fetal red
blood cells are removed, but because most of the bilirubin is in the extravascular compartment, 75% of total
body bilirubin remains and can cause a rebound rise
in serum levels soon after the exchange, necessitating a repeat procedure. The inherent risks of exchange
transfusion are substantial, however (Table 6.9),15 and
experience with the technique is declining. As many as
1-in-20 infants undergoing exchange transfusion may
die and 1-in-4 suffer non-fatal complications. Much
of this morbidity and mortality is found in preterm

Table 6.9 Potential complications of exchange transfusion
Haematological : Over-anticoagulation with hemorrhage, anemia, neutropenia, thrombocytopenia
Cardiac : Volume overload, congestive heart failure, hypertension, arrhythmia, arrest
Metabolic : Acidosis, hypocalcemia, hypoglycemia, hyperkalemia, hypernatremia,
Vascular : Thromboembolic events, necrotizing enterocolitis,
vessel perforation
Infectious : Bacterial, viral, malarial
Other : Hypothermia, apnea, bowel perforation

babies, emphasizing again the massive impact that
ante-natal management has had on this condition by
delaying the gestation at which the baby needs to be
born. Furthermore, intra-uterine transfusions provide
the fetus with red blood cells not at risk of immunemediated hemolysis. By the third IUT, the fetal blood
will be almost entirely RhD negative. At birth therefore, these babies paradoxically are less likely to need
exchange transfusion.
Avoiding exchange transfusion is clearly beneficial.
The use of intravenous immunoglobulin is well established now in the treatment of neonatal alloimmune
thrombocytopenia and HDN. Although a number of
mechanisms are possible, the main action is thought
to be a blockade of Fc receptors in the reticuloendothelial system. IVIG reduces carboxyhemoglobin levels, a
sensitive indicator of hemolysis. Although IVIG is prepared from multiple donors, and is extremely expensive, its use as an adjunct in moderate-to-severe HDN
seems justified. A Cochrane systematic review in 2002
concluded that IVIG significantly reduces the need for
exchange transfusion (RR = 0.28), and reduces the
number of exchanges needed when they cannot be
avoided.16 However, better quality studies are few in
number and there has been a call for larger randomized trials. It should be used only as an adjunct to phototherapy and 0.5–1.0 g/kg is usually given as a single
dose soon after delivery.
The baby remains at risk of developing anemia for
some months, for two reasons. Firstly, maternal antibodies circulate for 4–6 months and continue to cause
low-grade hemolysis. Secondly, intrauterine and newborn transfusions may suppress normal erythropoiesis
and it may be a number of months before reticulocytes appear. During this time, “top-up” blood transfusions may be required, although these carry minimal risk in comparison with exchange transfusion.

Chapter 6. Red cell alloimmunization

Regular recombinant erythropoietin (EPO) injections
can be used during this time to limit the number of
top-up transfusions required.

Rhesus D variants
RHD is a complex gene and much variation exists
within it, particularly between racial groups. Understanding this is crucially important for a number of
reasons. Phenotypic tests of RhD status examine how
blood from an individual behaves when it is added
to serum containing anti-D antibodies. Agglutination
indicates that the individual is RhD positive. Genotypic tests of RhD status look for key DNA sequences
from RHD. The entire gene cannot be examined, so
sections from a variety of coding exons are chosen for
multiplication, using the polymerase chain reaction.
If a PCR product is produced, then the assumption is
made that the individual is phenotypically RhD positive.
RhD variants can confuse this. The common cause
for RhD negativity in Africans is an allele called the
RHD pseudogene (RHD␺ ) which contains a 37 base
pair insert in exon 4 and a nonsense mutation in exon
6, which effectively make the protein non-functional.
Phenotypically, these individuals are RhD negative,
but genetic tests might, for example, amplify exon 7
successfully and give a false positive result for RhD
status. Amplifying more exons, such as exon 5, would
allow clarification because this exon is amplified from
the normal RHD but not RHD␺ . This has particular relevance to non-invasive pre-natal RhD testing.
Knowledge of the racial origin of the woman’s partner is clearly important. A second common African
variant is the RHD/CE hybrid allele, of which there
are more than 20. In these alleles, entire segments
of the RHCE gene have been substituted into RHD.
These red blood cells will agglutinate with polyclonal
serum, but fail to react with monoclonal antibodies
raised specifically against the extracellular loops coded
by the missing exons. Although certain RHD exons
will amplify with standard RHD probes, not all will
(because they are missing), and this allows them to
be distinguished from true RhD positive individuals
using genetic rather than serological tests.
These RHD/CE hybrids are usually known as “partial D” alleles. The changes usually affect a long string
of amino acids which is always located on the erythrocyte surface. The protein is altered so dramatically in these external “antigenic” portions that it is

not recognized by anti-D and these individuals are
prone to true RhD isoimmunization if exposed to normal RhD positive red cells. In the majority of cases,
women with these alleles should be treated like RhD
negative women and offered RhD negative blood if it
is required and anti-D if the fetus is possibly or definitely RhD positive. It is very important that genetic
tests in this group do not falsely classify them as RhD
positive and a variety of strategies are in place in most
laboratories to prevent this happening. The most common European partial D variant is DNB, caused by a
missense mutation, which alters one amino acid in the
sixth extracellular loop of the protein.
The second group of RHD variants is known as
“weak D.” The changes within these alleles substitute amino acids in the transmembranous portions
of the protein. The surface antigenic sites are unaltered; however, integration of the weak D protein into
the cell membrane is hindered or rendered unstable, effectively reducing the number of RHD antigenic
sites expressed per red cell. The effect is quantitative
rather than qualitative. Blood from these individuals
will eventually show agglutination with anti-D if given
more time and assisted by the addition of anti-human
globulin reagent. Weak D type 1 is the most common
European weak D variant and is caused by a single missense mutation at amino acid 270. Most women with
weak D variants (type 1–3) can be given RhD positive
blood and do not need prophylactic anti-D, although
there are a few exceptions.
The term “Du” was previously applied to variants
of RHD. In view of the complexity of the situation,
and the consequences of treating women as RhD positive when in fact they carry a variant which puts them
at risk of isoimmunization against RhD, it is recommended that advice is taken from the laboratory performing the serological and molecular tests in each
case where a variant is identified.

Other red cell antibodies
Table 6.10 lists some of the other red cell antigens,
which have been documented to be the target of maternal antibodies, resulting in hemolytic disease. Those
highlighted are the most significant from a clinical
perspective. Certain red cell antibodies never cause
hemolysis. Regional blood transfusion services will
advise where rare antibodies are discovered on antenatal screening.

Section 2. Feto-maternal alloimmune Syndromes

Table 6.10 Other red cell antigens implicated in
fetal/neonatal hemolytic disease
Kell
Rhesus c, C and E
A and B
Fya (Duffy) and Fyb
Jka (Kidd) and Jkb
S, M and U
Lua

types, effectively diluting their effect on red blood
cells. Secondly, cell-surface expression is incomplete
during gestation and develops gradually, thus limiting the risk before birth. Jaundice caused by ABO
incompatibility is usually mild and readily treated with
phototherapy.

The future

Isoimmunization against the Kell antigen deserves
special mention. Although there are four separate Kell
antigens, Kell 1 causes most concern. Anti-Kell1 antibodies are the second most common cause of fetal
immune mediated hemolysis and early onset anemia
and hydrops have been well documented. Nine out of
10 of the general population are Kell1 negative and
only 1 in 20 babies of Kell1 negative women are Kell1
positive. The Kell antigen is expressed on red cell
progenitors in the bone marrow and it is via these
cells that anti-Kell1 antibodies are able to suppress
hematopoiesis, as well as causing hemolysis. This made
ante-natal surveillance with amniocentesis unreliable
because ⌬ OD450 of amniotic fluid acts only as a surrogate for hemolysis and cannot estimate the impact that
the antibodies have on erythropoeisis. Fortunately, this
problem is not shared by MCA PSV, which is used in
exactly the same way as in RhD isoimmunization. In
approximately half of all cases of Kell isoimmunization the cause is a previous blood transfusion where
cross-matching did not take account of Kell status
of the woman, or donor. The remainder result from
FMH occurring at the delivery of a previous Kell positive baby. Absolute levels of anti-Kell antibodies are
less useful in the prediction of disease severity. Noninvasive pre-natal diagnosis (NIPD) for fetal Kell status is available through the BTS laboratory in Bristol
and is particularly helpful because most Kell positive
individuals are heterozygous. NIPD is also possible for
the Rhesus c and E antigens, although currently relatively few tests have been performed, when compared
with RhD, meaning that the degree of diagnostic certainty is less. Anything more than mild disease is very
unlikely with RhE antibodies.
Women who have the blood group O quite commonly have antibodies to the A- and B-antigens,
although these are more likely to be of the IgM class
which does not cross the placenta. Anti-A and antiB IgGs can reach the fetal circulation, but only mild
hemolysis is the general rule for two reasons. Firstly,
these antigens are expressed on a wide variety of cell

The successes of recent years have not brought research
and progress in the prevention and management of
RhD hemolytic disease to a close. Although polyclonal
anti-D is a safe product, it is pooled from various
donors and anxieties about viral and prion disease
transfection continue. The infection of hundreds of
Irish women with hepatitis C in 1977–78 following the
administration of contaminated anti-D illustrates this
point all too well and currently 4 out of 10 women in
the UK receiving anti-D will in fact be carrying a RhD
negative fetus. The anti-D for these women is unnecessary, unpleasant, expensive, and not without a degree
of risk.
Limiting the administration of anti-D to only those
RhD negative women carrying a RhD positive fetus is
a worthy goal. The techniques used for non-invasive
pre-natal fetus RhD testing (see above) are time consuming and expensive, although very accurate. Application of this technology to all pregnant RhD negative women is impractical. Mass screening requires an
automated test, and robotic systems have been developed and tested recently with very promising results.
Results from a Bristol study have given a detection
rate for fetal RhD positive status of 99.7% for a false
positive rate of 2%.5 These false positives represent a
group of RhD negative women who would continue to
receive anti-D unnecessarily, but 2% nevertheless is a
significant improvement on 38%. Of greater concern
are the false negative results (i.e. failure to detect fetal
RhD positivity) of which there were only 3 cases from
nearly 1200 RhD positive pregnancies. Although these
women would be at a three fold greater risk of isoimmunization because they would not receive ante-natal
prophylactic anti-D (they would still receive post-natal
anti-D), mathematical modeling shows that this actually equates to only one extra case of hemolytic disease
in 86 000 future pregnancies.
An alternative approach to improve on the
safety of anti-D prophylaxis is to use recombinant monoclonal antibodies (mAb) produced from
hybridoma or human B-cell lines, instead of polyclonal

Chapter 6. Red cell alloimmunization

antibodies collected from human serum. A number of
these cell lines exist and progress to date has recently
been summarized by Kumpel.17 Results with some of
the mAb are encouraging and D-immunization can
be prevented in RhD negative volunteers transfused
with RhD positive cells. A “clean” and effective
recombinant anti-D mAb may be on the horizon, but
hurdles still exist, not least obtaining ethical approval
for large-scale trials.
More distant are further exciting possibilities.18
Mutated recombinant anti-D monoclonal antibodies
have been designed and produced, which are able to
bind to the RhD antigen but have a much lower affinity for the Fc␥ receptor on macrophages than normal anti-D.19 These mutated antibodies would displace
endogenous anti-D from its binding sites on the RhD
antigen. Complement mediated lysis, hemolysis and
phagocytosis could all be reduced. There are many difficulties to overcome. The lifespan of these antibodies
is limited and very high and frequent maternal admin-

istrations might be needed for transplacental transfer
to maintain sufficient levels in the fetus.
A welcome move in neonatal care would be the
avoidance altogether of exchange transfusion. The
use of IVIG seems to have made some headway
with this however a further option is close at hand.
Competitive heme oxygenase inhibitors, such as tinmesoporphyrin, have recently undergone phase III
trials and are already available for certain conditions.
This structural analog of heme competitively blocks
heme-oxygenase, a rate limiting enzyme in bilirubin
production. Heme is left unaltered to be excreted in
bile. It does not pass through the blood–brain barrier
and does not accumulate in tissues. Several randomized trials have confirmed that these substances can
prevent and block jaundice progression in the newborn. In the future these drugs may result in a reduction in the need for phototherapy, and exchange transfusion for HDN may become a procedure confined to
the history books.

Section 2. Feto-maternal alloimmune Syndromes

References
1. Crowther C, Middleton P. Anti-D administration after
childbirth for preventing Rhesus alloimmunisation.
Cochrane Database of Systematic Reviews 1997 (2)
CD000021.
2. Royal Institute for Clinical Excellence. Guidance on
the use of routine antenatal anti-D prophylaxis.
Guideline no. 22. RCOG Press, 2002.
3. Jabara S, Barnhart KT. Is Rh immune globulin needed
in early first-trimester abortion? A review. American
Journal of Obstetrics and Gynecology 2003; 188:
623–627.
4. National Institute for Clinical Excellence. Guidance on
the use of routine antenatal anti-D prophylaxis for
RhD-negative women. Technology Appraisal
Guidance no. 41. www.nice.org.uk, 2002.
5. Finning K, Martin P, Summers J et al. Effect of
high-throughput RHD typing of fetal DNA in maternal
plasma: on use of anti-RHD immunoglobulin in RHD
negative pregnant women: prospective feasibility
study. British Medical Journal 2008; 336: 816–818.
6. Bennett PR, Le Van Kim C, Colin Y et al. Prenatal
determination of fetal RhD type by DNA
amplification. New England Journal of Medicine 1993;
329: 607–610.
7. Sikkel E, Vandenbussche FPHA, Oepkes D et al.
Amniotic fluid ⌬ OD450 values accurately predict
severe fetal anaemia in D-alloimmunisation. Obstetrics
and Gynecology 2002; 100: 51–57.
8. Nicolaides KH, Soothill PW, Clewell W, Rodeck CH.
Rh disease: intravascular fetal blood transfusion by
cordocentesis. Fetal Therapy 1986; 1: 185–192.
9. Lo YMD, Corbetta N, Chamberlain PF et al. Presence
of fetal DNA in maternal plasma and serum. Lancet
1997; 350: 485–487.
10. Chitty LS, van der Schoot, Hahn S, Avent ND. SAFE –
The special non-invasive advances in fetal and
neonatal evaluation network: aims and achievements.
Prenatal Diagnosis 2008; 28: 83–88.

11. Daniels G, Finning K, Martin P, Summers J. Fetal
blood group genotyping. Present and future. Annals
of The New York Academy Science 2006; 1075:
88–95.
12. Mari G, Deter RL Carpenter RL et al. For the
collaborative group for the assessment of the blood
velocity in anaemic fetuses. Non-invasive diagnosis by
Doppler ultrasonography of fetal anaemia due to
maternal red-cell alloimmunisation. New England
Journal of Medicine 2000; 342: 9–14.
13. Pereira L, Jenkins TM, Berghella V. Conventional
management of maternal red cell alloimmunisation
compared with management by Doppler assessment of
middle cerebral artery peak systolic velocity. American
Journal of Obstetrics and Gynecology 2003; 189:
1002–1006.
14. Ruma MS, Moise KJ, Kim E et al. Combined
plasmapheresis and intravenous immune globulin for
the treatment of severe maternal red cell
alloimmunisation. American Journal of Obstetrics and
Gynecology, 2007; 196: 138.e1–138.e6.
15. Jackson JC. Adverse events associated with exchange
transfusion in healthy and ill newborns. Paediatrics
1997; 99: E7.
16. Alcock GS, Liley H. Immunoglobulin infusion for
isoimmune haemolytic jaundice in neonates.
Cochrane Database of Systematic Reviews 2002; (3):
CD003313.
17. Kumpel BM. Efficacy of RhD monoclonal antibodies
in clinical trails as replacement therapy for
prophylactic anti-D immunoglobulin: more questions
than answers. Vox Sanguinis 2007; 93: 99–111.
18. Urbaniak SJ. Noninvasive approaches to the
management of RhD haemolytic disease of the fetus
and newborn. Transfusion 2008; 48: 2–5.
19. Nielson LK, Green TH, Sandlie I et al. In vitro
assessment of recombinant, mutant anti-D
immunoglobulin G devoid of haemolytic activity for
treatment of on-going haemolytic disease of the fetus
and newborn, Transfusion 2008; 48: 12–19.

Section

Thromboembolism and
anticoagulation

Section 3
Chapter

Thromboembolism and anticoagulation

Acute management of suspected
thromboembolic disease in pregnancy
Andrew J. Thomson and Ian A. Greer

Introduction
Antenatal and postnatal venous thromboembolism
(VTE) is around 10 and 25 times more common
respectively, than in non-pregnant women of the
same age and is the major cause of direct maternal
mortality in the developed world. European studies
have consistently found the pregnancy-related VTE
mortality to be 8.5 – 14 per million live births.1,2
In the United Kingdom, sequential reports from
Confidential Enquiries into Maternal Deaths have
demonstrated that VTE remains the main direct cause
of maternal death and have highlighted failures in
obtaining objective diagnoses and employing adequate
treatment.3 Fatal pulmonary embolism (PE) arises
from deep venous thrombosis (DVT), many cases of
which are not recognized clinically and are only identified at post-mortem following a maternal death.3
The subjective, clinical assessment of DVT and PE is
particularly unreliable in pregnancy and a minority of
women with clinically suspected VTE has the diagnosis confirmed when objective testing is employed.4
Acute VTE should be suspected during pregnancy
in women with symptoms and signs consistent with
possible VTE,4–6 particularly if there are other risk
factors for VTE (see Tables 8.1 and 8.2).7–11 The symptoms and signs of VTE include leg pain and swelling
(usually unilateral), lower abdominal pain, low grade
pyrexia, dyspnoea, chest pain, haemoptysis and
collapse.

Epidemiology of VTE during pregnancy
Virchow’s triad for VTE consists of alterations in normal blood flow (stasis), trauma or damage to the vascular endothelium and alterations in the constitution
of blood (hypercoagulability), and describes the three

broad categories of factors that contribute to thrombosis. During normal pregnancy, hypercoagulability
results from increases in the levels of factor VIII and
fibrinogen, reduction in protein S levels, a resistance
to activated protein C and impaired fibrinolysis. Studies assessing blood flow velocity in the lower limbs in
pregnancy have shown an extensive reduction in flow
of up to 50% by 29 weeks’ gestation, reaching its nadir
at 36 weeks. The changes in both blood flow velocity
and coagulation factors may persist for up to 6 weeks
after delivery. The third component of Virchow’s triad,
damage to the vascular endothelium, arises during the
course of vaginal or abdominal delivery – whilst VTE
can occur at any stage of pregnancy, the puerperium is
the time of greatest risk.
Almost 90% of cases of DVT occur in the left leg
in pregnancy, in contrast to the non-pregnant situation, where only 55% occur on the left.4 This may
reflect compression of the left iliac vein by the right
iliac artery and the ovarian artery, which cross the vein
only on the left side. Over 70% of DVTs in pregnancy
arise in the iliac and femoral veins rather than the calf
veins, whereas in non-pregnant patients only about 9%
arise in the ilio-femoral area. This is of importance
since ileo-femoral DVTs are more likely to result in PE
than are calf vein thromboses.

Assessment and diagnosis of acute
VTE in pregnancy
Clinical diagnosis of both DVT and PE is unreliable.
In non-pregnant patients where DVT is suspected, the
diagnosis is confirmed in about 20–30% of cases when
objective testing is performed. During pregnancy, clinical assessment is even more unreliable since many of
the symptoms and signs of VTE, such as leg swelling,

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

Section 3. Thromboembolism and anticoagulation

Table 7.1 – Symptoms and signs of VTE in pregnancy
Deep venous thrombosis
r leg pain or discomfort
r tenderness
r swelling
r lower abdominal pain
r increased temperature and oedema
r elevated white cell count
Pulmonary thromboembolism
r chest pain
r dyspnoea
r haemoptysis
r tachycardia
r focal signs in the chest
r raised jugular venous pressure
r collapse
r abnormalities on the chest X-ray
r symptoms and signs associated with DVT

chest pain and dyspnea, are commonly found in normal pregnancy (Table 7.1). As a consequence, the accuracy of clinical diagnosis falls to about 8% for DVT and
to less than 5% for suspected PE.4–6
It is therefore essential that objective testing is performed in women with suspected VTE. Failure to identify VTE will place the mother’s life at risk, whilst
unnecessary treatment is associated with risks, inconvenience and costs during the pregnancy and may also
have implications for her future health care (including
future use of oral contraception and hormone replacement therapy, and thromboprophylaxis in future pregnancies). There are no well-designed large clinical trials to support the management of suspected VTE in
pregnancy and guidelines are therefore empirical and
based on extrapolation from studies performed in
non-pregnant patients.1,2,4
If there is a delay in obtaining objective tests, the
woman should be commenced on anticoagulant therapy, unless contraindicated, until testing can be performed.

Diagnosis of DVT

Compression Duplex ultrasound of the entire proximal venous system is the optimal initial diagnostic
test for DVT in pregnancy. If the initial ultrasound
shows an abnormality in the popliteal or femoral veins,
the diagnosis of proximal DVT is confirmed and anticoagulant treatment should be commenced and continued. A normal ultrasound does not exclude a calf
DVT and therefore, if ultrasound is negative and a high
level of clinical suspicion exists, the patient should
remain anticoagulated and the ultrasound repeated in

one week or an alternative diagnostic test employed. If
repeat testing is negative, anticoagulant treatment can
be discontinued.4
For the diagnosis of iliac vein thrombosis, which
may present with back pain and/or swelling of
the entire limb, pulsed Doppler, magnetic resonance venography or conventional contrast venography should be considered.

Diagnosis of PE
When a woman with suspected PE is haemodynamically stable, a chest X-ray (CXR) should be performed.
CXR may identify other pulmonary disease such as
pneumonia, pneumothorax or lobar collapse. Whilst
the CXR is normal in over half of pregnant patients
with objectively proven PE, abnormal features caused
by PE include atelectasis, effusion, focal opacities,
regional oligaemia or pulmonary edema. The radiation dose to the fetus from a CXR performed at any
stage of pregnancy is negligible. If the CXR is abnormal, ventilation perfusion (V/Q) scanning is unreliable and CT pulmonary angiography (CTPA) should
be performed.4–6
If the CXR is normal, the authors recommend that
bilateral Doppler ultrasound leg studies should be performed. A diagnosis of DVT may indirectly confirm a
diagnosis of PE and since anticoagulant therapy is the
same for both conditions, further investigation is not
usually necessary. This would limit the radiation doses,
particularly associated with CTPA, given to the mother
and her fetus.4
The choice of technique for definitive diagnosis
(V/Q scan or CTPA) may depend on factors such as
local availability and guidelines, and should usually
be made after discussion with a radiologist. During
pregnancy the ventilation component of the V/Q scan
can often be omitted, thereby minimizing the radiation dose for the fetus. In the United Kingdom, the
British Thoracic Society5 recommends CTPA as firstline investigation for non-massive PE in non-pregnant
patients. This technique has potential advantages over
radionuclide (V/Q) imaging including better sensitivity and specificity, (at least in non-pregnant patients)
and a lower radiation dose to the fetus (see section
below). In addition it can identify other pathology
such as aortic dissection. The main disadvantage of
CTPA is the high radiation dose to the maternal breasts
associated with an increased lifetime risk of developing breast cancer. This is particularly relevant when it is

Chapter 7. Acute venous thromboembolism

known that only around 5% of such investigations will
have a positive result. In addition, conventional CTPA
may not identify small peripheral PEs, although this
is overcome by the latest multidetector row spiral CT
techniques. In contrast to CTPA, V/Q scanning may
be delayed because of availability of isotope. Despite
these potential advantages of CTPA, many authorities,
including the authors, continue to recommend V/Q
scanning, where possible, as first-line investigation in
pregnancy because of its high negative predictive value
in this situation, its substantially lower radiation dose
to pregnant breast tissue, and because most pregnant
women in the UK will not have co-morbid pulmonary
pathology.4,12

Radiation exposure associated with
diagnostic tests
CTPA delivers less radiation to the fetus than V/Q
scanning during all trimesters of pregnancy. It has
been estimated that the risk of fatal cancer to the age
of 15 years is ⬍1/1,000,000 after in utero exposure
to CTPA and 1/280 000 following a perfusion scan.
While CTPA is associated with a lower risk of radiation for the fetus, this must be offset by the relatively
high radiation dose (20 mGy) to the mother’s thorax
and in particular breast tissue. The delivery of 10 mGy
of radiation to a woman’s breast increases her lifetime risk of developing breast cancer. It has been estimated that the increased risk is 13.6% (background
risk 1/200), a figure that has been cited widely (4,
12). More recently, authorities have suggested that this
risk is an overestimate. Nevertheless, breast tissue is
especially sensitive to radiation exposure during pregnancy, and it therefore seems sensible to recommend
that lung perfusion scans should be considered the
investigation of first choice for young women, especially if there is a family history of breast cancer or
the patient has had a previous chest CT scan. Radiation exposure from pulmonary angiography is approximately 0.5 mSv to fetus, and 5 to 30 mSv to mother.

D-dimer testing in pregnancy
Outwith pregnancy, a normal plasma D-dimer level
has been shown to have excellent negative predictive value in patients with a low clinical probability
score for VTE. However levels increase physiologically
throughout pregnancy, becoming elevated at term and
in the post-natal period in most healthy pregnant

women. Furthermore, D-dimer levels are increased if
there is a concomitant problem such as pre-eclampsia,
preterm labour, and placental abruption. Thus the
probability of a negative result is lower and objective testing is more often required. For this reason
guidelines produced by the Royal College of Obstetricians and Gynaecologists in the United Kingdom,4
do not recommend that D-dimer levels are evaluated
in pregnant women with suspected VTE. In contrast,
the European Society of Cardiology13 recommend that
D-dimer levels should be measured, as a proportion of
patients will have a normal result and be able to avoid
unnecessary imaging. It should be noted, however, that
although the SimpliRED test has been reported to have
a negative predictive value of 100% in pregnancy, false
negative results have been reported.

Thrombophilia testing in acute VTE
pregnancy
Almost half of all women who have an episode of
VTE in pregnancy, will have an underlying heritable or acquired thrombophilia.14 The prevalence rates
for thrombophilias in European populations is shown
in Table 12.2 and the relative risk of each condition,
shown by metaanalysis, is shown in Table 7.2. Performing a thrombophilia screen in the acute stages of
thrombosis may give misleading results and is not routinely recommended. Levels of antithrombin, protein
C and protein S may fall, particularly if thrombus is
extensive. In addition, protein S levels fall in normal
pregnancy and an acquired activated protein C resistance is found with the APC sensitivity ratio test in
around 40% of pregnancies, due to the physiological
changes in the coagulation system. Clearly, genotyping for factor V Leiden and prothrombin G20210A will
not be affected by pregnancy or thrombus. Whilst the
results of a thrombophilia screen will not influence the
immediate management of acute VTE, they might provide information that can influence the duration and
intensity of anticoagulation, such as when antithrombin deficiency or antiphospholipid syndrome is identified.

Initial treatment of VTE in pregnancy
Before anticoagulant therapy is initiated, blood should
be taken for a full blood count and coagulation screen.
Urea, electrolytes and liver function tests should also

Section 3. Thromboembolism and anticoagulation

Table 7.2. Risk of pregnancy associated VTE in women with
underlying thrombophilia (14)

Thrombophilic defect

Relative Risk of
thrombosis

Factor V Leiden (heterozygote)

8.32

Factor V Leiden (homozygote)

34.4

Prothrombin 20210A (heterozygote)

6.8

Prothrombin 20210A (homozygote)

26.4

Protein C deficiency

4.76

Protein S deficiency

3.19

Antithrombin deficiency

4.69

be checked to exclude renal or hepatic dysfunction,
which are cautions for anticoagulant therapy.
The treatment of VTE in pregnancy is heparin.
Vitamin K antagonists are rarely employed in this
setting as they cross the placenta and are associated
with increased pregnancy loss, a specific embryopathy and other abnormalities in the first trimester, as
well as fetal haemorrhagic complications and central nervous system anomalies at any stage of pregnancy. Although for many years, unfractionated heparin (UFH) was the standard anticoagulant used during and outwith pregnancy, it has now largely been
replaced by low molecular weight heparin (LMWH).
Meta-analyses of randomised controlled trials (RCTs)
in non-pregnant patients indicate that LMWHs are
more effective and are associated with a lower risk
of haemorrhagic complications and lower mortality
than unfractionated heparin in the initial treatment
of DVT. A meta-analysis of RCTs has shown equivalent efficacy of LMWH to unfractionated heparin in
the initial treatment of PE. A systematic review of
LMWH in pregnancy has confirmed its efficacy and
safety in the management of acute thrombosis and
in the provision of thromboprophylaxis. Furthermore,
compared with UFH, LMWH is associated with a substantially lower risk of heparin-induced thrombocytopenia, haemorrhage and osteoporosis.1,2,4,15–17 Neither UFH nor LMWH cross the placenta and both are
safe for breast feeding.
Whilst several LMWH preparations are available,
most experience currently exists with enoxaparin, dalteparin and tinzaparin. In non-pregnant patients with
acute VTE, LMWH is usually administered in a once
daily dose. In view of recognised alterations in the
pharmacokinetics of dalteparin and enoxaparin during pregnancy, a twice daily dosage regimen is recom-

mended for these LMWHs in the treatment of VTE
in pregnancy, (enoxaparin 1mg/kg twice daily; dalteparin 100 units/kg twice daily). Preliminary biochemical data suggest that once daily administration
of tinzaparin (175 units/kg) may be appropriate in the
treatment of VTE in pregnancy. Whichever, preparation of LMWH is employed, the woman should be
taught to self-administer the drug by subcutaneous
injection, allowing further management on an outpatient basis until delivery.4
In the very early management of DVT, the leg
should be elevated and a graduated elastic compression stocking applied to reduce oedema. Once full anticoagulation has been commenced, the woman should
be encouraged to mobilize whilst wearing compression hosiery as this has been shown to reduce pain and
swelling in the affected leg. Studies in non-pregnant
patients have shown that early mobilization, with compression therapy, does not increase the likelihood of
developing PE and helps prevent the development of
post-thrombotic syndrome. For patients with persisting leg oedema after DVT, class II compression hosiery
is more effective than class I stockings. Where DVT
threatens leg viability through venous gangrene, the
leg should be elevated, anticoagulation given and consideration given to surgical embolectomy or thrombolytic therapy.4

Monitoring of LMWH therapy
Experience indicates that satisfactory anti-Xa levels
(peak anti-Xa activity, 3 hours post-injection, of 0.5–
1.2 u/ml) are obtained using a weight-based regimen
and monitoring of anti-Xa is not routinely required
in patients with VTE on therapeutic doses of LMWH,
particularly as there are concerns over the standardization and accuracy of anti-Xa monitoring. There may
be a case for monitoring levels at extremes of body
weight (⬍50 kg and ≥90 kg), and women with other
complicating factors including renal disease and recurrent VTE.4,16
Guideline documents from North America2 recommend that routine platelet count monitoring is
not required in obstetric patients who have received
only LMWH as heparin induced thrombocytopenic
thrombosis is not a feature in pregnancies managed
exclusively with LMWH. If unfractionated heparin
is employed, or if the obstetric patient is receiving
LMWH after first receiving unfractionated heparin, or
if she has received unfractionated heparin in the past,

Chapter 7. Acute venous thromboembolism

the platelet count should ideally be monitored every 2–
3 days from day 4 to day 14, or until heparin is stopped,
whichever occurs first.

Maintenance treatment of VTE
Women with ante-natal VTE can be managed with
subcutaneous LMWH for the remainder of the pregnancy using LMWH administered subcutaneously. It
is our practice to continue with the initial dose regimen throughout pregnancy despite the pregnancyassociated weight gain, (since LMWH does not cross
the placenta and therefore the weight of the fetoplacental unit is not relevant). If LMWH therapy
requires monitoring, for example in extremes of body
weight or renal impairment, the aim is to achieve a
peak anti-Xa activity, three hours post-injection of
0.5 –1.2 u/ml.
It is not yet established whether the initial dose of
LMWH can be reduced to an intermediate dose after
an initial period of several weeks of therapeutic anticoagulation, although this practice has been successfully employed in some centres. Outwith pregnancy in
patients with underlying malignancy, a reduction in
dose has been shown to be safe after 4 weeks of therapeutic anticoagulation. Although there have been no
studies directly comparing these two types of dosing
strategies in pregnant women, this type of modified
dosing regimen may be useful in pregnant women at
increased risk of bleeding or osteoporosis.18

Management at the time of delivery
For women on therapeutic anticoagulation, a planned
delivery, either through induction of labour or elective caesarean section, allows accurate timing of events
and minimizes the risk of the woman having to deliver
on full anticoagulation. The dose of LMWH should
be reduced to a once daily thromboprophylactic dose
on the day before induction of labour or caesarean
section. When a woman presents whilst on a therapeutic, twice daily regimen of LMWH, regional techniques should not usually be employed for at least 24
hours after the last dose of LMWH. LMWH should
not be given for at least four hours after the epidural catheter has been removed and the cannula should
not be removed within 12 hours of the most recent
injection.4
For delivery by elective caesarean section, the treatment doses of LMWH should be omitted for 24
hours prior to surgery. A thromboprophylactic dose of

LMWH (enoxaparin 40mg; dalteparin 5000iu; tinzaparin 75 iu/kg) should be given by three hours postoperatively (⬎4 hours after removal of the epidural catheter, if appropriate), and the treatment dose
recommenced that evening. There is an increased risk
of wound haematoma following caesarean section with
both unfractionated heparin and LMWH of around
2%. For this reason, wound drains should be considered at caesarean section, and the skin incision should
ideally be closed with staples or interrupted sutures to
allow easy drainage of any haematoma.
If the thrombosis occurred in the last week of pregnancy, consideration should be given to the use of
unfractionated heparin (since it can be relatively easily reversed using protamine sulfate and has a short
duration of action). If spontaneous labour occurs in
women receiving therapeutic doses of subcutaneous
unfractionated heparin, careful monitoring of the anticoagulant effect by measuring the activated partial
thromboplastin time (APTT) is required. Subcutaneous unfractionated heparin should be discontinued
12 hours before and intravenous unfractionated heparin stopped 6 hours before induction of labour or
regional anaesthesia. It should be noted however that
the aPTT is less reliable in pregnancy due to increased
levels of FVIII and heparin binding proteins, which
can lead to an apparent heparin resistance.4,18

Postpartum anticoagulation and
duration of anticoagulation therapy
In the United Kingdom, it is recommended that in
non-pregnant patients, anticoagulant therapy should
be continued for 6 weeks for calf vein thrombosis
and 3 months for proximal DVT or PE when VTE
has occurred in relation to a temporary risk factor,
and 6 months for a first episode of idiopathic VTE.4
The presence of ongoing risk factors and the safety of
LWMH have led authorities to propose that anticoagulant therapy should be continued for the duration of
the pregnancy and until at least six weeks postpartum
and to allow a total duration of treatment of at least
3 months. Both heparin and warfarin are satisfactory
for use postpartum – in our experience most women
prefer to use LMWH (which can be used with once
daily dosing postpartum) because they have become
accustomed to its administration, and they appreciate
the convenience of not having to attend clinics to have
their INR checked. Before discontinuing treatment
the ongoing risk of thrombosis should be assessed

Section 3. Thromboembolism and anticoagulation

Figure 7.1. Management of women
with clinically suspected massive PTE

•
•

Emergency call to multi-disciplinary resuscitation team

•

Heparinise with intravenous unfractionated heparin

•

IV fluids and inotropic support

•

Inform on-call obstetric team immediately for consideration of early delivery

Oxygen administered

Transfer to intensive therapy area

Diagnosis made by emergency CTPA or echocardiogram
If the patient becomes periarrest at any stage, consider thrombolysis
without imaging.

Negative investigations:
Search for other diagnosis

CTPA confirms significant PE
or
Cardiac echo confirms RV dilatation/dysfunction

If persistent hypotension (SBP < 90mmHg), consider:-

•

Thrombolysis

If thrombolysis is contraindicated, consider

•
•

Percutaneous catheter fragmentation
Surgical embolectomy

including a review of personal and family history of
VTE and any thrombophilia screen results. Thrombophilia screening should be discussed and arranged
if required.
Neither heparin (unfractionated or LMWH) nor
warfarin is contraindicated in breastfeeding. There are
little published data on whether LMWHs are secreted
in breast milk, although extensive experience of enoxaparin in the puerperium reports no problems during
breastfeeding and other heparins are known not to
cross the breast. Furthermore, neither unfractionated
heparin nor LMWH are orally active and no effect
would therefore be anticipated in the fetus.
The post-thrombotic syndrome is a common complication following DVT.18 It is found in over 60%
of cases followed up over a median of 4.5 years. It is
characterized by chronic persistent leg swelling, pain,
a feeling of heaviness, dependant cyanosis, telangiec-

tasis, chronic pigmentation, eczema, associated varicose veins and in some cases lipodermatosclerosis,
and chronic ulceration. Symptoms are made worse
by standing or walking and improve with rest and
recumbancy. The syndrome is more common where
there is a recurrent DVT, with obesity and where
there has been inadequate anticoagulation. It is recommended that graduated elastic compression stockings (class II) should be worn on the affected leg for
two years after the acute event to reduce the risk of
post-thrombotic syndrome. This recommendation is
based upon studies in non-pregnant patients where
such therapy reduces the incidence of post thrombotic syndrome from 23% to 11% over 2 years. Graduated elastic compression stockings will improve the
microcirculation by assisting the calf muscle pump,
reducing swelling and reflux, and reducing venous
hypertension.4

Chapter 7. Acute venous thromboembolism

Management of massive,
life-threatening PE
Massive, life-threatening PE may be defined as embolus associated with haemodynamic compromise (a systolic blood pressure ⬍90 mmHg or a drop in systolic blood pressure of ≥ 40mmHg from baseline
for a period ⬎15 minutes), not otherwise explained
by hypovolaemia, sepsis or new arrhythmia. This is
an obstetric and medical emergency and hospitals
should have in place guidelines for the management
of non-haemorrhagic obstetric shock (see figure 7.1).
The collapsed, shocked pregnant woman needs to be
assessed by a multi-disciplinary resuscitation team of
experienced clinicians including senior obstetricians,
physicians and radiologists, who should decide on
an individual basis whether a woman receives intravenous unfractionated heparin, thrombolytic therapy
or thoracotomy and surgical embolectomy.
Oxygen should be administered and the circulation supported using intravenous fluids and inotropic
agents if required. Intravenous unfractionated heparin
is the traditional method of heparin administration
in acute VTE and remains the preferred treatment in
massive PE because of its rapid effect and extensive

experience of its use in this situation. The diagnosis
should be established using either portable echocardiogram or CTPA performed within 1 hour of presentation.
In massive life-threatening PE with haemodynamic compromise there is a case for considering
thrombolytic therapy as anticoagulant therapy will
not reduce the obstruction of the pulmonary circulation. After thrombolytic therapy has been given
an infusion of unfractionated heparin can be given.
There are now a large number of published case
reports on the use of thrombolytic therapy in pregnancy, streptokinase being the agent most frequently
employed. Streptokinase, and probably other thrombolytic agents, do not cross the placenta. No maternal deaths associated with thrombolytic therapy have
been reported, and the maternal bleeding complication rate is approximately 6%, which is consistent with
that in non-pregnant patients, receiving thrombolytic
therapy. Most bleeding events occur around catheter
and puncture sites, and, in pregnant women, from the
genital tract. If the patient is not suitable for thrombolysis or moribund, a discussion with the cardiothoracic
surgeons with a view to urgent thoracotomy should be
had.4

Section 3. Thromboembolism and anticoagulation

References
1.

Bates SM, Ginsberg JS. How we manage venous
thromboembolism during pregnancy. Blood 2002; 100:
3470–8.

Bates SM, Greer IA, Pabinger I et al. Venous
thromboembolism, thrombophilia, antithrombotic
therapy, and pregnancy: American College of Chest
Physicians Evidence-Based Clinical Practice
Guidelines (8th Edition). Chest 2008; 133: 844S–
886S.

Royal College of Obstetricians and Gynaecologists.
Thromboembolic disease in pregnancy and the
puerperium: acute management. RCOG Green Top
Guideline 2007, No 28.

British Thoracic Society Standards of Care Committee
Pulmonary Embolism Guideline Development Group.
British Thoracic Society guidelines for the
management of suspected acute pulmonary embolism.
Thorax 2003; 58: 470–84.

Lewis, G. ed. The Confidential Enquiry into Maternal
and Child Health. Saving mothers’ lives: reviewing
maternal deaths to make motherhood safer 2003–2005.
The seventh report on Confidential Enquiries into
Maternal Deaths in the United Kingdom. London:
CEMACH, 2007.

Scarsbrook AF, Evans AL, Owen AR, Gleeson FV.
Diagnosis of suspected venous thromboembolic
disease in pregnancy. Clinical Radiology 2006; 61:
1–12.
Royal College of Obstetricians and Gynaecologists.
Reducing the risk of thrombosis and embolism during
pregnancy and the puerperium. RCOG Green Top
Guideline 2009, No 37.

Jacobsen AF, Skjeldestad FE, Sandset PM. Ante- and
postnatal risk factors of venous thrombosis: a
hospital-based case-control study. Journal of
Thrombosis and Haemostasis 2008; 6: 905–12.

Lindqvist P, Dahlbäck B, Marŝál K. Thrombotic risk
during pregnancy: a population study. Obstetrics and
Gynecology 1999; 94: 595–9.

10. James AH, Jamison MG, Brancazio LR, Myers ER.
Venous thromboembolism during pregnancy and the
postpartum period: incidence, risk factors, and
mortality. American Journal of Obstetrics and
Gynecology 2006; 194: 1311–5.
11. Knight M; on behalf of UKOSS. Antenatal pulmonary
embolism: risk factors, management and outcomes.
BJOG: An International Journal of Obstetrics &
Gynaecology 2008; 115: 453–461.
12. Cook JV, Kyriou J. Radiation from CT and perfusion
scanning in pregnancy. British Medical Journal 2005;
331: 350.
13. The Task Force for the Diagnosis and Management of
Acute Pulmonary Embolism of the European Society
of Cardiology. Guidelines on the diagnosis and
management of acute pulmonary embolism. European
Heart Journal 2008; 29: 2276–2315.
14. Robertson, L. and Wu, O. and Langhorne, P. and
Twaddle, S. and Clark, P. and Lowe, G.D.O. and
Walker, I.D. and Greaves, M. and Brenkel, I. and
Regan, L. and Greer, I.A. Thrombophilia in pregnancy:
a systematic review. British Journal of Haematology
2006; 132: 171–196.
15. Greer IA, Nelson-Piercy C. Low-molecular-weight
heparins for thromboprophylaxis and treatment of
venous thromboembolism in pregnancy: a systematic
review of safety and efficacy. Blood 2005; 106: 401–7.
16. Thomson AJ, Walker ID, Greer IA. Low molecular
weight heparin for the immediate management of
thromboembolic disease in pregnancy. The Lancet
1998; 352: 1904.
17. James AH. Prevention and management of venous
thromboembolism in pregnancy. American Journal of
Medicine 2007; 120: S26–S34.
18. Greer I, Hunt BJ. Low molecular weight heparin in
pregnancy: current issues. British Journal of
Haematology 2005; 128: 593–60.
19. McColl D, Ellison J, Greer IA. et al. Prevalence of the
post thrombotic syndrome in young women with
previous venous thromboembolism. British Journal of
Haematology 2000; 108: 272–274.

Section 3
Chapter

Thromboembolism and anticoagulation

Thromboprophylaxis
Sarah Germain and Catherine Nelson-Piercy

Introduction and epidemiology
Venous thromboembolism (VTE) remains the leading
direct cause of maternal death in the UK. In the latest CEMACH report “Saving Mothers Lives: Reviewing maternal deaths to make motherhood safer, 2003–
5”1 there were 33 deaths from VTE and eight from
cerebral vein thrombosis. Although the absolute numbers of both fatal and non-fatal VTE in pregnancy
and the puerperium are small, with an overall incidence of approximately 2 per 1000 births,2,3 many
are preventable. Successive enquiry reports have highlighted the need to identify risk factors for VTE early
in pregnancy and ensure adequate thromboprophylaxis is employed. The Royal College of Obstetricians
and Gynaecologists have published guidelines regarding thromboprophylaxis covering both the ante-natal
and post-natal periods4,5 and these have recently been
updated.6 Despite this, a recent case control study
of 143 cases of ante-natal pulmonary embolism (PE)
in the UK, by the UK Obstetric Surveillance System
(UKOSS), demonstrated that, although nine of the
women should have received ante-natal thromboprophylaxis according to national guidelines, only 33%
had actually done so, and 50% of the six women who
had a PE while on prophylactic low molecular weight
heparin (LMWH) were receiving lower than recommended doses.7
Traditionally, VTE has been considered a complication of late pregnancy and Post-Cesarean section.
The Confidential Enquiries into Maternal Deaths have
shown that this is by no means the case, with twothirds of ante-natal fatal pulmonary VTE in 2003–
2005 occurring in the first trimester, and just over
half of the post-natal deaths from pulmonary VTE
after vaginal delivery.1 A study from the USA found
that 44% of deep vein thromboses DVTs in pregnancy
occurred in the first trimester,8 and a more recent

Spanish study similarly found that 40% of ante-natal
VTE were in the first trimester.9 These all emphasize
the need for risk assessment pre-pregnancy and institution of prophylaxis if appropriate in early pregnancy.
Although numerically most VTE occurs ante-natally,
the highest risk per day is during the immediate postpartum period, and this is demonstrated by a cohort
study from the USA, which showed that the annual
incidence of VTE was five times higher among postpartum compared to pregnant women.3
As clinicians we therefore ideally need to identify women at risk prior to conception, or at least
early in pregnancy, then establish their level of risk,
and finally initiate an appropriate thromboprophylaxis
regimen, which extends into the puerperium. We also
need to be aware that a woman who starts pregnancy
as “low risk” of VTE may develop or acquire risk
factors as pregnancy progresses, and thromboprophylaxis may need to be introduced at that point or after
delivery.

Pathogenesis and risk factor
assessment
Pregnancy itself puts all women at higher risk of
VTE, with a four to ten-fold increase compared to
an age-matched non-pregnant female population.3,5
This is primarily related to the pro-coagulant changes
that occur during pregnancy to promote hemostasis
post-delivery,2 and are evident from early in the first
trimester. Other components of Virchow’s triad are
also present, namely increased venous stasis and vascular trauma, the latter particularly around the time of
delivery. Superimposed on this background risk, are a
range of additional risk factors which may either predate the pregnancy or develop during the pregnancy or
puerperium, and can be persistent or transient.

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

Section 3. Thromboembolism and anticoagulation

Table 8.1 Additional risk factors for VTE in pregnancy and the
∗
puerperium

Table 8.2 Quoted adjusted odds ratios for individual risk
factors

Pre-existing
Age ⬎35 years
Obesity (BMI ⬎ 30 kg/m2 , either pre-pregnancy or in early
pregnancy)
Parity ≥3
Gross varicose veins
Paraplegia
Sickle cell disease
Smoking
ART
Inflammatory disorders (e.g. inflammatory bowel disease)
Some medical disorders (e.g. nephrotic syndrome, certain
cardiac diseases)
Myeloproliferative disorders (e.g. essential thrombocythemia,
polycythemia vera)

Risk factor for VTE

aOR

95% CI

Previous VTE 11

24.8

17.1–36

Age ⬎ 35 14

1.3

1.0–1.7

BMI ⬎ 30 12

5.3
4.4

2.1–13.5
3.4–5.7

Smoking 13

2.7

1.5–4.9

Parity ≥3 14

2.4

1.8–3.1

Medical conditions11

2.0–8.7

New onset or transient
Ovarian hyperstimulation syndrome
Hyperemesis
Dehydration
Severe infection (e.g. pyelonephritis)
Long-distance travel
Immobility (⬎ 4 days bed rest)
Pre-eclampsia
Excessive blood loss/blood transfusion
Surgical procedure in pregnancy or puerperium (e.g. ERPC,
postpartum sterilization)
Specific to postpartum VTE only
Cesarean section (especially in labor)
Prolonged labor
Midcavity instrumental delivery
Immobility after delivery

Sickle cell disease, SLE, heart
disease, anemia, infection,

2.0–3.2

Hyperemesis

2.51

Immobility 13

7.7 (an)
10.8 (pn)

3.2–19
4–28.8

Pre–eclampsia13

3.1

1.8–5.3

+Fetal growth restriction

5.8

2.1–16

ART 13

4.3

2.0–9.4

Twins13

2.6

1.1–6.2

APH 13

2.3

1.8–2.8

PPH 13

4.1

2.3–7.3

Cesarean section 14

3.6

3.0–4.3

Varicose veins

2.4

1.04–5.4

Transfusion11

7.6

6.2–9.4

∗

Adapted from ref. 6
VTE = venous thromboembolism; BMI = body mass index;
ERPC = evacuation of retained products of conception; ART =
Assisted Reproductive therapy.6

100

The major additional risk factors are a previous
VTE and/or a documented thrombophilia.2 A history
of thrombosis increases the risk of pregnancy-related
VTE to 2%–12%.2 De Stefano et al. found recurrence
rates following a single previous DVT or PE of 5.8% for
ante-natal and 8.3% for post-natal VTE.10
Thrombophilias may either be heritable
(antithrombin deficiency, Protein C deficiency,
Protein S deficiency, Factor V Leiden, or Prothrombin gene variant) or acquired (antiphospholipid
syndrome, including lupus anticoagulant and anticardiolipin antibodies). Up to 50% of women who develop
VTE during pregnancy or the postpartum period have
an underlying thrombophilia.11 The relative risk of
VTE in pregnancy varies depending on the thrombophilia, but can be as high as ten fold with antithrombin deficiency. The most important determinant of

VTE risk in a pregnant woman with a thrombophilia
is a personal or family history of VTE.
Additional risk factors are detailed in Table 8.1,
adapted from the Royal College of Obstetricians Green
Top Guideline on Reducing the risk of Thrombosis and
Embolism during pregnancy and the puerperium.6
All are accepted VTE risk factors, but the degree of
increased VTE risk associated with them varies, as
indicated in Table 8.2. Of particular importance are
obesity and increasing maternal age.
The growing problem of maternal obesity in association with VTE is highlighted in the findings from
both the most recent Confidential Enquiry into Maternal Deaths (2003–5)1 and the UKOSS study of antenatal PE,7 that have already been discussed. The latter demonstrated that one of the main risk factors was
a BMI ⬎30 with an adjusted odds ratio (OR) of 2.65
(95% confidence interval (CI) 1.09–6.45). Similarly, a
recent case control study from Denmark of 129 cases
of VTE in pregnancy or the puerperium demonstrated
an adjusted OR of 5.3 (95% CI 2.1–13.5) for obesity
(BMI ⬎30), with a higher risk of PE (adjusted OR 14.9,

Chapter 8. Thromboprophylaxis

95% CI 3.0–74.8) than of DVT (adjusted OR 4.4, 95%
CI 1.6–11.9)12 The only other risk factors that reached
statistical significance in these two studies were multiparity (adjusted OR 4.03, 95% CI 1.60–9.84)7 and
current smoking (adjusted OR 2.7, 95% CI 1.5–4.9),12
although Knight et al. highlight the difficulty of obtaining sufficient power to show other associations, even in
a large national study such as theirs, as ante-natal PE is
still a relatively rare condition.7

Management
Management strategy
The management strategy for thromboprophylaxis
during pregnancy and the postpartum period is
detailed in Table 8.3. All women should have an assessment of risk factors for VTE, in early pregnancy, or
ideally pre-pregnancy. It is particularly important to
identify those with a previous VTE and/or known
thrombophilia. If there is a past history of VTE, then
the details of presentation, means of diagnosis, and
drug treatment and length of course should be determined. If deemed appropriate, women with a previous VTE should be screened for both heritable and
acquired thrombophilia prior to pregnancy, as interpretation of some of the tests (especially protein S) is
unreliable in pregnancy. The other risk factors detailed
in Table 8.1 should also be considered, as well as any
family history of VTE in a first-degree relative.
Following discussion with the woman, a written
plan can then be made for ante-natal thromboprophylaxis, if required, and prescribed if the woman is
already pregnant. Women should be taught how to
self-administer subcutaneous LMWH. If the woman is
not yet pregnant, the prescription can still be given, so
LMWH can be started as soon as a pregnancy test is
positive. This is particularly for women in the very high
and high risk groups, as the pregnancy-related increase
in VTE risk starts from the beginning of the first
trimester.8,9 The recommendations should be detailed
in a letter copied to both the GP and patient, so prophylaxis can be started without the woman needing to
come back to the hospital clinic first.
It is important to remember that the VTE risk
may change for a particular woman as pregnancy progresses, for example, if she develops pre-eclampsia,
and the risk factor assessment should be repeated if
there is any change in circumstances, and LMWH initiated as appropriate. It may be that the additional

risk factor is only temporary, for example hyperemesis
gravidarum, and the original regimen can be returned
to once the condition or situation has resolved.
Issues regarding anticoagulant use peri-delivery
should be discussed, including epidural timing, and
again documented in the ante-natal notes (discussed
further in Chapter 10). Assessment by an obstetric
anesthetist in the ante-natal period if the woman
is using ante-natal LMWH, especially if at high
prophylactic or therapeutic dose, should be part of
this process.
A clear plan for postpartum prophylaxis should
also be documented in the obstetric notes. The level
of risk will need to be reassessed post-natally, as it
may be increased depending on the mode of delivery
and any associated complications. Women on longterm warfarin are usually managed with high-dose
prophylactic LMWH for the first week post-natally
and then converted back to warfarin. Follow-up
should be arranged for those women with a previous
VTE who have not had a thrombophilia screen, once
they have completed the 6-week post-natal anticoagulant course, so they can be investigated for an underlying thrombophilia.

Non-pharmacological and pharmacological
measures used
Thromboprophylaxis
involves
both
nonpharmacological and pharmacological measures,
and the various modalities and drugs that can be used
are discussed below.

Non-pharmacological
Non-pharmacological measures include appropriate
hydration, early mobilization after surgery or delivery,
graduated compression stockings (TEDS), and pneumatic compression boots. The aim is to improve blood
flow and decrease stasis in the femoral and popliteal
vessels.
No randomized controlled trials have been carried out in pregnancy to study the efficacy of TEDS
or pneumatic compression for thromboprophylaxis,
but the latter have been shown to be cost-effective
when used at Cesarean section,16 and trials outside
of pregnancy have demonstrated significant reduction in incidence of VTE when used peri-operatively.
These measures are also not associated with the potential hemorrhagic side effects of the pharmacological

101

Section 3. Thromboembolism and anticoagulation

Table 8.3 Management strategy for thromboprophylaxis in pregnancy and the postpartum period
Pre-pregnancy

Assess women with prior history of VTE and/or known thrombophilia
Consider VTE risk factors when giving pre-pregnancy counseling to women with other medical problems
Make plan for future pregnancy regarding thromboprophylaxis and document in letter to GP and patient
Consider giving high risk women prescription for LMWH to start with positive pregnancy test

Ante-natal

Assess all women for VTE risk factors at booking
Determine need for ante-natal thromboprophylaxis and prescribe if required
Liaise with obstetric hematologist and/or obstetric physician
Discuss implications of anticoagulants peri-delivery
Arrange assessment by obstetric anesthetist if using ante-natal LMWH
Document plan for peri-delivery and postpartum periods in obstetric notes
Reassess woman’s risk status, need for LMWH, and LMWH dose if events change in pregnancy (e.g. acquire additional
risk factors detailed in Table 8.1)

Peri-delivery

Liaise with obstetric anesthetist regarding timing of regional analgesia and anesthesia.

Postpartum

Reassess woman’s risk status depending on mode of delivery and any obstetric complications
Prescribe postpartum prophylaxis if required
Arrange follow-up for thrombophilia testing once off anticoagulants if previous VTE and not already tested

VTE = venous thromboembolism; LMWH = low molecular weight heparin.

agents discussed below. They can be of particular use
in patients requiring Cesarean section or having prolonged bedrest.

Pharmacological
Aspirin, heparin, and warfarin are the main pharmacological agents to be used in thromboprophylaxis,
and each will be discussed in turn. Although they differ in their ability to cross the placenta, all are safe to
use in breast-feeding mothers.17

Aspirin
Aspirin inhibits the enzyme cyclo-oxygenase in
platelets, thus reducing thromboxane production and
platelet aggregation. Aspirin is known to be effective
in reducing the risk of VTE in both surgical and
medical patients but no randomized controlled trials
have been carried out looking at the use of aspirin as
thromboprophylaxis in pregnancy. However studies
of its use for a range of other indications in pregnancy
have shown it to be safe at low dose (75 mg).17 It may
therefore be reasonable to consider low dose aspirin
for women who have an increased risk of VTE, but not
high enough to warrant LMWH, although its use for
this indication is controversial, and recent guidelines
do not recommend it.6,17

Heparin

102

Neither unfractionated (UH) nor low molecular
weight (LMWH) heparin cross the placenta, so there
are no adverse effects on the fetus.

LMWHs are now the anticoagulants of choice in
the UK for prophylaxis and treatment of VTE in pregnancy, for the vast majority of cases. LMWH have
a longer half-life and increased bioavailability, which
allows once daily dosing for prophylaxis. UH is usually
only used for thromboprophylaxis if there is an allergy
to LMWH or in renal failure.
Hospitals differ in the LMWH used, and Table 8.4
details the recommended, prophylactic and therapeutic doses for the different LMWHs available. Doses
need to be adjusted according to maternal weight in
early pregnancy.18
UH is a heterogenous mixture of high molecular weight molecules (3 000–30 000 daltons), whereas
LMWHs are a derivative of UH with molecular weights
of 4 000–5 000 daltons. This affords LMWH a number of advantages over UH, including predictable and
reliable pharmacokinetics, a higher ratio of anti-Xa
to anti-IIa activity, providing good antithrombotic
effect with possibly a lower risk of bleeding. They also
have less of an effect on platelet aggregation, function and activation, and bind platelet factor 4 less well,
hence reducing the risk of both early and late heparininduced thrombocytopenia (HIT).19
Heparin-induced osteoporosis is an important
risk, especially if heparin is required throughout the
ante-natal period, but the risk is much lower with
LMWH (0.04%) than UH (2%).19 Also, even if there
is loss in bone density during pregnancy (up to
5%), this usually improves on stopping breast feeding postpartum. LMWHs in prophylactic doses have
not been associated with a reduction in bone density

Chapter 8. Thromboprophylaxis

∗

Table 8.4 Ante-natal prophylactic and therapeutic doses of different low molecular weight heparins

Dose
(based on early pregnancy
maternal weight)

Enoxaparin∗∗

Dalteparin

Tinzaparin

Standard prophylactic
⬍50 kg
50–90 kg
91–130 kg
131–170 kg

20 mg od
40 mg od
60 mg od or 40 mg bd
80 mg od or 40 mg bd

2500 units od
5000 units od
7500 units od or 5000 bd
10,000 units od or 5000 units bd

3500 units od
4500 units od
7000 units od

High prophylactic

40 mg bd

5000 units bd

4500 units bd

Treatment/therapeutic
⬍50 kg
50–90 kg
⬎90 kg

1 mg/kg bd
40 mg bd
60–80 mg bd
100 mg bd

100 units/kg bd
5000 units bd
6000–8000 units bd
10 000 units bd

175 units/kg od

∗ Adapted from4,14 ; ∗∗ 100 units/mg
od = once daily; bd = twice daily.

over and above what would be expected in normal
pregnancy.20
Of women, 1%–2% may develop a local allergic skin rash with LMWH. The first-line pragmatic
approach is to substitute with an alternative LMWH,
but there may be cross-reactivity, as also occurs with
UH.19 If this is the case, or if other complications such
as HIT develop, then consideration of alternative heparinoids such as danaparoid or the synthetic pentasaccharide fondaparinux is appropriate. However, HIT
has never been described with LMWH in a pregnant
woman.2,17

Warfarin
Warfarin crosses the placenta and is teratogenic.17
The characteristic “warfarin embryopathy” includes
chrondrodysplasia punctata, nasal hypoplasia, and
short proximal limbs. The period of greatest risk is
between 6 and 12 weeks. Warfarin is also associated
with an increased rate of miscarriage and stillbirth,
and in the third trimester a significant risk of fetal
intracerebral hemorrhage and maternal retroplacental
bleeding, especially after 36 weeks’ gestation. Use in
the second trimester has been linked to microcephaly
and neurological abnormalities, which may be due to
over-anticoagulation of the fetus.
Because of these effects, warfarin is not used routinely during pregnancy, and most women requiring thromboprophylaxis or treatment can be managed
with LMWH or unfractionated heparin. There are
some women though, who that require continued full
anticoagulation with warfarin throughout pregnancy

until planned delivery. The only non-controversial
indication for use of warfarin in pregnancy is for
women with a metal prosthetic heart valves (particularly of the older type and those in the mitral position), whose thrombotic risk is extremely high, and in
whom valve thrombosis carries a high mortality rate
(see Chapter 9).
In other high-risk women the decision is more difficult, and the risk/benefit balance for the particular
individual has to be considered. This would include
those with a known thrombophilia and VTE or cerebral arterial thrombosis despite full dose LMWH. A
compromise option here is to use LMWH for the highest risk periods for warfarin side effects, namely, the
first trimester and after 36 weeks’ gestation, and convert back to warfarin for the second and early third
trimesters. This obviously requires very close supervision and thorough discussion with the woman.

Recommendations
Although VTE in pregnancy and postpartum is a
major cause of maternal morbidity and mortality, the
absolute risk for most women is low and there are few
randomized controlled trials of thromboprophylaxis
in the pregnancy-related setting on which to base recommendations. A Cochrane review published in 2002,
looking at trials of thromboprophylaxis during pregnancy and the early puerperium, concluded there was
insufficient evidence on which to base recommendations, as the number of trials and sample sizes were too
small.21 A number of professional bodies have drawn

103

Section 3. Thromboembolism and anticoagulation

Table 8.5 Ante-natal VTE risk factor assessment and thromboprophylaxis management

Risk level

Risk factors

Recommended thromboprophylaxis

Very high

Previous VTE on long-term warfarin +/– thrombophilia
antithrombin deficiency or APS

High prophylactic or therapeutic dose LMWH
Requires specialist management by experts in hemostasis
and pregnancy

High

Previous recurrent VTE
Previous VTE:
b plus thrombophilia
b plus family history of VTE
b plus other risk factor(s)∗
b on COCP or during pregnancy
b at unusual site
Asymptomatic thrombophilia:
b Antithrombin deficiency
b Combined defects
b Homozygous factor V Leiden
b Homozygous prothrombin gene defect
b Compound heterozygote
Three or more other risk factors∗
Three or more risk factors plus admission to hospital

Prophylactic dose LMWH

Intermediate

Previous single VTE without family history,
thrombophilia or other risk factor(s)∗
Other asymptomatic thrombophilias including APS
(not covered above)

Close surveillance
Advise to keep mobile and avoid dehydration consider
TEDS

Lower

less than 3 other risk factors∗

Advise to keep mobile and avoid dehydration
Consider TEDS

∗ See Table 8.1
VTE = venous thromboembolism; COCP = combined oral contraceptive pill; LMWH = low molecular weight heparin;
APS = anti-phospholipid syndrome; TEDS = graduated compression stockings.

up guidelines,4–6,17 many recommendations are based
on low grade evidence. However, recent data show
that risk assessment and allocation of thromboprophylaxis according to such guidelines is efficacious and cost effective, with few clinically significant adverse events.22,23

Ante-natal

104

Recommendations for ante-natal assessment and
management of VTE risk are given in Table 8.5. It
is important to remember that there are probably
many heritable thrombophilias as yet undiscovered
and therefore unable to be confirmed with in vitro
testing. So, if the history is suspicious, for example,
previous VTE in an unusual site or previous recurrent
VTE especially associated with as family history of
VTE, it is sensible to treat these women as high risk,
even without an identifiable thrombophilia, and give
both ante-natal and post-natal LMWH prophylaxis.
In comparison, women with a single previous VTE
related to a temporary but non-estrogen related
(pregnancy or the combined oral contraceptive pill)

risk factor, who have no identifiable thrombophilia
or additional current risk factor, do not require antenatal LMWH.24 A recent study from Italy of 88 women
who became pregnant after a single previous episode
of VTE, with no ante-natal thromboprophylaxis,
demonstrated no recurrence of VTE in pregnancy if
the initial VTE was related to transient risk factors
other than pregnancy or oral contraceptive use.10 This
contrasted with a 7.5% recurrence rate if the first VTE
was unprovoked, or estrogen related.
Women who are on long-term warfarin outside
pregnancy because of previous VTE or stroke associated with APS are at very high risk of recurrence, and
should be managed with high prophylactic or sometimes full therapeutic LMWH ante-natally, depending
on the clinical history, under the care of an expert in
hemostasis and pregnancy.
Even women without a previous VTE or identified
thrombophilia may require ante-natal thromboprophylaxis due to other risk factors. Those with at least
three of the risk factors detailed in Table 8.1 should
be considered for LMWH ante-natally and those
with two risk factors require LMWH during hospital

Chapter 8. Thromboprophylaxis

Table 8.6 Post-natal VTE risk factor assessment and thromboprophylaxis management

Risk level

Risk factors

Recommended prophylaxis

Very high

Previous VTE on long-term warfarin +/–
thrombophilia antithrombin deficiency or APS

LMWH until re-established on warfarin

High

Any other previous VTE
Asymptomatic thrombophilia:
b Antithrombin deficiency
b Combined defects
b Homozygous factor V Leiden
b Homozygous prothrombin gene defect
b Compound heterozygote
Extended major pelvic or abdominal surgery
(e.g. CS hysterectomy)
Paralysis of lower limbs

6 weeks prophylactic dose LMWH

Intermediate

Emergency CS in labor
Any CS plus any other risk factor
Three or more other risk factors∗
Asymptomatic thrombophilia:
b Heterozygous factor V Leiden
b Heterozygous prothrombin gene defect

7 days prophylactic dose LMWH∗∗
Consider extending if other risk factors or
positive family history

Lower

CS not in labor no other risk factors
Less than 3 other risk factors∗

Early mobilization and avoidance of dehydration
Consider TEDS

∗ See Table 8.1; ∗∗ See text – there is minimal evidence to determine optimum duration of postpartum thromboprophylaxis.
VTE = venous thromboembolism; LMWH = low molecular weight heparin; CS = Cesarean section; TEDS = graduated compression
stockings.

admissions. This stresses the importance of repeated
risk assessment as pregnancy progresses. It also highlights the need to assess all pregnant women for their
thromboembolic risk, not just women under the care
of hematologists with previous VTE and/or thrombophilia.
Interestingly, the UKOSS7 study demonstrated that
approximately one-third of women with an ante-natal
PE had no classical risk factors for VTE disease (apart
from pregnancy), and only 9 of the 99 who did would
have been eligible for thromboprophylaxis under the
2004 RCOG guidelines.5 Knight et al. discuss inclusion of women with at least two risk factors being eligible, but also acknowledge that this requires an estimated 9% of maternities to receive LMWH, or 5.5% if
just multiparity was added as a risk factor. They suggest
further studies are required to assess the cost benefit of
this inclusion.7

Post-natal
Table 8.6 details recommendations for post-natal
assessment and management of VTE risk. The most
recent RCOG guidelines supersede an order guideline
regarding women who have had a Cesarean section.4,6
There is significant variation between units as to how
aggressively these are applied, with some prescribing

LMWH to all who have undergone Cesarean section,
while others just for emergency Cesarean section in
labor.
It is important to remember that thromboprophylaxis should not be limited to those who have delivered
by Cesarean section, as women die from VTE even
after a normal vaginal delivery. Women with at least
two persisting risk factors from Table 8.1 should be
considered for postpartum LMWH.
The immediate postpartum period is the time of
highest risk for VTE, and post-natal LMWH should be
continued for at least 7 days. Recently published data,
assessing changes in thromboelastography parameters
in the postpartum period suggest the risk is high for 7
days.25 There is also emerging evidence that it may take
several weeks for the hypercoagulable state of pregnancy to return to non-pregnant levels,24 therefore for
high risk patients, for example, those with a previous
VTE or thrombophilia, it is standard practice to continue LMWH for six weeks.

Dilemmas – current research and
future direction

r Identification of other thrombophilias
r Cost–benefit analysis of extending ante-natal
thromboprophylaxis

105

Section 3. Thromboembolism and anticoagulation

r Investigation of the optimal duration of
postpartum thromboprophylaxis in women with
risk factors other than previous VTE or
thrombophilia.

Summary

r Venous thromboembolism (VTE) is the
leading direct cause of maternal mortality
in the UK, but many cases are potentially
preventable.
r Risk factors for VTE should be identified
pre-pregnancy, or at least early in
pregnancy, and reassessed throughout
pregnancy and the puerperium, as level of risk
may change.

106

r Pregnancy itself is a risk factor for VTE, and
additional risk factors include previous VTE,
thrombophilia, and obesity.
r Thromboprophylaxis should be introduced
depending on the level of risk. Guidelines are
given for both ante-natal and post-natal
management, and in particular for the highest risk
period immediately postpartum.
r Thromboprophylaxis includes both
non-pharmacological and pharmacological
measures, mainly low molecular weight heparin.
r Further research is required to identify additional
thrombophilias, assess whether
thromboprophylaxis should be extended, and
determine optimal duration for postpartum
thromboprophylaxis.

Chapter 8. Thromboprophylaxis

References
1. Confidential Enquiry into Maternal and Child Health.
Saving Mothers’ Lives: reviewing maternal deaths to
make motherhood safer, 2003–5. The Seventh Report of
the Confidential Enquiries into Maternal Deaths in the
United Kingdom. London, RCOG Press, 2007.
2. James AH. Prevention and management of venous
thromboembolism in pregnancy. American Journal of
Medicine 2007; 120: S26–S34.
3. Heit JA, Kobbervig CE, James AH, Petterson TM,
Bailey KR, Melton LJ 3rd . Trends in the incidence of
venous thromboembolism during pregnancy or
postpartum: a 30-year population-based study. Annals
of Internal Medicine 2005; 143: 697–706.
4. Royal College of Obstetricians and Gynaecologists.
Report of the RCOG Working Party on prophylaxis
against thromboembolism in gynaecology and obstetrics.
London, RCOG Press, 1995.
5. Royal College of Obstetricians and Gynaecologists.
Thromboprophylaxis during pregnancy, labour and after
normal vaginal delivery. Guideline no.37. London,
RCOG Press, 2004.
6. Royal College of Obstetricians and Gynaecologists.
Reducing the risk of thrombosis and embolism during
pregnancy and the puerperium. Guideline no.37.
London, RCOG Press, 2009.
7. Knight M on behalf of UKOSS. Antenatal pulmonary
embolism: risk factors, management and outcomes.
British Journal of Obstetrics and Gynecology 2008; 115:
453–461.
8. James AH, Tapson VF, Goldhaber SZ. Thrombosis
during pregnancy and the postpartum period.
American Journal of Obstetrics and Gynecology 2005;
193: 216–219.
9. Blanco-Molina A, Trujillo-Santos J, Criado J
et al. Venous thromboembolism during pregnancy or
postpartum: findings from the RIETE Registry.
Thrombosis and Haemostasis 2007; 97: 186–
190.
10. De Stefano V, Martinelli I, Rossi E et al. The risk
of recurrent venous thromboembolism in pregnancy
and puerperium without antithrombotic
prophylaxis. British Journal of Haematology 2006; 135:
386–391.
11. Nelson SM, Greer IA. Thrombophilia and the risk for
venous thromboembolism during pregnancy, delivery,
and puerperium. Obstetric Gynecologic Clinics of North
America 2006; 33: 413–427.
12. James AH, Jamison MG, Brancazio LR, Myers ER.
Venous thromboembolism during pregnancy and the

postpartum period: incidence, risk factors, and
mortality. American Journal of Obstetrics and
Gynecology 2006; 194: 1311–5.
13. Larsen TB, Sorensen HT, Gislum M, Johnsen SP.
Maternal smoking, obesity, and risk of venous
thromboembolism during pregnancy and the
puerperium: a population-based nested case-control
study. Thrombosis Research 2007; 120: 505–
509.
14. Jacobsen AF, Skjeldestad FE, Sandset PM. Ante- and
postnatal risk factors of venous thrombosis: a
hospital-based case-control study. Journal of
Thrombosis and Haemostasis 2008; 6: 905–912.
15. Lindqvist P, Dahlbäck B, Marŝál K. Thrombotic risk
during pregnancy: a population study. Obstetrics and
Gynecology. 1999; 94: 595–599.
16. Casele H, Grobman WA. Cost-effectiveness of
thromboprophylaxis with intermittent pneumatic
compression at Cesarean delivery. Obstetrics and
Gynecology 2006; 108: 535–540.
17. Bates SM, Greer IA, Pabinger I et al. Venous
thromboembolism, thrombophilia, antithrombotic
therapy, and pregnancy. American College of Chest
Physicians Evidence-Based Clinical Practice
Guidelines (8th edn.). Chest 2008; 133: 844S–886S.
18. British National Formulary 55. London, BMJ Group
and RPS Publishing, 2008.
19. Greer IA, Nelson-Piercy C. Low-molecular-weight
heparins for thromboprophylaxis and treatment of
venous thromboembolism in pregnancy: a systematic
review of safety and efficacy. Blood 2005; 106:
401–407.
20. Carlin AJ, Farquharson RG, Quenby SM et al.
Prospective observational study of bone mineral
density during pregnancy: low molecular weight
heparin versus control. Human Reproduction 2004; 19:
1211–1214.
21. Gates S, Brocklehurst P, Davis LJ. Prophylaxis for
venous thromboembolic disease in pregnancy and the
early postnatal period. Cochrane Database Systematic
Review 2002; (2):CD001689.
22. Johnston JA, Brill-Edwards P, Ginsberg JS et al.
Cost-effectiveness of prophylactic low molecular
weight heparin in pregnant women with a prior
history of venous thromboembolism. American
Journal of Medicine 2005; 118: 503–514.
23. Bauersachs RM, Dudenhausen J, Faridi A et al. Risk
stratification and heparin prophylaxis to prevent
venous thromboembolism in pregnant women.
Thrombosis and Haemostasis 2007; 98: 1237–1245.

107

Section 3. Thromboembolism and anticoagulation

24. Brill-Edwards P, Ginsberg JS, Gent M et al. Safety of
withholding heparin in pregnant women with a
history of venous thromboembolism. Recurrence of
Clot in this Pregnancy Study Group. New England
Journal of Medicine 2000; 343: 1439–1444.

108

25. Maybury H J, Waugh J J S, Gornall A, Pavord S. There
is a return to non-pregnant coagulation parameters
after four not six weeks postpartum following
spontaneous vaginal delivery. Obstetric Medicine 2008;
1: 92–94.

Section 3
Chapter

Thromboembolism and anticoagulation

Prosthetic heart valves
Claire McLintock

Introduction
Patients with mechanical heart valves require longterm anticoagulation, but the choice of anticoagulant
for these women during pregnancy presents a major
challenge. Oral anticoagulants such as warfarin and
coumadin are the most effective agents for prevention of maternal thromboembolism, but freely cross
the placenta and are teratogenic. They also cause late
fetal loss in as many as one in ten pregnancies. Anticoagulation with unfractionated heparin (UFH) and
low molecular weight heparin (LMWH), which do not
cross the placenta will reduce the risk of these adverse
fetal outcomes. However, there is concern that these
drugs are less effective at preventing maternal valve
thrombosis and systemic thromboembolism. Therein
lies the challenge: the anticoagulant that is safest for
the mother’s physical health carries the greatest potential risk for her infant. Many women will choose a
treatment that is safest for her baby, even if her own
health may be compromised. Despite being advised
that thrombotic complications may necessitate urgent
valve-replacement surgery or lead to major neurological sequelae, many women are reluctant to take
oral anticoagulants during pregnancy when informed
about fetal risks. At worst, some women are noncompliant with all therapy, causing even greater maternal risk. This chapter addresses the management of
pregnancy in women with mechanical heart valves and
discusses the maternal and fetal risks associated with
the different anticoagulant options, to enable clinicians
and women to make the most informed choice in this
challenging clinical situation.

Indication for valve replacement
The most common indications for replacement of a
native heart valve are congenital valvular disease and

rheumatic heart disease. While the incidence of congenital valvular disease is relatively stable at around
0.25% of births, there is marked variation in the rates
of rheumatic fever and rheumatic heart disease (RHD)
across different countries. Rheumatic fever is certainly
more common in resource-poor countries and communities, but it is also prevalent in countries such
as New Zealand and Australia with high rates in
the indigenous Aboriginal and Maori people in these
countries and in Pacific Island people in New Zealand
and the Pacific region (Table 9.1).1–3
Table 9.1 Rates of acute rheumatic fever in different regions of
the world

Country

Ethnicity

Rate of acute
rheumatic fever
in children aged
5–14 y (rate per
100 000)

Sub-Saharan
Africa1

13.4

South-central
Asia1

54.0

New Zealand2

Maori
Pacific Island peoples
European

30.4
77.7
1.7

Australia3

Aboriginal and
Torres Strait
Islander peoples
Other Australian
people

162–375
1.0

China1

21.2

Consideration of valve type in women
of child bearing age
For patients who require heart valve replacement,
the alternatives include bioprosthetic valves – either
homograft (human tissue) or heterograft (porcine or

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

109

Section 3. Thromboembolism and anticoagulation

Table 9.2 Mechanical valve types

Valve type
Ball and cage

Starr-Edwards

Valve related factors

Older type ball-cage valves, i.e.
Starr–Edwards valve
Valve in mitral position
≥2 prosthetic valves

Clinical risk factors

Previous thromboembolism
Atrial fibrillation
Dilated left atrium
Heart failure

Tilting disc

Bjork-Shiley
R

Medtronic-Hall

Bileaflet

St Judes
R

CarboMedics
R

On-X
R

ATS valve

bovine tissue) – or mechanical valves (Table 9.2). The
major advantage of bioprosthetic valves for women
of child-bearing age is that, in the absence of other
thromboembolic risk factors, oral anticoagulant therapy is not required. The disadvantage is the high
rate of structural valve deterioration, the commonest reason for replacement of bioprosthetic valves.
Structural deterioration of bioprosthetic valves in the
mitral position occurs more commonly than of those
in the aortic or tricuspid position and was reported
in 84% of bioprosthetic mitral valves by 10 years.
Although valve deterioration occurred more rapidly
in younger patients, it was not further accelerated
by pregnancy.4 In contrast, structural valve failure is
extremely uncommon with mechanical heart valves,
but the risks of valve thrombosis and systemic thromboembolism mean that patients must take long-term
oral anticoagulant therapy. When considering which
type of valve to use in young women, clinicians should
take into account the impact of the decision on management and outcome of future pregnancies.5

Prevention of thromboembolism

110

Table 9.3 Risk factors for thromboembolism in patients with
prosthetic valves

An overall approach to management of women with
prosthetic heart valves is summarized in Figure 9.1.
Thromboembolic complications of mechanical valves
include valve thrombosis, causing valve obstruction or systemic embolization, mainly cerebrovascular accidents (CVA) but also myocardial infarction or
embolization into peripheral arteries. Systemic thromboembolism can develop from either obstructed or
non-obstructed valves. Prevention of these complications is the main indication for long-term anticoagulation. Outside of pregnancy, the rate of major systemic embolization in patients with mechanical valves
is around 1% per year in patients taking warfarin, 2%
per year in patients taking aspirin, and 4% per year in
patients on no anticoagulation.6 Additional risk factors for thromboembolism are shown in Table 9.3. The

prothrombotic changes of pregnancy further increase
the risk of thromboembolism with events reported in
4% of women taking warfarin during pregnancy.7

Anticoagulant management
during pregnancy
Available guidelines accurately state that continuation
of oral anticoagulants (OAC) is the safest option for
the mother for prevention of valve thrombosis and recommend that they should be used during pregnancy
in spite of the known fetal risks (Table 9.4). Most suggest continuation of OAC throughout pregnancy, perhaps substituting adjusted-dose UFH or LMWH during the first trimester, the teratogenic risk-period.8–10
However, ongoing use of OAC is associated with a significant risk of late fetal loss, as high as 10%, mainly
as a result of complications from fetal anticoagulation.
Maternal concern relating to these fetal risks and the
desire for a healthy baby often means that women opt
for the most dangerous option of all, taking no anticoagulation during pregnancy. The recent American
College of Chest Physicians Guidelines on Anticoagulant Therapy 8 is the first published guideline to recommend either adjusted dose LMWH or unfractionated heparin (UFH) throughout pregnancy or until the
beginning of the 13th week of pregnancy, switching to
warfarin until close to delivery.
Unfractionated heparin and LMWH are recommended for prevention and treatment of thrombosis
during pregnancy in women who are at risk of thromboembolism from other causes.8 These agents do not
cross the placenta, are not teratogenic and have no
fetal anticoagulant effect. LMWH with its more predictable anticoagulant effect and better side effect profile has supplanted UFH for treatment and prevention of venous thromboembolism during pregnancy.
Data relating to the safety and efficacy of LMWH in
prevention of thromboembolic complications in pregnant women with mechanical heart valves are limited,

Chapter 9. Prosthetic heart valves

Table 9.4 Summary of published guidelines of management of anticoagulation in pregnant women with mechanical heart valves

Warfarin

LMWH

UFH

Aspirin

American College of
Chest Physicians
20088

Only for women at very high
risk thromboembolism, i.e.
history thromboembolism,
older type valve in mitral
position

Can be used throughout
pregnancy. Twice daily
dose-adjusted,
manufacturers peak
anti-Xa level (1.0 U/ml)
4h post-dose

Can be used throughout
pregnancy. Initial dose
17 500–20 000 U every
12 h. 6 hour
post-injection aPTT 2x
baseline or anti-Xa
levels 0.35–0.70 U/mL

75–100 mg
throughout
pregnancy for
high risk
women

American Heart
Association 20089

Reasonable to avoid 6–12
weeks’ gestation.
Embryopathy risk 4–10%.
INR target 3.0 (range
2.5–3.5). Discontinue 2–3
weeks before planned
delivery

Can be used 6–12 weeks’
gestation. Twice daily,
dose-adjusted with 4 h
post-dose target
anti-Xa levels 0.7–1.2
U/ml

Can be used 6–12 weeks’
gestation. Continuous
iv UFH or
dose-adjusted s/c UFH
bid. Starting dose s/c
17 500–20 000 U every
12h. Target aPTT 2x
baseline

75–100 mg
during second
and third
trimesters

European Society of
Cardiology 200710

Favored in first trimester if
dose ≤5 mg. Favored
anticoagulant during the
second and third trimester
until week 36.

Currently not
recommended –
insufficient safety and
efficacy data

Close monitoring is
required when used

Not discussed

All guidelines recommend full discussion of the risks and benefits of anticoagulant regimens.

*
*
*
*

Assess thrombotic risk: valve
type, position, previous
thromboembolism, cardiac
rhythm, left atrial size
Discuss anticoagulant
regimens
Option 1: therapeutic dose
LMWH throughout, i.e.
enoxaparin 1 mg / kg bid
Option 2: therapeutic dose
LMWH weeks 6−12, warfarin
from week 13
Option 3: warfarin
throughout

Early clinical review
when pregancy
confirmed
*
*
*

Decide on anticoagulant
regimen
Option 1, 2 or 3
Start low dose aspirin: 100−
150 mg

*
*
*
*

Option 1 or 2: stop warfarin
before 6 weeks
Start therapeutic dose LMWH

Ongoing management
for option 2 (and 3)
*

Measure anti-Xa levels day 3−5
Target anti-Xa: trough (predose) 0.4−0.7 IU and peak
(4 hours postdose) 0.7−1.2 IU

Monthly monitoring of anti-Xa
levels

Further discussion of anticoagulant
regimen at week 13.

Pre- pregnancy
counseling

Switch LMWH to warfarin at
week 13 (option 2)

* INR
* At least monthly INR

measurement: target range
determined by valve type

Planned delivery at 38−39
weeks

Mode of delivery detrmined
by obstetric indications

Last dose LMWH 36 hours
pre-delivery

Start intravenous UFH 24
hours prior to delivery

Switch to therapeutic dose
enoxaparin at week 34−36

1st trimester
Option 1 or 2

All groups

Fig. 9.1 Approach to management of women with mechanical heart valves in pregnancy.

but some studies suggest that it is less effective than
warfarin at preventing maternal thromboembolism.
A major systematic review by Chan and co-workers7
compared the maternal and fetal outcome of different anticoagulant regimens in 1234 pregnancies in 976
women in women with mechanical heart valves from
28 studies of women conducted between 1966 and
1997. Comparisons were made across four broad categories of anticoagulant regimens:
r oral anticoagulants throughout pregnancy (792
pregnancies);

r unfractionated heparin in the first trimester
followed by oral anticoagulants (230 pregnancies);
r unfractionated heparin throughout pregnancy (21
pregnancies);
r antiplatelet agents or no anticoagulant therapy
(102 pregnancies).
Since that time, a number of further studies have been
published,11–13 including reviews of pregnancy outcomes in women taking LMWH at some stage during
pregnancy.14,15 A summary of published data of maternal and fetal outcomes in pregnancies where women

111

Section 3. Thromboembolism and anticoagulation

Table 9.5 Maternal and fetal outcomes in pregnancies in women with mechanical heart valves related to anticoagulant
management approach

Miscarriage
N (%)

Stillbirth∗
(Σ )

Warfarin
embryopathy

Maternal
thromboembolic
complications

Anticoagulant regimen

Pregnancies

Warfarin throughout pregnancy7,12,13,27

983

253
(25.7%)

13/127§
(10.2%)

39/740
(5.3%)

36
(3.7%)

Warfarin-UFH-warfarin7,11,13

285

66
(23.4%)

3/43§
(7.0%)

6/229
(2.6%)

28
(9.9%)

Warfarin-LMWH-warfarin15

4
(7.1%)

1/51
(1.9%)

5
(8.9%)

LMWH throughout11,13,15

2
(8.3%)

1
(4.2%)

∗
§

pregnancies included in Chan review excluded as stillbirths not reported as a separate group.
excluding spontaneous miscarriages and terminations of pregnancy.

have received different anticoagulant regimens is presented in Table 9.5.

Anticoagulation with warfarin
during pregnancy
Oral anticoagulants, such as warfarin and acenocoumarol, are the most effective agents for prevention
of valve thrombosis and systemic thromboembolism
during pregnancy in women with mechanical heart
valves. Disadvantages of oral anticoagulants include
teratogenicity, high rates of spontaneous abortion and
late fetal loss as well as neurological abnormalities in
surviving infants.

Fetal effects of warfarin

112

Exposure to warfarin in the first trimester causes warfarin embryopathy in as many as 6% of infants. Hall
and co-workers16 recommended that nasal hypoplasia and stippled epiphyses should be the minimal features required to classify a case as warfarin embryopathy. These features are not described in women
who substitute heparin for warfarin between 6–12
weeks of pregnancy or who are exposed to warfarin
from the second trimester onwards. A recent review of
63 published cases of warfarin-related abnormalities17
described skeletal anomalies in 81% of cases (n=51)
with mid facial hypoplasia described in 47 infants and
epiphyseal calcific stippling of long bones, vertebrae,
calcanei, or phalanges in 32 infants. Breathing and
feeding problems were present in 24 of 47 infants who
had severe midfacial hypoplasia. The period of exposure to warfarin, common to infants who developed

embryopathy, was between 6 and 9 weeks’ gestation.
Long-term follow-up information was available on 20
of 46 children in this cohort who survived the neonatal period with abnormalities persisting in about half
of the children with midline hypoplasia and spinal
deformities. These teratogenic effects are unlikely to
be caused by inhibition of Vitamin K-dependent clotting proteins by warfarin, as these are not produced
by the fetal liver until after 12–14 weeks’ gestation.
Vitamin K-dependent proteins are also important in
development of bone and cartilage and inhibition of
these proteins may account for this teratogenic effect of
warfarin.
Microcephaly, cerebral atrophy, hydrocephalus,
optic atrophy, and intracranial hemorrhage are among
the central nervous system abnormalities described
in 1% of liveborn infants exposed to warfarin during
pregnancy. The prevalence of long-term neurological
problems, such as developmental delay and low IQ, in
infants who appear normal after in-utero exposure to
warfarin is still debated.

Warfarin and pregnancy outcome
Warfarin exposure during pregnancy is associated
with increased rates of spontaneous miscarriage and
a 10% rate of late fetal loss, Table 9.5. The majority
of late losses are thought to be due to fetal intracranial hemorrhage (ICH) as a result of anticoagulation
of the fetus. Warfarin freely crosses the placenta so
it will inhibit fetal vitamin K-dependent clotting proteins that are produced from 12–14 weeks’ gestation.
As vitamin K levels in the fetus are one-tenth of the
levels in the mother, the dose of warfarin taken by the

Chapter 9. Prosthetic heart valves

mother to achieve a therapeutic INR will cause severe
over-anticoagulation in the fetus, possibly with INR
levels as high as those that develop in elderly patients
on warfarin who become vitamin K deficient when
they are unwell, stop eating, and take antibiotics.

Effect of warfarin dose on
pregnancy outcome
Vitale and co-workers18 described increased fetal complication rates in women with mechanical heart valves
taking sodium warfarin doses of ⬎5 mg during pregnancy compared to women on lower doses. This dosedependent relationship was further examined in an
updated publication of their pregnancy cohort including 71 pregnancies in 52 consecutive patients.19 Fetal
losses were reported in 78.8% of 33 pregnancies in
women taking ⬎5 mg warfarin, including 21 spontaneous abortions between 12 and 20 weeks’ gestation
and five stillbirths after 20 weeks’ gestation. Of 38 pregnancies in women taking ≤5 mg of warfarin daily only
two spontaneous abortions (5.3%) were reported. The
cause of spontaneous abortion or stillbirth was not
reported. Rates of classical embryopathy did not differ between groups with skeletal anomalies reported in
three infants; two with nasal hypoplasia (one in a term
infant in the ≤5 mg warfarin group and one in a spontaneously aborted fetus in the ⬎5 mg group) and cervical spine abnormalities reported in a spontaneously
aborted fetus from the higher dose group. This dosedependent relationship with adverse fetal outcome is
not described in all studies4 and the clinical impact
of such an association is uncertain as OAC dosage is
determined by maternal INR.

Anticoagulation with heparin
during pregnancy
Unfractionated heparin and
antiplatelet agents
Unfractionated heparin and antiplatelet agents such
as aspirin were used as alternatives to warfarin for
anticoagulation of pregnant women with mechanical heart valves prior to the development of LMWH.
Although use of antiplatelet agents avoided the risk
of congenital anomalies the thromboembolic complications were reported in 29% of pregnancies where
women took antiplatelet agents alone.7 Similarly, no

fetal anomalies were reported in pregnancies where
UFH was substituted for warfarin prior to six weeks
gestation and continued until completed 12 weeks’ gestation. Variation in rates of thromboembolic complications with UFH are likely to be confounded by differences in heparin dosing, anticoagulant monitoring,
and duration of treatment. Thromboembolic complications occurred in 9.9% of these pregnancies where
women received UFH (Table 9.5) including events in
the two recent studies11,13 where the dose of UFH was
adjusted to keep the activated partial thromboplastin time (aPTT) at two to three times baseline levels. In 21 pregnancies where UFH was used throughout, thromboembolic complications were reported in
33%.7 Concerns with maternal safety with UFH provided the impetus to explore the efficacy of LMWH
with its more predictable anticoagulant effect as an
alternative anticoagulant for women with mechanical
heart valves who do not wish to take warfarin during
pregnancy. Long-term use of therapeutic dose UFH
also increases the risk of osteoporosis and heparininduced thrombocytopenia.

Maternal and fetal outcomes in women
receiving LMWH during pregnancy
Limited data are available of pregnancy outcome in
women with mechanical heart valves treated with
LMWH during pregnancy; outcomes in women taking either LMWH during the first trimester only or
throughout pregnancy account for only 4.1% and
2.1% of published reports, respectively (Table 9.5). In
their review of published studies of pregnancy outcome in women with mechanical heart valves who
received LMWH during pregnancy, Oran and coworkers reported seven episodes of valve thrombosis
and two CVAs in 81 pregnancies, a complication rate
of 11.1%.15 Single case reports and small case series
accounted for six of the thromboembolic events in this
review. Limiting analysis to studies including ≥5 pregnancies the rate of valve thrombosis was 4.8%. Dose
adjustment of LMWH in response to anti-Xa levels
appears to be a key factor in management of anticoagulation with LMWH in pregnant women with mechanical heart valves: six of the seven valve thromboses in
Oran’s review occurred in women who did not have
anti-Xa levels measured.
Another consideration with heparin use is the
risk of heparin-induced thrombocytopenia and bone

113

Section 3. Thromboembolism and anticoagulation

Table 9.6 Quick-reference guide: estimates of maternal and fetal risks with different anticoagulant regimens for pregnant women with
mechanical heart valves

Anticoagulant regimen

Maternal valve
thrombosis and systemic
thromboembolism

Spontaneous
abortion

Late fetal
loss

Warfarin
embryopathy

Warfarin continued throughout pregnancy

Low

High

Moderate

Dose-adjusted UFH weeks 6–12, then warfarin

Moderate

High

Moderate

No risk

Therapeutic-dose LMWH weeks’ 6–12 with
anti-Xa monitoring, then warfarin

? Low

Moderate

Low

No risk

Therapeutic-dose LMWH throughout
pregnancy started ⬍6 weeks’ gestation
with anti-Xa monitoring

? Low

Moderate

Low

No risk

Low risk ⬍5% Moderate risk = 5%–10% High
risk ⬎10%

114

mineral density loss. This is less common with LMWH
than with UFH.

Table 9.7 Anticoagulant options offered to women with
mechanical heart valves attending high-risk medical ante-natal
clinic, National Women’s Health, Auckland City Hospital,
New Zealand

Role of aspirin

Option 1. Substitution of warfarin with therapeutic dose
enoxaparin (1 mg/kg bd) and aspirin 150 mg before 6
weeks’ gestation, continued until planned delivery

Outside of pregnancy, low dose aspirin (100–150 mg)
is recommended in addition to warfarin for patients
with mechanical heart valves with other thromboembolic risk factors (Table 9.3). Low dose aspirin has been
shown to reduce the risk of major thromboembolic
events such as valve thrombosis and CVA in patients
at the expense of an increase in minor, but not major
bleeding.20 The prothrombotic changes of pregnancy
could also be considered as an additional risk factor
for thromboembolism supporting the use of low dose
aspirin during pregnancy in addition to warfarin or
heparin.
A simplified summary of risks and benefits associated with UFH, LMWH and warfarin is provided in
Table 9.6.
Although there are limited published data relating
to use of LMWH for anticoagulation during pregnancy
in women with mechanical heart valves, it is possible that thromboembolic complications may in part
relate to suboptimal LMWH doses. Regular monitoring of anti-Xa levels with dose-adjustment of
LMWH may provide more effective thromboprophylaxis. The high risk ante-natal medical clinic at
National Women’s Health, Auckland City Hospital has
published one of the largest single case series of women
with mechanical heart valves treated with enoxaparin
and aspirin during pregnancy.21 Women attending this
clinic are informed that the safest option for them is
not to become pregnant but those who choose to proceed with pregnancy have in-depth counseling of the

Option 2. Substitution of warfarin with therapeutic dose
enoxaparin before 6 weeks until 12 completed weeks,
reverting to warfarin until 34–36 weeks, then therapeutic
dose enoxaparin until planned delivery. Aspirin 150 mg
throughout pregnancy.
Option 3. Warfarin and aspirin 150 mg throughout pregnancy,
switching to therapeutic enoxaparin and aspirin at 34–36
weeks’ gestation until planned delivery.

Low-dose aspirin (100–150 mg) recommended for all women

maternal and fetal risks and benefits of different anticoagulant agents and provide written consent for their
choice of one of three anticoagulant regimens (Table
9.7). Regular anticoagulant monitoring is carried out
with target therapeutic ranges of INR and anti-Xa levels as listed in Table 9.8. Low-dose aspirin is recommended for all women.
The thromboembolic complications reported with
dose-adjusted UFH and the variable dose–response
rates with this agent suggest that it may be a less reliable
alternative to LMWH. Given the significant advantages in terms of fetal outcome, perhaps monitored
therapeutic-dose LMWH for prevention of maternal valve thrombosis and systemic thromboembolism
could be an attractive alternative to warfarin for use in
this clinical setting.22 Certainly, it would seem premature to contraindicate use of therapeutic dose LMWH
during pregnancy for these women23 given the lack
of an acceptable alternative anticoagulant and also the

Chapter 9. Prosthetic heart valves

Table 9.8 Anticoagulant monitoring for women with mechanical heart valves attending high-risk medical ante-natal clinic, National
Women’s Health, Auckland City Hospital, New Zealand

Drug

Laboratory test

Valve type

Anticoagulant target
range

Enoxaparin (1 mg/kg bd)

Anti-Xa levels – 3–5 days after first
dose then monthly

All valves

Trough (pre-dose) 0.4–0.7 IU
Peak (4 h post-dose) 0.7–1.2 IU

Warfarin

Monthly INR

Starr–Edwards valves
Bileaflet and tilting disc valves

3.0–4.5
2.5–3.5

continued emergence of safety and efficacy data of its
use in this clinical setting. In the absence of a randomized clinical trial, clinicians must rely on bestpractice guidelines based on existing evidence and
experience.
Other critical issues to consider when deciding on
anticoagulant management in pregnant women with
mechanical heart valves include the cost of LMWH
and access to laboratory testing for anti-Xa levels. In
countries with limited resources it may be more appropriate to treat women with warfarin but these decisions
will rest with individual clinicians.

Management of labor and delivery

Other management issues

Women on oral anticoagulants

Management of women in the peri-delivery period
requires close clinical monitoring, given the bleeding risks associated with therapeutic anticoagulation.
The mode of delivery should be determined by obstetric indications but vaginal delivery is preferable to
Cesarean section as it minimizes time off therapeuticdose anticoagulation – the increased risk of bleeding
from the operative site necessitates a delay in restarting anticoagulation postpartum. A planned delivery
allows for better control and adjustment of anticoagulation.

Regular follow-up during pregnancy is essential to
make sure frequent monitoring of anticoagulation is
done, and careful clinical assessment to detect any cardiac complications. Table 9.9 lists important aspects of
care. The infants of mothers with congenital heart disease have a five- to tenfold increased risk of congenital
heart defects themselves and careful anatomy scan or
fetal echocardiogram is required at 20–24 weeks’ gestation.

Women should stop warfarin by 34–36 weeks’ gestation to allow normalization of the infant’s INR and
minimize the risk of fetal intracranial hemorrhage at
delivery. Although cessation of OAC leads to normalization of the woman’s INR within 3–4 days, the process often takes much longer in the infant as it is likely
to be over-anticoagulated given the immaturity of its
coagulation system. Anticoagulant options for women
who have taken warfarin until the peri-delivery period
include:

Table 9.9 Summary of ante-natal care required for women
with mechanical heart valves: other maternal and fetal issues

Management of pregnancies in women with mechanical heart
valves: other issues

(1) Regular patient review: every 4 weeks until 28 weeks’
gestation, every two weeks until 34 weeks’ gestation then
weekly until delivery
(2) Urgent assessment if patient concerns about possible
thromboembolism, change in cardiac symptoms,
bleeding, preterm labor
(3) Clinical assessment to include enquiry about symptoms of
systemic thromboembolism, heart failure, cardiac rhythm,
maternal hypertension.
(4) Careful auscultation of maternal heart at each visit: new
murmurs.
(5) Maternal echocardiogram every trimester or more
frequently if clinical concern
(6) Fetal surveillance: regular scans for fetal growth, evidence
of intracranial hemorrhage

continuous intravenous unfractionated heparin
aiming for aPTT 2–3x baseline;
dose-adjusted subcutaneous UFH bid aiming for
aPTT 2–3x baseline;
therapeutic dose LMWH with monitoring of
anti-Xa levels (see Table 9.8).
Tables 9.10 and 9.11 summarize peri-delivery management of anticoagulation for planned vaginal delivery or cesarean section, respectively. Gradual reintroduction of intravenous UFH, as outlined in Tables
9.10 and 9.11, is the preferred option for anticoagulation in the immediate postpartum period. Intravenous UFH has the advantage over LMWH of allowing more flexible dose adjustment including more

115

Section 3. Thromboembolism and anticoagulation

Table 9.10 Management of anticoagulation prior to induction
of labor for planned vaginal delivery in women on
subcutaneous LMWH (National Women’s Health, Auckland City
Hospital, New Zealand)
(1) 36 hours prior to planned CS – last dose s/c LMWH
(2) 24 hours prior to induction – start IV UFH infusion 5000 U
bolus dose then 1200 U/h
(3) Check aPTT 6-hourly – target aPTT 2–3 × baseline∗
(4) Discontinue iv UFH when woman in established labor
(5) If regional analgesia required stop iv UFH 4 hr prior to
epidural catheter placement, check aPTT back to baseline
prior to placement, restart iv UFH 3 hours after catheter
placement
(6) 4–6 h postpartum, restart iv UFH at 500 U/h (no bolus dose)
if no bleeding concerns then increase dose by 250 U every
4–6 hours until aPTT 2–3 × baseline∗
(7) Start oral anticoagulant on first postpartum day if
uncomplicated vaginal delivery or day 2–3 if Cesarean
section or other bleeding complications
∗

therapeutic range may vary between centers, being dependent
on the sensitivity of the aPTT reagent used. It should also be
noted that the aPTT can be less reliable in pregnancy due to
increased levels of factor VIII and heparin binding proteins.

Table 9.11 Management of anticoagulation prior to elective
Cesarean section in women on subcutaneous LMWH (National
Women’s Health, Auckland City Hospital, New Zealand)
(1) 36 hours prior to planned CS – last dose s/c LMWH
(2) 24 hours prior to planned CS – start iv UFH infusion 5000 U
bolus dose then 1200 U/h
(3) Check aPTT 6-hourly – target aPTT 2–3 × baseline∗
(4) Stop iv UFH 4 hr prior to epidural catheter placement for
regional anesthesia, check aPTT back to baseline prior to
placement.
(5) 6–12 h post-delivery, restart iv UFH at 500 U/h (no bolus
dose) if no bleeding concerns, then increase dose by 250 U
every 4–6 hours until aPTT 2–3 × baseline∗
(6) Delay starting oral anticoagulant until epidural catheter
removed if remains in situ for post-operative pain
management
∗

116

rapid and complete reversal of its anticoagulant effect
in the event of clinically significant bleeding in the
postpartum period, especially as oral anticoagulants
are restarted. Some centers may lack the experience of
managing intravenous UFH infusions so an alternative
may be to use intermediate doses of LMWH i.e. enoxaparin 40mg daily then twice daily until the first few
days postpartum until the INR is therapeutic. However, not only is there the concern of a higher risk of
valve thrombosis with low doses of LMWH, if bleeding does occur, the anticoagulant effect cannot be completely reversed.

In the case of an obstetric emergency such as
preterm labor or placental abruption, rapid reversal of anticoagulation is required. Management
of heparin reversal with protamine is outlined in
Chapter 10. Effective reversal of oral anticoagulation
requires administration of 5.0–10.0 mg vitamin K1
intravenously, as well as prothrombin concentrate or
fresh frozen plasma.

Prevention of infective endocarditis
Antibiotic prophylaxis against infective endocarditis is
recommended for women with prosthetic heart valves
following vaginal delivery or Cesarean section, given
the high risk of adverse outcome should this complication occur.24 Women who have prosthetic heart
valves as a result of rheumatic heart disease require
secondary prevention of acute rheumatic fever and
rheumatic heart disease for a minimum of 10 years
after the most recent episode or until 30–40 years
of age (whichever is longer). Benzathine penicillin-G,
benzylpenicillin, or erthyromycin are considered safe
in pregnancy.

Management of valve thrombosis
Development of neurological symptoms, chest pain or
symptoms of heart failure, or detection of a new cardiac murmur warrants exclusion of valve thrombosis
using transthoracic or transesophageal echocardiography. Minor valve thromboses can often be managed by increasing the intensity of anticoagulation,
but for more severe thromboses valve replacement
surgery is usually required, although some centers
have reported success with thrombolysis using rtPA
or streptokinase.25 Women with valve thrombosis
should be managed jointly by the obstetric and cardiothoracic surgical team. Readers are referred to
recent review articles discussing management issues in
pregnancy.26–28

Summary
The clinical challenge of managing anticoagulation in
pregnancy in women with mechanical heart valves is
set to continue. These women and the clinicians caring
for them are faced with a true dichotomy: the choice
of warfarin or heparin. Taking warfarin in pregnancy
minimizes the risk of thrombotic complications but
carries a high-risk of adverse fetal outcome; LMWH

Chapter 9. Prosthetic heart valves

has an uncertain maternal risk but the fetal complications are very low. Experience suggests that, given
the choice, many women will opt for what is safer
for the baby – especially if it is their first pregnancy
or they have had a bad fetal outcome with warfarin
previously. However, women who develop complications with valve thrombosis or stroke due to suboptimal anticoagulation also place their babies at risk.

Data relating to the safety and efficacy of LMWH continues to emerge and in the absence of a randomized trial, data from registries such as the International
Registry of Pregnancies in Women with Mechanical
Heart Valves developed by the ISTH Subcommittee on
Women’s Health Issues in Thrombosis and Haemostasis may help provide clearer guidance for management
in the future.

117

Section 3. Thromboembolism and anticoagulation

References
1. Carapetis JR, Steer AC, Mulholland EK, Weber M. The
global burden of group A streptococcal diseases. The
Lancet Infectious Diseases. 2005; 5: 685–694.
2. Atatoa-Carr P, Lennon D, Wilson N. Rheumatic fever
diagnosis, management, and secondary prevention: a
New Zealand guideline. New Zealand Medical Journal
2008; 121: 59–69.
3. Australian Institute of Health and Welfare. Rheumatic
heart disease: all but forgotten in Australia except
among Aboriginal and Torres Strait Islander peoples.
Australian Institute of Health and Welfare: Australian
Government 2004.
4. Sadler L, McCowan L, White H. Pregnancy outcomes
and cardiac complications in women with mechanical,
bioprosthetic and homograft valves. BJOG: An
International Journal of Obstetrics and Gynaecology
2000; 107: 245–253.
5. Bach DS. Choice of prosthetic heart valves:update for
the next generation. Journal of the American College of
Cardiology 2003; 19; (10): 1717–1719.
6. Cannegieter SC, Rosendaal FR, Briet E.
Thromboembolic and bleeding complications in
patients with mechanical heart valve prostheses.
Circulation 1994; 89: 635–641.
7. Chan WS, Anand S, Ginsberg JS. Anticoagulation of
pregnant women with mechanical heart valves: a
systematic review of the literature. Archives of Internal
Medicine 2000; 160: 191–196.
8. Bates SM, Greer IA, Pabinger I et al. Venous
thromboembolism, thrombophilia, antithrombotic
therapy, and pregnancy: American College of Chest
Physicians Evidence-Based Clinical Practice
Guidelines (8th edn). Chest 2008; 133: 844S–886S.
9. Bonow RO, Carabello BA, Chatterjee K et al. focused
update incorporated into the ACC/AHA 2006
guidelines for the management of patients with
valvular heart disease: a report of the American
College of Cardiology/American Heart Association
Task Force on Practice Guidelines. 2008; 52: e1–142.
10. De Caterina R, Husted S, Wallentin L et al.
Anticoagulants in heart disease: current status and
perspectives [see comment]. European Heart Journal
2007; 28: 880–913.
11. Kawamata K, Neki R, Yamanaka K et al. Risks and
pregnancy outcome in women with prosthetic
mechanical heart valve replacement. Circulation
Journal 2007; 71: 211–213.

118

12. Lee JH, Park NH, Keum DY et al. Low molecular
weight heparin treatment in pregnant women with a
mechanical heart valve prosthesis. Journal of Korean
Medical Science 2007; 22: 258–261.

13. Meschengieser SS, Fondevila CG, Santarelli MT,
Lazzari MA. Anticoagulation in pregnant women with
mechanical heart valve prostheses. Heart 1999; 82:
23–26.
14. James AH, Brancazio LR, Gehrig TR et al.
Low-molecular-weight heparin for
thromboprophylaxis in pregnant women with
mechanical heart valves. Journal of Maternal–Fetal and
Neonatal Medicine 2006; 19: 543–549.
15. Oran B, Lee-Parritz A, Ansell J. Low molecular weight
heparin for the prophylaxis of thromboembolism in
women with prosthetic mechanical heart valves during
pregnancy. Thrombosis and Haemostasis 2004; 92:
747–751.
16. Hall JG, Pauli RM, Wilson KM. Maternal and fetal
sequelae of anticoagulation during pregnancy.
American Journal of Medicine 1980; 68: 122–138.
17. Van Driel D, Wesseling J, Sauer PJJ et al. Teratogen
update: feta; effects ofter in utero exposure to
coumarins overview of cases, follow-up findings, and
pathogenesis. Teratology 2002; 66: 127–140.
18. Vitale N, De Feo M, Cotrufo M. Anticoagulation for
prosthetic heart valves during pregnancy: the
importance of warfarin daily dose. European Journal of
Thoracic and Cardiovascular Surgery 2002; 22: 656.
19. Cotrufo M, De Feo M, De Santo LS et al. Risk of
warfarin during pregnancy with mechanical valve
prostheses. Obstetrics and Gynecology 2002; 99:
35–40.
20. Turpie AGG, Gent M, Laupacis A et al. A comparison
of aspirin with placebo in patients treated with
warfarin after heart-valve replacement. New England
Journal of Medicine 1994; 329: 524–529.
21. Rowan JA, McCowan LM, Raudkivi PJ, North RA.
Enoxaparin treatment in women with mechanical
heart valves during pregnancy. American Journal of
Obstetrics Gynecology 2001; 185: 633–637.
22. McLintock C, North RA, White HD. Prosthetic heart
valves and pregnancy. Circulation 2003; 108: e159–
60.
23. American College of Obstetricians and Gynecologists
Committee on Obstetric Practice. ACOG committee
opinion. Safety of Lovenox in pregnancy. Number 276,
October 2002. Committee on Obstetric Practice.
International Journal of Gynaecology and Obstetrics
2002 ; 79: 299–300.
24. Wilson W, Taubert KA, Gewitz M et al. Prevention of
infective endocarditis: guidelines from the American
Heart Association: a guideline from the American
Heart Association Rheumatic Fever, Endocarditis, and
Kawasaki Disease Committee, Council on
Cardiovascular Disease in the Young, and the Council

Chapter 9. Prosthetic heart valves

on Clinical Cardiology, Council on Cardiovascular
Surgery and Anesthesia, and the Quality of Care and
Outcomes Research Interdisciplinary Working Group.
Circulation 2007; 116: 1736–1754.
25. Leonhardt G, Gaul C, Nietsch HH et al. Thrombolytic
therapy in pregnancy. Journal of Thrombosis and
Thrombolysis 2006; 21: 271–276.
26. Dieter RS, Dieter RA, Jr., Dieter RA, 3rd et al.
Prosthetic heart valve thrombosis: an overview.
Wisconsin Medical Journal 2002; 101: 67–69.

27. Lengyel M, Horstkotte D, Voller H, Mistiaen WP,
Working Group Infection Thrombosis, Embolism and
Bleeding of the Society of Heart Valve Disease.
Recommendations for the management of prosthetic
valve thrombosis. Journal of Heart Valve Disease 2005;
14: 567–575.
28. Nassar AH, Hobeika EM, Abd Essamad HM et al. Usta
IM. Pregnancy outcome in women with prosthetic
heart valves. American Journal of Obstetrics and
Gynecology 2004; 191: 1009–1013.

119

Section 3
Chapter

Thromboembolism and anticoagulation

Management of anticoagulants at delivery
Christina Oppenheimer and Paul Sharpe

Introduction
This chapter will address the practical obstetric and
anesthetic management of women on prophylactic
heparin and therapeutic anticoagulation in the peripartum period, and the dilemmas for obstetricians,
anesthetists, and hematologists. Also considered will
be issues surrounding use of thrombolytic agents in
pregnancy and unusual but complex situations such as
cardiopulmonary bypass in pregnancy.
Increasing use of prophylactic anticoagulants in
pregnancy, both for venous thromboprophylaxis and
to modify fetal risk, as in antiphospholipid syndrome, means that more women are now reaching the
peri-partum period on anticoagulants, usually a low
molecular weight heparin. Therapeutic doses are used
for treatment of acute venous thromboembolic events,
prevention of thromboembolism in women with cardiac disease including mechanical heart valves, acute
cardiac events and cardiomyopathy, and those on longterm anticoagulation outside of pregnancy for a variety
of other indications. This situation necessitates careful
assessment of risks, close multidisciplinary discussion
and planning, and expert management by the medical and midwifery teams during labor or Cesarean section. Careful discussion of risks and therapeutic decisions with the patient and her partner are also essential.

Thromboprophylaxis
The use of a variety of anticoagulants in obstetric practice has been increasing steadily over the last 15 years,
bringing with it increasing awareness of the need for
attention to thromboprophylaxis, but also the need
to adapt and plan for the risks associated with this.
Thromboprophylaxis guidance from the Royal College of Obstetricians and Gynaecologists (RCOG)1

120

and the recent report from the Confidential Enquiry
into Maternal Health (CEMACH)2 have both significantly raised awareness of the importance of risk
assessment for venous thromboembolism, and hence
increased use of low molecular weight heparins in particular (Chapter 8). The recent and rapid rise in prevalence of obesity in women of child-bearing age also
brings many women into a high risk category necessitating use of general as well as pharmacological antithrombotic measures.

Antiplatelet agents
Increasing use of low-dose aspirin, firstly following the
publication of the CLASP trial in 1994,3 then the work
on anti-phospholipid syndrome in recurrent miscarriage by Regan and colleagues, as well as a gradual
increase in numbers of women on long-term aspirin
for medical conditions, such as previous stroke, has led
to guidance on aspirin use around the time of delivery.
The use of aspirin 75 mg up to the time of labor and
delivery is not contraindicated on either obstetric or
anesthetic grounds. At this dose there is no increased
risk of bleeding either at vaginal or Cesarean delivery, nor is there evidence of any increase in the risk
of vertebral canal hematoma after spinal or epidural
block insertion. However, it must be remembered that,
if used with either a heparin or warfarin in the postpartum period, there may be an additive effect and particular care should be taken with timing of dose
administration and epidural catheter removal. There
is no contraindication to breastfeeding on aspirin at
75 mg daily, in particular, there is no evidence of risk
of Reye’s syndrome at this dose.
The safety in pregnancy of other antiplatelet agents
such as clopidogrel or ticlopidine at usual therapeutic doses has not been established and they are rarely
used. Both the indication for use and a clear plan of

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

Chapter 10. Management of anticoagulants at delivery

management to minimize risk at the time of delivery,
including whether to interrupt the treatment, should
be made on an individual case basis with consultation
among responsible obstetrician, physician, and anesthetist.
Non-steroidal anti-inflammatory drugs are largely
contra-indicated in pregnancy and should not be used
around the time of delivery.

in a midwifery-led setting provided at least 8 hours has
elapsed since the last dose by delivery. If women in this
situation request a home confinement, this should be
discussed on an individual basis.

Induction of labor

Most of the women requiring prophylactic doses of
anticoagulant will be given one of the low molecular weight heparins. This is usually given once daily,
although those with a particularly high thrombotic
risk, obesity, or mechanical heart valves may be on
twice daily doses.
While increasing numbers of women being prescribed LMWH for thromboprophylaxis leads to
increased experience for staff in managing such pregnancies, it also necessitates sufficient knowledge for
safe practice, particularly around the time of labor and
delivery. This would take into account the half-life,
increased clearance in pregnancy and different therapeutic index of different heparins.4 Underpinning all
of the clinical management described below must be
collaboration, clear local guidelines, knowledge and
education of medical and midwifery staff, as well as
written information for patients and individual care
plans based on a woman’s particular risk factors.

If an induction of labor is planned, the LMWH should
be omitted at the start of the process. If a morning dose is usual, the dose on the day of induction
should be omitted, or if an evening dose, that on the
evening before should be taken as usual, but none further until after delivery. If an omission of more than
48 hours is thought to be contraindicated, intermittent
doses of subcutaneous UFH, 5000u, could be considered 8 hourly until artificial rupture of the membranes
(ARM) is possible, although this is rarely needed. UFH
has a half-life of about 3 hours, so that an anticoagulant
effect at a level suitable for regional anesthetic blockade is possible around that point. With prophylactic
LMWH this delay needs to be 8–12 hours depending
on dose as discussed below. Attention should be paid to
general antithrombotic measures including compression hose, hydration, and mobility.
A prolonged induction process is more likely in
primigravida and those with an unfavorable cervix at
the start of the induction. Thus, where possible, careful assessment should be made to try to delay induction until the cervix is more favorable. If this is not
possible, there will then be the dual additional risk of
a prolonged period where the woman is only covered
by general antithrombotic measures and an increased
risk of instrumental delivery or Cesarean section (CS)
inherent in an induced labor.

Labor and delivery

Practical issues

In the presence of increased risk of venous thromboembolism, the optimum management will be to aim
for a spontaneous onset of labor as these are on average shorter and have a lower risk of operative delivery. Clearly, this advantage needs to be weighed against
other obstetric and medical risks if present.
A standard approach is to omit the LMWH at
the onset of labor and ensure general antithrombotic
measures, including adequate hydration, with early
recourse to intravenous fluids if necessary, mobilization, with passive movements or massage if mobility
restricted by epidural, and the wearing of graduated
compression stockings. If a woman has no additional
complications other than a need for LMWH thromboprophylaxis, there is no contraindication to delivering

At prophylactic doses of LMWH or UFH, there is no
contraindication to intramuscular analgesics during
labor or to intramuscular syntometrine at delivery. The
LMWH should be recommenced 3–6 hours after delivery, providing hemostasis is secured and the timing
around removal of epidural catheter is considered, as
described below.
It should also be emphasized that, for standard prophylactic doses of LMWH, no significant increase in
the risk of intra- or postpartum bleeding, paravaginal hematoma or prolonged lochia has been shown;
however, it is good practice to ensure adequate and
timely use of uterotonics and early suturing of any tear
or episiotomy. There is no contraindication to pudendal or perineal block for analgesia. Also, there is no

Low molecular weight
heparins (LMWH)
Obstetric aspects

121

Section 3. Thromboembolism and anticoagulation

increased risk of excess surgical bleeding at Cesarean
section, although risk of wound hematoma may be
slightly increased.

Breastfeeding
There should be a clear plan of the length of time
to continue the LMWH postpartum. The woman
should be reassured that there is no contraindication
to breastfeeding since, although the LMWH will be
present in breastmilk, it will be broken down in the
gastric acid before absorption can occur.

Anesthetic aspects
One of the major concerns about the use of heparins in the peripartum period in relation to anesthesia is the risk of vertebral canal hematoma and
its severe sequelae. This was first raised as a significant issue in publications from the USA,5 and subsequently there has been much study and debate leading to some standardization and guidance.6,7 Over
40 cases of vertebral canal hematoma were reported
in the American literature in 1997–1998, from a 5year observation period, in patients given enoxaparin,
mostly following epidural, spinal, or lumbar puncture needle insertion. However, subsequent European
reports include only two such cases with vertebral
canal hematoma. The incidence has been estimated at
1 in over 2 million in a European study, but about
1 in 15 000 in American studies. Numerous factors
including timing of LMWH in relation to needle
insertion or epidural catheter removal, and dosing
schedules have been implicated. Other studies have
suggested that technical difficulties with needle and
particularly epidural catheter insertion may also be
associated with increased risk of hematoma, particularly multiple insertion attempts or blood-stained tap.8
(Table 10.1)

Table 10.1 Estimated incidence for vertebral canal
hematoma in spinal and epidural anesthesia

Epidural Spinal

122

Without heparin
Atraumatic
Traumatic

1/220 000 1/320 000
1/20 000 1/29 000

Heparin given after procedure
Atraumatic
Traumatic
UFH more than 1 hr after puncture
UFH less than 1 hr after puncture

1/70 000
1/2 000
1/100 000
1/8 700

1/100 000
1/2 900
1/150 000
1/13 000

Thromboprophylaxis for elective operative procedures
Single shot spinal versus epidural catheter techniques
Single shot subarachnoid anesthesia remains a popular
choice amongst obstetric anesthetists to provide operative conditions for elective Cesarean section. This is
primarily because of the superior nature of the quality
of anesthesia produced when compared with epidural
anesthesia. Because there is no catheter to be removed,
this technique facilitates the use of a single agent in the
post-operative period. A minimum of 2 hours should
be allowed to elapse between completing the subarachnoid injection and the administration of a LMWH.
The use of an epidural catheter in isolation, for
example to provide more controlled onset of anesthesia in patients with cardiovascular instability, is less
common in the elective setting. The introduction of
the combined spinal epidural (CSE) has allowed a flexible approach. The epidural space is located with a
Tuohy needle using a loss of resistance technique, usually with a 16- or 18-gauge needle. A spinal needle, of
smaller gauge such as 26, is then introduced through
the Tuohy needle to pierce the dura and enter the
subarachnoid space. Once this thinner spinal needle
has been removed, the epidural catheter can then be
threaded through the Tuohy needle into the epidural space. This technique has provided great flexibility. Administration of reduced doses of local anesthetic into the subarachnoid space, with further doses
administered via the epidural catheter allows excellent control over the cardiovascular system. The presence of an epidural catheter also allows the anesthetist
to provide additional doses of anesthetic in cases of
inadequate anesthesia or prolonged operative delivery. However, as this technique involves a catheter in
situ at the end of surgery, the operative team are presented with two options in women requiring thromboprophylaxis: (1) administration of a single anticoagulant, most commonly a LMWH or (2) a combination
approach.
If a LMWH is given as a sole agent, it is preferable
to wait 2–4 hours after removal of the Tuohy needle.
The epidural catheter should be left in situ until the
drug levels have reached a safe trough (see Table 10.2).
A combination of unfractionated heparin followed by
LMWH allows the epidural catheter to be removed in a
highly monitored environment in the Recovery Ward
and also permits more controlled reversal of heparin
effects if required in cases of massive post partum
hemorrhage. When unfractionated heparin is given at
the end of Cesarean section, local data suggest that

Chapter 10. Management of anticoagulants at delivery

coagulation profiles are equal to pre-operative values
at 4 hours post-operation, providing there has been no
significant peripartum bleed.

Labor analgesia
It is quite common for a “safe window” to be present
to administer epidural analgesia as a result of antenatal assessment and a plan documented to omit or
appropriately manage heparin once labor starts. For
prophylactic dosing, there is current guidance on timings for insertion of epidural blockade. This should
be utilized regularly as part of ante-natal planning,
and agreed between obstetric, anesthetic, and hematology teams. The demonstrated increase in clearance
of LMWH in the pregnant woman allows a slightly different regime to surgical patients in general. A scheme
allowing neuraxial analgesia 8–10 hours following the
administration of 2500 units and 12 hours following
a dose of 5000 units of dalteparin for example, would
be appropriate (see Table 10.2). The ability to predict an appropriate time is harder when higher doses
of LMWH have been used, or in situations where a
woman presents on full anticoagulant therapy, but in
general this needs to be at least 24 hours after the last
dose. An assessment based on previous anti-Xa levels if available, and on a risk–benefit analysis for each
patient and situation is needed.
If anticoagulation is to commence after delivery,
an adequate gap should be left after the removal of
the epidural catheter before the heparin is administered, usually around 4 hours for LMWH. Whilst most
anesthetists accept a 2-hour interval between catheter
Table 10.2 Guidance for relative timings of heparin and
epidural or spinal block
A. Subcutaneous prophylactic dose
unfractionated heparin
Catheter placement or removal ⬎ 2–4 h after injection
Delay next dose until ⬎ 2 h after catheter insertion
or ⬎4 h after removal
B. Intravenous infusion of unfractionated heparin
Catheter placement ⬎ 4 h after stopping infusion, when aPTT
back to baseline
Restart infusion ⬎2 h after catheter insertion
or ⬎4 h after removal
C. Low molecular weight heparin
Spinal or epidural catheter insertion
⬎ 8 h after last injection – low dose
⬎ 12 h after last injection – intermediate dose
⬎ 24 h after last injection – full anticoagulation
Remove epidural catheter
12 h after any dose
Delay next dose until ⬎2 h after catheter insertion or
subarachnoid injection
or ⬎ 4 h after catheter removal

insertion and heparin injection, as this can easily be
lengthened if the procedure is difficult or traumatic,
4 hours is preferred after removal as the catheter can
pull on established clot and stir up bleeding, even when
insertion had been apparently uneventful.

Full anticoagulation
Women who are fully anticoagulated at the time of
labor and delivery include those with significant risks
of morbidity – a recent venous thromboembolic event,
those normally on long-term warfarin for a variety
of conditions, cardiac disease including mechanical
valves, ischemic heart disease and cardiomyopathy,
and symptomatic homozygous or combination heritable thrombophilias.9–11
The management of these women with a need
for therapeutic levels of anticoagulation in the peripartum period involves balancing the risks of excessive bleeding and difficulties with analgesia and
anesthesia against the risk of thrombosis associated with the underlying condition. This requires
a careful and individualized approach and thorough forward planning, by a team including obstetrician, midwife, anesthetist, and hematologist, and
in full consultation with the patient (table 10.3).
This is particularly important for the small number
of women in whom the usually recommended temporary peripartum reduction in level of anticoagulation may be considered unsafe.
If it has been necessary to use warfarin during the
pregnancy, this should be stopped by 34–36 weeks’
gestation, to allow correction of the fetal coagulopathy, which takes longer than that of the mother,
and minimize the risk of intracranial hemorrhage at
delivery. The most common practice is to replace it
with therapeutic dose LMWH and monitor with antiXa levels as well as clinically. The anti Xa level is
checked at 3–5 days following the first dose and, if in
the desired therapeutic range does not usually need
repeating.
If vaginal delivery is intended, planned induction of labor once the cervix is sufficiently favorable
should be considered. This allows more accurate timing of events and minimizes the risk of delivery whilst
fully anticoagulated. The LMWH should be omitted
on the morning of induction. If prophylaxis needs to
be continued for the day of labor, 5000iu UFH can
be given subcutaneously 8 hourly. If treatment doses
are necessary, an infusion of UFH at 1200 u/h should
be started. APTT should be checked at 4 hours after the

123

Section 3. Thromboembolism and anticoagulation

124

infusion was commenced, aiming for the therapeutic
range as determined by the local laboratory. It should
be noted that the aPTT is less reliable in pregnancy due
to increased levels of factor VIII and heparin binding
proteins.
This regime has been shown to be useful when
stopped 1–6 hours (typically 4) pre-labor, with minimal obstetric or anesthetic complication.12 However,
if this interruption prior to labor is not appropriate,
the UFH infusion should be stopped when cervical
dilatation reaches 5 cm dilatation in primipara, or an
appropriate pre-planned dilatation depending on previous labor experience in multipara. It should also be
stopped if increased unscheduled blood loss is noted
or if urgent Cesarean section is required.13
Protamine sulphate must be immediately available
as should the regime for its use, and 4 units of crossmatched red cells should be available. A protamine sulphate regime is described in Table 10.4.
Graduated compression stockings should be worn
throughout induction and labor and throughout the
inpatient stay, mobility should be encouraged and
hydration ensured.
After delivery, the third stage of labor should be
actively managed with oxytocin, given by intravenous
bolus 10 u followed by an infusion of 40 u over 4 hours.
Perineal tears or episiotomy should be repaired immediately with careful attention to hemostasis and heparin restarted as soon as hemostasis has been secured.
A high index of suspicion must always be maintained when caring for these women with regard
to hemorrhagic complications during pregnancy. The
incidence of antepartum hemorrhage and abruption
is not increased, but is more likely to be significant.
A careful plan for investigation and management of
unexplained abdominal pain or unscheduled bleeding
must be available in the notes.
Other issues which must be considered are the
inadvisability of pudendal block, the potential maternal risks of instrumental delivery if labor occurs
spontaneously and without the lapse of time since
the last dose of heparin, and the importance of
careful postpartum observations for development of
hematomata.
If the woman is on warfarin when labor starts, the
INR must be checked urgently and if greater than 3 or
if a Cesarean section is required, reversed with vitamin K and if more urgent, prothrombin complex concentrate as well. In this instance, a cord blood clotting screen must be taken after delivery as intravenous

Table 10.3 Recommended thromboprophylactic and
anti-hemorrhagic measures
General measures:

Optimize anticoagulation ante-natally
Formulate intrapartum care plan
Multidisciplinary intrapartum care
Senior staff involvement in care
throughout
Aim for vaginal delivery whenever
possible

Specific
thromboprophylactic
measures:

Mobilize during labor
Maintain hydration
Graduated compression stockings
throughout labor and postpartum

Specific
anti-hemorrhagic
measures:

Minimize maternal soft tissue trauma
(e.g. difficult instrumental delivery,
perineal trauma)
Active third stage with intravenous
oxytocics
Prompt management of
complications (e.g. retained
placenta, perineal trauma)

Vitamin K may be required for the neonate. Early
involvement of the neonatologists is essential.

Anesthetic considerations
Analgesia and anesthesia in this group of women is a
major challenge and early involvement of senior anesthetic staff is essential in planning intrapartum care.
Regional analgesia is clearly contraindicated because
of the risk of vertebral canal hematoma. A review
by Loo et al. documented an overall incidence of
0.2–0.3/100 000 following obstetric epidural analgesia, with coagulation abnormalities being identified as
a major risk factor.14

Analgesia for labor
The role of a consistent intrapartum care provider, particularly midwifery, is of high importance – one-toone care from a single carer has been shown to reduce
analgesia requirements and will also reduce the many
anxieties associated with labor in such a situation.13
Other guidance aids decision making, concerning the
need for continuity of anticoagulation, or a window to
reduce the risk of hemorrhage and alter the range of
analgesics available.15
Pharmacological intervention includes inhaled
analgesia in the form of entonox, or administration
of systemic opioids. In a fully anticoagulated woman
intramuscular injection is contraindicated, so bolus or
patient-controlled intravenous administration of opioid analgesia is a suitable option.

Chapter 10. Management of anticoagulants at delivery

Traditionally, opioid therapy used by most delivery suites across the United Kingdom was intermittent bolus administration of intramuscular drugs such
as pethidine. The desire to match the pharmacokinetics of opioids to the time course of the cyclical pain
associated with labor has led to the investigation of
shorter-acting opioids administered in small repeated
doses. Patient-controlled analgesia pumps deliver a
small pre-set dose of opioid via an intravenous cannula.
Fentanyl is a synthetic phenylpiperidine derivative
and is a highly selective mu opioid agonist. When given
by the intravenous route, the dose is effective within
2 to 5 minutes. Fentanyl is highly lipid soluble and
therefore the drug in the plasma rapidly redistributes
to fat-rich areas. This accounts for the short duration
of fentanyl in clinical practice. If large doses are given,
in, for example, a prolonged labor, then the reservoir
for redistribution becomes full. The duration of action
therefore becomes exaggerated from each subsequent
dose, behaving more like morphine. The time to clinical action means that it can be difficult for the laboring
woman to coincide the analgesic action with the peak
of each contraction.
Remifentanil is another synthetic opioid of the
anilidopiperidine group. It has an ultra-short duration of action due to unique metabolism by plasma
esterases. It has a peak onset of 1 to 3 minutes, thus
making the timing relative to contractions easier to
manage. Remifentanil can be given by bolus dose,
or a combination of background infusion with supplemental bolus doses. The pharmacokinetic profile
of remifentanil means that, of all the opioids available, it should most closely match the time profile
of a contraction. Following a recent survey of opioid use in labor, remifentanil is now the most commonly used opioid in the UK for women laboring with a live fetus. The recommended dose varies
between studies, and whilst the presence of a background infusion seems to increase analgesia, it is
often at the expense of increased adverse events.
The narrow therapeutic window means respiratory
depression, and indeed apnea, are significant risks.
The degree of monitoring required is often well in
excess of that which can be offered on a delivery
suite.

Anesthesia for operative delivery
General anesthesia is associated with significant morbidity and even mortality, for example, the risk of fail-

ing to intubate and protect the trachea is increased by
a factor of ten during pregnancy. This must be considered in a risk–benefit assessment for each individual
woman presenting for delivery.
Advice regarding timing of regional anesthesia relative to anticoagulant dose is the same as for regional
analgesia (see Table 10.2), although the use of a single
shot spinal anesthetic technique with a fine gauge needle may be considered earlier than standard guidance
if the woman has other significant anesthetic risk factors.
The anesthetist must be aware of the increased risk
of bleeding at Cesarean section in those women taking
higher doses of LMWH. The introduction of cell salvage machines in obstetric practice to collect autologous blood may help in the management of these
patients.
Other potential complications of regional anesthesia include :
Problems with blood patches
Accidental puncture of the dura mater occurs in 0.5%
to 2.0% of all epidural procedures. Of these women
70% will develop a post-dural puncture headache. If
this headache is severe, it is often treated with an epidural blood patch. Maternal blood is withdrawn aseptically from a suitable vein and introduced via a Tuohy
needle into the epidural space. As this requires further passage of a Tuohy needle, the same time delay
should be introduced after LMWH administration to
avoid bleeding risk. If the woman is fully anticoagulated in the post-delivery period, it should not be done.
If headache is severe, then a hiatus in anticoagulant
therapy may be considered; however, it is unlikely that
the risks of coming off anticoagulant therapy will be
outweighed by the treatment of the headache. Review
of the current literature would suggest that continuing
on prophylactic doses of LMWH does not affect the
ability of the epidural blood patch to treat the headache
effectively.
Cauda equina syndrome
The cauda equina is formed from the terminal nerve
roots of the spinal cord, after the spinal cord has formally terminated around the L2 lumbar disc space.
Compression of the cauda equina presents in a common pattern. Table 10.5 shows the symptoms and signs
classically displayed by patients with a cauda equina
syndrome. Patients on anticoagulant therapy whom
have had regional analgesia or anaesthesia should be

125

Section 3. Thromboembolism and anticoagulation

Table 10.4 Treatment of severe heparin overdosage
Protamine sulphate regime

(1) heparin infusion
25 mg–50 mg after stopping heparin infusion
(1 mg of protamine sulphate neutralizes 80–100 units of heparin)
(2) heparin bolus:
1.00–1.5 mg/100 IU Heparin can be given if 30 minutes have elapsed
0.5–0.7 mg/100 IU Heparin can be given if 30–60 minutes have elapsed
0.25 mg–0.375 mg IU Heparin can be given if 2 hours have elapsed
(3) subcutaneous heparin injection:
1–1.5 mg/100 IU heparin. 25–50 mg by slow IV injection and the remainder by slow IV infusion
over 8–16 hours (or the expected duration of absorption of heparin), or 2 hourly divided doses.
(4) heparin during extracorporeal circulation:
1.5 mg per 100 IU heparin. Sequential APTTs may be needed to calculate correct dosage

Dose in renal/hepatic impairment.

For hepatic impairment, seek further advice
No dose adjustment necessary for renal impairment

Note

Excessive doses of protamine can have an anticoagulant effect

Table 10.5 The cauda equina syndrome
Low back pain
Bilateral, occasionally unilateral, sciatica
Perineal numbness (saddle numbness)
Bladder dysfunction
Bowel dysfunction
Variable lower limb weakness and sensory loss

carefully monitored in the post delivery period. Imaging of a potential lesion is usually undertaken with
MRI scanning. Cauda equina syndrome is a medical emergency, and in the case of hematoma, requires
urgent surgical opinion to schedule evacuation of the
clot, as delay is likely to increase the risk of residual
neurological dysfunction.

Thrombolysis and bypass

126

It has been traditional to consider pregnancy and the
puerperium as contraindications to the use of thrombolytic agents. However, in the situations where they
are needed, the life of the mother is likely to be at
high risk, as in cardiac compromise following massive
central pulmonary embolus (PE), or acute myocardial
infarction. In this situation, the balance of risks needs
careful consideration, but if thrombolysis is likely to be
life-saving, it should not be delayed as if hemorrhage
occurs it can be managed with a combination of surgical and hematological techniques. Clearly for massive
central PE targeted injection via catheter is the ideal,
if appropriate staffs are available, but there are reports
of successful thrombolysis without this.16 Life-saving

treatment should not be delayed or withheld for theoretical risk which cannot be substantiated. Senior and
experienced decision making is essential – for example
if a Cesarean section is required as part of resuscitation
in a woman collapsed from PE or MI, the timing and
management of thrombolytic agents would need to be
discussed carefully but swiftly.
The risk–benefit balance when cardio-pulmonary
bypass is contemplated in pregnancy is different.
Clearly, the seriousness of the situation giving rise
to the need for bypass, usually associated with cardiac surgery, combined with prolonged and significant
anticoagulation means that to facilitate optimization
of both maternal condition for anesthesia and surgery
and post-operative recovery, emptying of the uterus
prior to bypass is advisable. There are case reports
of successful prolongation of pregnancy in this situation, but also numerous reports published and unpublished of significant retroplacental bleeding, inability
to maintain maternal blood pressure or oxygenation,
and difficulty in maintaining good peri-operative conditions associated with attempting to continue pregnancy. Clearly, in a viable fetus these risks are unacceptable and delivery should be expedited. Prior to
this, serious consideration should be given to medical
or surgical termination.

Summary
Management of women with any degree of anticoagulation in the peripartum period is often challenging. However, attention to detail, careful planning and
documentation, departmental education and involving both appropriate disciplines as well as the patient,
should allow optimal management to be achieved.

Chapter 10. Management of anticoagulants at delivery

References
1. Royal College of Obstetricians and Gynaecologists.
Reducing the risk of Thrombosis and Embolism
during pregnancy and the puerperium. Guideline
no.37. London, RCOG Press, 2009.
2. Lewis G (Ed.). Saving Mothers’ Lives : reviewing
maternal deaths to make motherhood safer
2003–2005. The Seventh Report on Confidential
Enquiries into Maternal Deaths in the United
Kingdom. London, CEMACH, 2007.
3. CLASP (Collaborative Low dose Aspirin Study in
Pregnancy) Collaborative Group. CLASP: a
randomised trial of low-dose aspirin for the
prevention and treatment of pre-eclampsia among
9364 pregnant women. Lancet 1994; 343: 619–629.
4. James AH, Abel DE, Braucazio LR. Anticoagulants in
pregnancy. Obstetrical and Gynaecological Survey
2005; 61: 59–61.
5. Wysowski DK, Talarico L, Bacsanyi J, Botstein P.
Spinal and epidural haematoma and low molecular
weight heparin. New England Journal of Medicine 1998;
338: 1774–1775.
6. Scottish Intercollegiate Guidelines Network. Section 7
Spinal and epidural blocks; Section 9 Pregnancy and
puerperium. In: Prophylaxis of Venous
Thromboembolism. Publication No. 62. Edinburgh,
SIGN 2002; 24–26; 30–35.
7. Horlocker TT, Wedel DJ, Benzon H et al. Regional
anesthesia in the anticoagulated patient: defining the
risks. (The Second ASRA Consensus Conference on
Neuraxial Anesthesia and Anticoagulation.) Regional
Anesthesia and Pain Medicine 2003; 23: 172–197
8. Stafford Smith M. Impaired haemostasis and regional
anaesthesia. Canadian Journal of Anesthesia 1996; 43:
R129–R141.

9. Royal College of Obstetricians and Gynaecologists.
Thromboembolic disease in pregnancy and the
puerperium : acute management. Green-top guideline
28, 2007.
10. Asghar F, Bowman P. A Clinical Approach to the
Management of Thrombosis in Obstetrics Part 2 :
diagnosis and management of venous
thromboembolism. The Obstetrician and Gynaecologist
2007; 9: 3–8.
11. Gelson E, Johnson M, Gatzoulis M, Uebing A.
Cardiac Disease in Pregnancy Part 2 : Acquired Heart
Disease. The Obstetrician and Gynaecologist 2007; 9:
83–87.
12. Austin SK, Lambert J, Peebles D, Cohen H. Managing
peri-delivery anticoagulation in women on therapeutic
dose low molecular weight heparin : a role for
unfractionated heparin? Journal of Thrombosis and
Haemostasis 2007; 5: P-S-622.
13. Akkad A, Oppenheimer C, Mushambi M, Pavord S.
Intrapartum care for women on full anticoagulation.
International Journal of Obstetric Anaesthesia 2003; 12:
188–192.
14. Loo CC, Dahlgren G, Irestedt L. Neurological
complications of obstetric regional analgesia.
International Journal of Obstetric Anesthesia 2000; 9:
99–124.
15. Bates SM, Greer IA, Pabinger I, Sofaer S, Hirsh J.
Venous Thromboembolism, Thrombophilias,
Anti-thrombotic therapy and Pregnancy. American
College of Chest Physicians Evidence-Based Clinical
Practice Guidelines (8th edn). Chest 2008; 133:
844S–886S.
16. Leonhardt G, Gaul C, Nietsch HH et al. Thrombolytic
therapy in pregnancy. Journal of Thrombosis and
Thrombolysis 2006; 21: 271–276.

127

Section

Thrombophilia and fetal loss

Section 4
Chapter

Thrombophilia and fetal loss

Antiphospholipid syndrome
Sue Pavord, Bethan Myers, and Beverley Hunt

Introduction
Antiphospholipid syndrome (APS) is an autoimmune disorder characterized by vascular thrombosis
and/or obstetric morbidity in the presence of persistent antiphospholipid antibodies (aPL)1 namely, lupus
anticoagulant antibodies (LAC), anti-cardiolipin antibodies (aCL) and/or anti-␤2 -glycoprotein I antibodies. There are numerous potential complications for
pregnancy but with optimal management, good maternal health and a live birth rate of 80%–90% can be
achieved.
The syndrome produces a spectrum of disease,
both in terms of clinical manifestations and the
presence of other autoimmune conditions. Arterial,
venous, or small vessel thrombosis may occur, there is
an array of adverse obstetric outcomes and a number of
additional clinical features may be present, involving
organs such as the heart, skin, and central nervous system. The disease is classified as primary (PAPS) when
it occurs in the absence of any features of other autoimmune disease, and secondary where other autoimmune disease is present (SAPS). Predominantly, this
is systemic lupus erythematosus (SLE), but other conditions such as inflammatory bowel disease may be
involved.

Prevalence
The syndrome occurs most commonly in young to
middle-aged adults, with a mean age of onset of
31years. Women are more frequently affected, with a
female to male ratio of 5:1 which is even higher in
SAPS associated with SLE. There is no defined racial
predominance for APS, although an increased incidence of SLE occurs in African Americans and the Hispanic population. Among patients with SLE, the prevalence of aPL is 15% to 35%, but only around half of

these cases will have clinical features of antiphospholipid syndrome.

Pathophysiology and etiology
Different pathological mechanisms may be responsible
for the varying clinical manifestations. Recurrence of
complications often follows a similar pattern of disease
and recurrent thrombosis usually occurs in the same
vascular field, although this is not always the rule.
It was originally thought that aPL were directed
against negatively charged phospholipid, but it is now
clear that they target plasma proteins with affinity
for these anionic phospholipids. There is concordance
between the LAC, aCL, and anti-␤2 -GP I, antibodies;
however, they are not identical and some LAC antibodies react with phospholipids other than cardiolipin and
proteins other than ␤2 -GP I, whereas some aCL and
anti-␤2-GP I antibodies have no LAC activity. In general, LAC antibodies are more specific for the diagnosis of APS, but there is no association with particular
clinical manifestations and antibody type.
The main antigens involved are ␤2 -glycoprotein 1
(␤2 GPI) and prothrombin, although many more antigenic targets have been described. Antibodies to the
␤2 GPI are persistent and associated with thrombotic
complications, whereas those independent of ␤2 GPI
tend to be present only transiently, in association with
infectious diseases or drugs.

Procoagulant effects
␤2 GPI is a multifunctional apolipoprotein, which
contributes to the regulation of hemostasis as well
as other physiological processes. Not surprisingly
therefore, ␤2 GPI-dependent antibodies have been
associated with a number of different biological
effects (Table 11.1). These include direct cellular effects

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

131

Section 4. Thrombophilia and fetal loss

Table 11.1 Procoagulant effects of ␤2 GPI-dependent aPL

Up-regulation of the tissue factor pathway
Inhibition of the activated protein C pathway
Inhibition of antithrombin activity
Inhibition of fibrinolysis
Activation of endothelial cells
Enhanced expression of adhesion molecules by
endothelial cells with increased binding of
leukocytes
Activation and degranulation of neutrophilis
Potentiation of platelet activation
Enhanced platelet aggregation
Displacement of annexin V from cell membranes
caused by bound ␤2 GPI –antibody complexes, with
affinity for both anionic phospholipid expressed on
the surface of activated cells and heparin sulphatecontaining structures on non-activated cells. The binding of ␤2 GPI to anionic structures, through domain
5, induces the expression of new cryptic epitopes in
domain 1 and may increase the antigenic density,
two events that seem to be pivotal for the antibody
binding. Studies show that dimerization of ␤2 GPI
by anti-␤2 GPI antibodies causes a conformational
change in the molecule increasing its affinity for phospholipids by 100-fold. ␤2 GPI can bind to the lowdensity lipoprotein receptor, ApoER2, on the surface
of platelets and thus mediate platelet activation, with
increased thromboxane synthesis and platelet aggregation. In vitro endothelial cells and monocytes can also
be activated by aPL and ␤2 GPI binding, resulting in
tissue factor expression. In addition, in vitro studies
have shown some aPL cause interference with hemostatic factors such as IX, X, and XII, resistance to activated protein C, and a reduction in fibrinolysis from
antiplasmin or anti-tissue-type plasminogen activator
(tPA) activity. At this time, there is no clarity as to how
aPL cause thrombosis, and it is the subject of much
research.

Obstetric morbidity
132

The pathophysiological mechanisms underlying fetal
loss or morbidity also appear to be multiple. Due to
the wide spectrum of manifestations and heteroge-

neous findings in placental tissue in these patients, it
is unclear whether one or several aPL subgroups are
responsible for the varying phenotypes and whether
concurrent, aPL-independent genetic and environmental factors affecting the maternal–fetal interface, influence the potential pathogenicity of these
antibodies.
Early histological studies demonstrated decidual
vasculopathy and placental thrombosis. Displacement
of annexin V from trophoblasts contributes to a procoagulant state through acceleration of coagulation reactions. An elegant mouse model has demonstrated activation of complement through the classical pathway,
with consequent influx of inflammatory cells into tissues, mediating placental injury, and leading to fetal
loss and growth restriction. In this model, heparin prevented pregnancy loss by blocking activation of complement, rather than primarily via an anticoagulant
effect.2
Direct trophoblastic damage by aPL, independent
of mechanisms involving thrombosis and complement activation, has also been demonstrated recently.
Interaction of aPL with ␤2 GPI, exposed during trophoblast syncytium formation, has been shown to
cause inhibition of the intercytotrophoblast fusion
process, gonadotrophin secretion, and trophoblast
invasiveness. This mechanism has been hypothesized
to contribute to early pregnancy loss. There is evidence
of a significant reduction in intradecidual endovascular trophoblast invasion on analysis of the products of
conception (first-trimester failure) from APS patients.
The factors that determine whether aPL induce a
thrombotic or non-thrombotic disease phenotype in
the placenta are not known. It is likely that interplay
between patient background traits and distinct aPL
subgroups determines disease manifestation.

Clinical features
International consensus criteria for the classification
of definite APS were initially published in 19993 and
updated in 20061 (Table 11.2).

Thrombosis
Thrombosis is the most common presenting feature of
APS.4 Thrombosis may occur in both the venous and
arterial circulation as well as the microvasculature. It
can involve vascular beds that are infrequently affected
by other prothrombotic states and is independent of
atherosclerotic vascular disease.

Chapter 11. Antiphospholipid syndrome

Table 11.2 Summary of the revised classification criteria for the Antiphospholipid Syndrome (APS)1

APS is diagnosed if at least one clinical and one laboratory criteria are met (although not if there is less than 12
weeks or more than 5 years between the positive aPL test and the clinical manifestation).
Clinical criteria
1. Vascular thrombosis
One or more clinical episodes of arterial, venous, or microvascular thrombosis occurring in any tissue or
organ (superficial venous thrombosis is not included)
2. Pregnancy morbidity
(a) One or more unexplained deaths of a morphologically normal fetus at, or beyond, 10 weeks’ gestation
or
(b) One or more premature births of a morphologically normal neonate before 34 weeks’ gestation
because of eclampsia, severe pre-eclampsia or recognized features of placental insufficiency∗ .
(c) Three or more unexplained consecutive spontaneous abortions before the 10th week of gestation,
with maternal anatomic or hormonal abnormalities and paternal and maternal chromosomal causes
excluded.
Laboratory criteria
1. LAC present on two or more occasions at least 12 weeks apart detected according to the guidelines of the
International Society on Thrombosis and Haemostasis (10)
2. aCL of IgG and/or IgM isotype in serum or plasma, present in medium or high titer (i.e. ⬎40 GPL or MPL, or
⬎the 99th percentile), on two or more occasions, at least 12 weeks apart, measured by a standardized
ELISA
3. Anti-␤2 GP-I antibody of IgG and/or IgM isotype in serum or plasma (in titer ⬎the 99th percentile), present
on two or more occasions, at least 12 weeks apart, measured by standardized ELISA
∗

Generally accepted features of placental insufficiency include: (i) abnormal or non-reassuring fetal surveillance test(s), e.g. a non-reactive
non-stress test, suggestive of fetal hypoxemia, (ii) abnormal Doppler flow velocimetry waveform analysis suggestive of fetal hypoxemia,
e.g. absent end-diastolic flow in the umbilical artery, (iii) oligohydramnios, e.g. an amniotic fluid index of 5 cm or less, or (iv) a post-natal
birth weight less than the 10th percentile for the gestational age.

Venous thrombosis
aPL are found in approximately 2% of patients presenting with acute venous thromboembolism. Often, the
venous thrombosis occurs in an unusual site such as
the cerebral, retinal, splanchnic or axillary, and subclavian veins and APS can account for up to 70% of
such presentations. Venous thromboembolism, especially deep venous thrombosis of the legs, has been
shown to occur in around 30%–50% of patients with
APS during an average follow-up of less than 6 years.5
Following a first episode, the risk for future venous
thrombosis increases significantly.

Arterial thrombosis
The most common site of arterial thrombosis is the
central nervous system, with strokes and transient
ischemic attacks accounting for 50% of the arterial
events seen with APS. Myocardial infarction accounts

for around 20% and other vascular beds may be
involved including those of the lungs, retina, gastrointestinal tract, spleen, and extremities. In many cases
the event is otherwise unexplained, with no other identifiable risk factors for arterial disease, such as smoking, diabetes, or hypertension.

Obstetric complications
Complications during pregnancy, in addition to
maternal thrombosis, include recurrent spontaneous
abortions in the first trimester as well as adverse
outcomes occurring late in pregnancy. However, there
are women with aPL who have no problems at all in
pregnancy.

Early pregnancy loss
Pregnancy loss is one of the leading problems in
women’s health issues. Approximately one-third of all

133

Section 4. Thrombophilia and fetal loss

conceptions and 15% of clinically recognized pregnancies (⬍6 wk of gestation) fail to result in a live birth.
Of women, 5% experience two or more losses and 1%–
2% suffer with three or more. Up to half of the cases
remain unexplained after gynecological, hormonal,
and karyotypic analyses. Of the women who have
recurrent pregnancy loss, defined as three or more first
trimester miscarriages, 10% to 20% have detectable
aPL.6 These women potentially have a 90% risk of further fetal loss if left untreated.7 The diagnostic criteria for APS suggest that evaluation should begin
after the third consecutive early miscarriage, defined
by less than 10 weeks’ gestation (Table 11.2). However,
in practice, evaluation after two early miscarriages is
often initiated at the discretion of the physician.

Late complications of aPL in pregnancy
Complications occurring late in pregnancy relate to
placental dysfunction caused by aPL. The manifestations include pre-eclampsia, prematurity, fetal distress, intrauterine growth restriction, and fetal death.
Preterm delivery is associated with premature rupture
of the membranes or pre-eclampsia. The median rate
of gestational hypertension or pre-eclampsia is 30%–
50% in untreated women with previously diagnosed
APS but falls to 10% with effective management. In
contrast, aPL are not found in a significantly higher
proportion of general obstetric patients presenting with pre-eclampsia.8 HELLP syndrome (hemolysis, elevated liver enzymes, and low platelets) may
occur, being associated with pre-eclampsia/eclampsia
in most cases and seems to occur earlier than in
women without APS, often in the second trimester.

Other clinical manifestations
In addition to thrombosis and obstetric morbidity,
there are a number of additional clinical manifestations which are not included in the official definition of
APS. These include abnormalities of skin (particularly
livedo reticularis), cardiac valves, central nervous system, kidneys, and hematological disturbances such as
thrombocytopenia and a positive direct Coombs test
with occasional cases of clinical hemolytic anemia.

Thrombocytopenia
134

Many patients with APS have thrombocytopenia
(platelets⬍100 × 109 /L). The pathogenic antibodies
are directed towards epitopes on platelet membrane

glycoproteins and are distinct from antiphospholipid
antibodies. Conversely, aPL are found in approximately 25% of patients with chronic autoimmune
thrombocytopenia. They do not confer a different clinical phenotype initially, but the persistence of lupus
anticoagulant in these patients has been found to be
an important risk factor for subsequent development
of APS.9

CNS affects
A variety of neurological manifestations may occur,
mostly secondary to cerebrovascular infarcts. The clinical features depend upon the caliber and location of
the vessels occluded and include multi-infarct dementia, psychomotor agitation and insomnia, movement
disorders such as chorea, dystonia, oral dyskinesias
and speech impairment, transverse myelitis, seizures,
migraine, psychosis, and optic neuritis.

Valve defects
Up to 30% of patients with APS have minor valvular abnormalities, which usually do not cause hemodynamic disturbance. Non-bacterial thrombotic endocarditis (Libman–Sacks endocarditis) is a rare disorder
characterized by sterile, thrombotic vegetations of the
heart valves and can occur rarely in APS. These thrombotic lesions carry significant embolic potential.

Catastrophic antiphospholipid
syndrome (CAPS)
This term defines a severe accelerated form of APS
that results in multi-organ failure from widespread
thromboses, which are usually microvascular rather
than large vessel occlusions. The pathogenesis
appears dependent on a multi-hit phenomenon, with
infection, trauma or surgery, drug administration,
or warfarin withdrawal exacerbating an already
procoagulable state. In 50% of cases no triggering
factor is identified and in some it may be relatively
minor such as a biopsy. Around one-quarter suffer
with disseminated intravascular coagulation (DIC),
contributing to “thrombotic storm” and end organ
damage. Severe thrombocytopenia is common. Acute
adult respiratory distress syndrome (ARDS) occurs in
one-third of patients and death in around 50%, mainly
from cardiac or respiratory failure, despite treatment
with anticoagulation and plasma exchange.

Chapter 11. Antiphospholipid syndrome

Laboratory evaluation
Limitations to laboratory testing include lack of laboratory standardization for aPL and the heterogenous nature of the antibodies resulting in low specificity of the assays. Many healthy individuals can have
aPL without thrombosis or obstetric morbidity; indeed
aPL are found in 3%–5% of the normal population.
They are often transient and associated with infection
or drugs. The clinical importance of these aPL is uncertain. To satisfy the APS laboratory classification criteria, a patient has to be persistently positive for either
one of the assays – anticardiolipin antibody, lupus anticoagulant, or ␤2 GPI antibody, for at least 12 weeks.
Testing should occur away from the acute event.

Lupus anticoagulant (LAC)
These antibodies are detected by their ability to prolong phospholipid-dependent coagulation reactions,
not corrected by mixing patient and normal plasmas.
As aPL showing LAC activity are heterogeneous, it
is recommended that at least two methods are performed; APTT and a direct antiXa assay such as the
dilute Russell viper venom assay (dRVVT). It should
be confirmed that the anticoagulant is directed against
protein bound to negatively charged phospholipids.
The International Society for Thrombosis and
Haemostasis has identified the following criteria for
the confirmation of a LAC.10
(1) Prolongation of a phospholipid-dependent
clotting assay
(2) Evidence of an inhibitor demonstrated by mixing
studies
(3) Confirmation of the phospholipid-dependent
nature of the inhibitor (platelet or other
phospholipid neutralization procedure).
Most LACs are directed against either ␤2 GPI or prothrombin and recently methods to distinguish those
associated with anti ␤2 GPI have been developed but
whilst these are highly specific, their sensitivity is low.
LAC has been shown to be the most relevant assay
in relation to vascular events and obstetric morbidity. The odds ratios for thrombotic risk ranges from 5
to 16.

Anticardiolipin antibodies (aCL)
Anticardiolipin antibodies are measured by ELISA,
although, again, concordance amongst laboratories is

poor and it is difficult to distinguish between the significant antibodies associated with ␤2 GPI from those
bound to other plasma proteins or directly bound
to cardiolipin. The correlation between aCL titer and
thrombotic risk is well established, with the IgG subtype having stronger association than IgM or IgA.11
The revised classification criteria for APS uses an IgG
aCL cut off of 40 U. Low levels of aCL, although statistically abnormal, may not be associated with a significant risk of thrombosis and in a systematic review of
the literature, Galli et al. observed no correlation with
venous thrombosis and only a weak correlation with
arterial thrombosis.12

Anti-beta 2 glycoprotein 1 antibodies
Anti-␤2 GPI antibodies are now included in the revised
criteria for the diagnosis of APS1 (Table 11.2). They
show better correlation with thrombosis than aCL
but there is a high false-positive rate. Recently, new
guidelines have been published for the performance of
an anti-␤2 GPI antibody ELISA, which might improve
standardization of the assay. However, the specificity
remains low as there are non-pathogenic antibodies that bind ␤2 GPI. Indeed, of the five domains of
␤2 GPI involved, only those antibodies directed against
domain 1 correlate with thromboembolic complications, with an odds ratio of around 18.

Principles of management
Individual treatment strategies for the management of
the antiphospholipid syndrome in pregnancy in part
depends on the assessment of a number of different
factors. These include:
r history of prior thrombosis;
r whether the thrombotic event was provoked or
spontaneous;
r whether the thrombotic event was venous or
arterial;
r history of obstetric morbidity alone;
r evidence of any organ damage;
r the presence of SLE or other autoimmune disease;
r other maternal risk factors, such as obesity and
maternal age.

Background
The first treatment used and studied for pregnant
patients with APS, was a combination of corticosteroids and low dose aspirin. Low-dose aspirin is

135

Section 4. Thrombophilia and fetal loss

known to be safe in pregnancy, in the first and second
trimesters, and also recognized to reduce risk of preeclampsia.
Corticosteroids were an obvious choice as an
immunosuppressant to suppress the antibodies
present. Small, early studies were encouraging. Subsequent studies comparing heparin and prednisolone
concluded that low dose heparin was preferable,
since, although effective, steroids induced significant
maternal morbidity, and more premature deliveries.13
Heparin (either unfractionated or low molecular
weight) is the standard anticoagulant in pregnancy for
prophylaxis and treatment of VTE. With improved
understanding of the mechanisms of action of
heparins and the pathophysiology of APS, it is a
logical drug of choice in this condition. In addition
to anticoagulant activity, it has anti-inflammatory
and anti-complement effects, both of which may
be involved in APS pathogenesis. In vitro heparin
also appears to enhance trophoblast development,
apparently limiting aPL attack on trophoblasts. Two
systematic reviews of small studies recommended
a combination of aspirin and heparin to reduce
fetal loss, concluding that this regime may reduce
pregnancy loss by 54%.14,15 Some authors, however,
question the role of pharmacological treatment in
improved live birth rate, as some studies showed no
difference between treatment and placebo arms, in
low-risk patients. Problems in interpretation are due
to the small size of studies, variable entry criteria, lack
of placebo arms, and absence of blinding.

r Review detailed medical and obstetric history.
r Document and confirm persistant aPL, assess
renal function and presence of thrombocytopenia
and/or anemia.
r Optimize the patient’s clinical state and
pharmacological treatment before pregnancy.
Advise postponing pregnancy if a thrombotic
event has occurred within the last 6 months, SLE
has been active or hypertension uncontrolled.
r Assess individual additional risk factors such as
obesity and maternal age and give a clear
indication to the patient regarding the degree of
risk for both thrombosis and obstetric
complications.
r Assess for the presence of anti-Ro or La
antibodies, even if no evidence of SLE. These
antibodies are associated with a 2% risk of
complete heart-block in the fetus and up to a 10%
risk of neonatal lupus. If found, fetal cardiology
assessment should be offered and any pregnancy
affected by complete heart block should be
managed by a specialized centre where there is a
pediatric cardiologist to manage the neonate.
r Provide contact information for prompt, early
referral at the onset of pregnancy and ensure a
clear understanding of the need to substitute
heparin for warfarin at the time of the first missed
period. Ideally, the woman can be provided with a
supply of LMW heparin and lessons in
self-injection so should they get pregnant, they
can switch quickly from warfarin.

Management of thrombosis
Pre-pregnancy management

136

Pre-pregnancy counseling should be offered, taking all
factors into consideration. Where necessary, recommendations should be made to improve general health
and reduce risk before a pregnancy is undertaken, such
as the need for weight loss, to wait at least 6 months following an acute thrombotic event or until SLE has been
quiescent for 3 months. There may be circumstances
where pregnancy should be actively discouraged, for
example, if pulmonary hypertension is present the
risk of maternal death is estimated at greater than
35%.
During this review, a clear proposed plan for pregnancy management should be outlined, both verbally and in writing. The following issues need to be
addressed:

The immediate management of thrombosis should be
the same in those with or without the antiphospholipid syndrome. It may, indeed, be the first indication
of underlying APS. Samples should not be sent for
thrombophilia testing at the time of an acute event. The
results may be misleading for several reasons, particularly during pregnancy, and the laboratory results do
not change initial management.

Management of women with
antiphospholipid antibodies and a
previous thrombotic event
These women may be on long-term anticoagulants,
although this depends on the circumstances of the
thrombotic event; if it was a single venous event with

Chapter 11. Antiphospholipid syndrome

a clear temporary provoking factor, a limited duration
of anticoagulation may have been given.
In the former case, a change from warfarin to heparin can be made once a pregnancy test is positive. This
requires the woman to be well motivated to check carefully in order to ensure that the substitution occurs
before 6 weeks’ gestation, to minimize risk of teratogenicity of warfarin. In the UK, it is usual to recommend a pregnancy test on the first or second day of
a missed period and then to switch to low molecular weight heparin (LMWH) either at intermediate
or full therapeutic doses, depending on the extent of
risk when all factors are taken into account. Low dose
aspirin, at 75 mg daily, is also given.
The patient requires regular, frequent review
throughout the pregnancy, and scans to assess fetal
growth should be performed throughout the second
half (monthly or more often as indicated). Uterine
artery Doppler flow measurements (between 20 and
24 weeks onwards) are a further useful tool to assess
platelet dysfunction. The absence of bilateral notching
is a good prognostic sign for fetal outcome.
Post-natally, the patient may be re-established on
warfarin or continued on LMWH for at least 6 weeks.

Management of women with multiple
previous venous events, or venous plus
arterial events
This group of women will likely be on long-term warfarin, possibly at a higher INR range (3–4s), and are
at very high risk during pregnancy. In this group,
pre-pregnancy counseling is particularly important, to
assess the extent of risk on an individual basis, and this
risk clearly conveyed to the patient and partner prior to
embarking on pregnancy. Low molecular weight heparin should be substituted for warfarin before 6 weeks’
gestation, in intermediate or full dose. Some groups
monitor anti-Xa levels, although they are not clearly
predictive of anti-thrombotic effect. Other management is as above, i.e. low dose aspirin, fetal growth
assessment by regular ultrasound scanning, and warfarin reintroduced post-natally.

Management of women with APS and
pregnancy morbidity
A Cochrane systematic review in 2005 assessed 13 trials published (between 1991 and 1999) on recurrent

pregnancy loss associated with aPL. They commented
that the quality of the studies was not good, which limited useful conclusions.

Aspirin
In obstetric patients, low dose aspirin has been used
to improve pregnancy outcome in those with hypertension, pre-eclampsia, preterm birth, and intrauterine growth restriction. The Cochrane review and
meta-analysis summarized the studies with aspirin.16
Three trials with aspirin alone showed no significant reduction in pregnancy loss; two studies using
unfractionated heparin and aspirin showed a significant improvement in fetal outcome compared with
aspirin alone; but in a further randomised controlled
trial, whilst high success rates were achieved with low
dose aspirin, the addition of LMWH did not provide further benefit.17 This latter study has been criticized as the laboratory criteria for APS were not
met.
However, these studies were done at a time when
LMWH was just being introduced, and it is difficult
to draw firm conclusions from this collection of data.
Subsequently obstetric hematology groups, have accumulated a large volume of experience with LMWH,
and it is considered safer and as effective as unfractionated heparin, and more convenient with a oncedaily dosing regimen. The risk of osteoporosis which
is as high as 2% with unfractionated heparin, is rarely
described with LMWH and prophylactic dose LMWH
and low dose aspirin has become standard practice.
Improved outcomes for women with previous late fetal
loss or early delivery due to placental insufficiency
have been confirmed and a recent metaanalysis supports the efficacy of this approach for recurrent pregnancy loss.18

Ultrasonography
In patients with poor obstetric history, pre-eclampsia
or evidence of fetal growth restriction, fetal growth
scans every 4 weeks, from 20 weeks is recommended, in addition to pharmacological treatments.
Studies have shown uterine artery Doppler to be
valuable in predicting placental dysfunction, i.e.
pre-eclampsia and intrauterine growth restriction
and the discovery of bilateral uterine artery notching
can allow the obstetrician to monitor the pregnancy
more closely. Multivariate analysis of data on 100
pregnancies demonstrated that a notched uterine
artery at the second trimester was the only predictor

137

Section 4. Thrombophilia and fetal loss

for adverse pregnancy outcome. Uterine artery
Doppler assessment should be performed with the
fetal scan at 20 weeks and 24 weeks. Le Thi Huong
et al.19 showed the predictive value of the umbilical
artery Doppler ultrasound examination for late pregnancy outcome, together with clinical examination
and laboratory tests in women with SLE and/or
APS.

with success, although several studies have not
demonstrated its benefit.20 It may also be useful in
APS cases who have additional indications for its
use, for example, in auto-immune
thrombocytopenia or Guillain–Barré syndrome.
(c) Successful use of plasma exchange to remove
antibodies temporarily has been reported, in high
risk pregnancies where plasma exchange was
administered.

Table 11.3 Summary of pharmacological management of APS
in pregnancy

Clinical feature

Management

APS with prior fetal
death or recurrent
pregnancy loss

Aspirin 75 mg od
LMWH prophylactic
dosage
Doppler ultrasound
for fetal assessment

APS with prior venous
or arterial thrombosis

Low molecular weight
heparin at
intermediate or
therapeutic dosage,
plus aspirin

Antiphospholipid
antibodies without
clinical features and
healthy previous
pregnancies

No treatment, or low
dose aspirin

Primigravida with
isolated aPL

Low dose aspirin and
fetal monitoring

APS with recurrent
thrombotic events

Full therapeutic dose
LMWH; consider
warfarin

Other treatment modalities
A summary of pharmacological management of APS
in pregnancy is provided in Table 11.3.
If there is continued pregnancy loss despite
prophylactic-dose heparin and low dose aspirin, the
following options should be considered.

138

(a) Increasing the dose of heparin to therapeutic
levels may be effective, although there are no trials
that have demonstrated this.
(b) There are anecdotal case reports of use of
intravenous immunoglobulin in refractory cases

Management of thrombocytopenia
associated with APS in pregnancy
This may be due to pre-eclampsia, HELLP syndrome
or worsening maternal idiopathic thrombocytopenic
purpura (ITP). This should be managed in the same
way as those complications occurring without APS
(see Chapters 4,17, and 18).

Management dilemmas
(1) Management of women who have isolated aPL,
with no prior pregnancy loss or thrombo-embolic
phenomena, do not generally merit ante-natal
pharmacological treatment, although low dose
aspirin is often used. The best predictor of
maternal and fetal outcome in APS pregnancies is
the previous obstetric history. Mothers with a
previously normal obstetric history despite aPL
can be reassured that any future pregnancy has a
low risk of complications.
(2) For women in an ultra-high risk group, such as
those with multiple unprovoked venous and
arterial event, or patients who develop a new
thrombotic event during pregnancy despite
anticoagulation, consideration should be given for
warfarin usage from the second trimester, or even
all trimesters. The risks of warfarin to the fetus
must be explained – including the teratogenic risk
in the first trimester and the ongoing risk of fetal
loss, hemorrhage and subtle neurological changes,
which have been described after its use in the
second and third trimester.
(3) Where in vitro fertilization (IVF) or other assisted
reproductive techniques are planned, LMWH
should be substituted for warfarin at the time at
which assisted reproductive procedure is
performed, i.e. at the time of egg transfer; if the
woman is not on anticoagulation prior to IVF,

Chapter 11. Antiphospholipid syndrome

then prophylactic LMWH and aspirin should be
used.
(4) For the rare seronegative APS, or SNAPS, where
typical clinical features occur in the absence of
measurable standard antibodies – expert clinical
judgment is required to make this diagnosis and
to determine need for treatment.
(5) Management of women with SLE in pregnancy is
one of the few indications for the use of
glucocorticoids during the pregnancy. SLE may
flare during pregnancy and increasing or starting
small doses of prednisolone is appropriate.
Hydroxychloroquine and azathioprine, standard
drugs for the management of SLE, are safe in
pregnancy and should be continued if the disease
is stable. Stopping such medications may lead to a
flare of SLE which could be harmful to mother
and fetus.
(6) Chorea gravidarum is a rare complication in
pregnancy and may be associated with primary or
secondary APS. It is thought to be due to
development of antibodies against components of

the basal ganglia, or rarely due to infarction of this
area. It is usually self-limiting and resolves
following pregnancy, although may recur in
subsequent pregnancies. When severe, a variety of
treatments have been described, including low
dose haloperidol, steroids, anticoagulants,
antiplatelet medication, or a combination of
treatments.21

Neonatal issues
Neonatal APS has been described, although the existence of this syndrome has not been fully accepted.
It is a rare occurrence, characterized by neonatal
thrombosis thought to be due to the transplacental
passage of maternal aPL.22 Ischemic stroke is the main
event described. In comparison to a high incidence
of thrombotic and obstetric complications in women
with APS, the aPL-associated thrombotic events in
neonates are extremely rare. There is no benefit from
routine screening for aPL in neonates born to mothers
with APS.

139

Section 4. Thrombophilia and fetal loss

References
1. Miyakis S, Lockshin MD, Atsumi T et al. International
consensus statement on an update of the classification
criteria for definite antiphospholipid syndrome (APS).
Journal of Thrombosis and Haemostasis 2006; 4:
295–306.
2. Girardi G, Redecha P, Salmon JE. Heparin prevents
antiphospholipid antibody-induced fetal loss by
inhibiting complement activation. Nature Med. 2004;
10: 1222–1226.
3. Wilson WA, Gharavi AE, Koike T et al. International
consensus statement on preliminary classification
criteria for definite antiphospholipid syndrome: report
of an international workshop. Arthritis and
Rheumatism 1999; 42: 1309–1311.
4. Cervera R, Piette JC, Font J et al. Euro-Phospholipids
Project Group. Antiphospholipid syndrome: clinical
and immunologic manifestations and patterns of
disease expression in a cohort of 1,000 patients.
Arthritis and Rheumatism 2002; 46: 1019–1027.
5. Asherson RA, Khamashta MA, Ordi-Ros J et al. The
“primary” antiphospholipid syndrome: major clinical
and serological features. Medicine (Baltimore) 1989;
68: 366–374.
6. Yetman DL, Kutteh WH. Antiphospholipid antibody
panels and recurrent pregnancy loss: prevalence of
anticardiolipin antibodies compared with other
antiphospholipid antibodies. Fertility and Sterility
1996; 66: 540–546.
7. Rai RS, Clifford K, Cohen H, Regan L. High
prospective fetal loss rate in untreated pregnancies of
women with recurrent miscarriage and
antiphospholipid antibodies. Human Reproduction
1995; 10: 3301–3304.
8. Dreyfus M, Hedelin G, Kutnahorsky R et al. Antiphospholipid antibodies and preeclampsia: a case-control
study. Obstetrics and Gynecology 2001; 97: 29–34.
9. Diz-Küçükkaya R, Hacihanefioglu A, Yenerel M et al.
Antiphospholipid antibodies and antiphospholipid
syndrome in patients presenting with immune
thrombocytopenic purpura: a prospective cohort
study. Blood 2001; 98: 1760–1764.
10. Brandt JT, Triplett DA, Alving B, Scharrer I. Criteria
for the diagnosis of lupus anticoagulants: an update.
On behalf of the Subcommittee on Lupus
Anticoagulant/Antiphospholipid Antibody of the
Scientific and Standardisation Committee of the ISTH.
Thrombosis and Haemostasis 1995; 74: 1185–1190.
11. Harris EN, Pierangeli SS. Revisiting the anticardiolipin
test and its standardization. Lupus 2002; 11: 269–275.

140

12. Galli M, Luciani D, Bertolini G, Barbui T. Lupus
anticoagulants are stronger risk factors for thrombosis

than anticardiolipin antibodies in the
antiphospholipid syndrome: a systematic review of the
literature. Blood 2003; 101: 1827–1832.
13. Cowchock FS, Reece EA, Balaban D et al. Repeated
fetal losses associated with antiphospholipid
antibodies: a collaborative randomised trial comparing
prednisone with low-dose heparin treatment.
American Journal of Obstetrics and Gynaeology 2003
101; 1319–1332.
14. Rai R, Cohen H, Dave M, Regan L. Randomised
controlled trial of aspirin and aspirin plus heparin in
pregnant women with recurrent miscarriage
associated with phospholipid antibodies. British
Medical Journal 1997; 314: 253–257.
15. Kutteh WH. Antiphospholipid antibody associated
recurrent pregnancy loss: treatment with heparin and
low-dose aspirin is superior to low dose aspirin alone.
American Journal of Obstetrics and Gynaecology 1996;
174: 1584–1589.
16. Empson M, Lassere M, Craig J, Scott J. Prevention of
recurrent miscarriage for women with
antiphospholipid antibody or lupus anticoagulant.
Cochrane Database of Systematic Reviews 2005; 002859.
17. Farquharson RG, Quenby S, Greaves M.
Antiphospholipid syndrome in pregnancy: a
randomised controlled trial of treatment. Obstetrics
and Gynecology 2002; 100: 408–413.
18. Mak A, Cheung MWL, Cheak AA, Ho RC.
Combination of heparin and aspirin is superior to
aspirin alone in enhancing live births in patients with
recurrent pregnancy loss and positive
anti-phospholipid antibodies: a meta-analysis of
randomized controlled trials and meta-regression.
Rheumatology 2010; 49: 281–288.
19. Le Thi Huong D, Weschler B, Vauthier-Brouzes et al.
The second trimester Doppler ultrasound examination
is the best predictor of late pregnancy outcome in
systemic lupus erythematosis and/or the
antiphospholipid syndrome. Rheumatology 2006; 45:
332–338.
20. Triolo G, Ferrante A, Ciccia F et al. Randomised study
of subcutaneous low molecular weight heparin plus
aspirin versus intravenous immunoglobulin in the
treatment of recurrent fetal loss. Arthritis and
Rheumatism 2003; 48: 728–731.
21. Cervera R, Asherson RA, Font J et al. Chorea in the
antiphospholipid syndrome. Clinical, radiologic and
immmunolgic characteristics of 50 patients from our
clinics and recent literature. Medicine 1997; 76(3):
203–212.
22. Boffa MC, Lachassine E. Infant perinatal thrombosis
and antiphospholipid antibodies: a review. Lupus 2007;
16: 634–641.

Section 4
Chapter

Thrombophilia and fetal loss

Thrombophilia and pregnancy loss
Isobel D. Walker

Introduction
Pregnancy loss is psychologically and emotionally
extremely difficult for the mother, her partner and
wider family. Couples who have had such an event,
have many questions including, what caused the pregnancy to fail, will it happen again, and what can be
done to minimize the risk of a recurrence?
It has been postulated that, in some cases, pregnancy failure may be, at least in part, due to inadequate placental circulation and that thrombophilia, by
increasing the risk of fibrin deposition or thrombosis within the placental circulation, may increase the
risk of pregnancy loss. This postulate has led to the
hypothesis that, for women with a history of pregnancy
loss with no identifiable cause other than an underlying thrombophilia, intervention with antithrombotics
may improve the outcome in subsequent pregnancies.

Epidemiology
Sporadic pregnancy loss is very common. It has been
estimated that 30%–50% of fertilized ova are spontaneously aborted with around 15% of clinically recognized pregnancies being lost before 24 weeks’ gestation. Around one in 20 women will suffer two or more
consecutive pregnancy losses and 1% suffers the loss
of three or more consecutive pregnancies (Table 12.1).
The observed incidence of recurrent pregnancy loss
is greater than the 0.34%, which may be expected by
chance, suggesting that some women are predisposed
to pregnancy loss.
The World Health Organization’s definition of miscarriage is a pregnancy which fails to progress, resulting in the death and expulsion of an embryo or fetus
weighing no more than 500 g (which corresponds to
a gestational age of 20 weeks or less). Unfortunately,
this definition is not used consistently and a huge range

of “definitions” have been employed in the many studies examining potential associations between thrombophilia and pregnancy loss. Recurrent pregnancy loss
is defined as the occurrence of three or more consecutive miscarriages. This includes women with primary recurrent pregnancy loss (with ≥ 3 consecutive pregnancy losses and no pregnancy proceeding
beyond 20 weeks’ gestation) and women with secondary recurrent pregnancy loss (with ≥ 3 consecutive pregnancy losses following a pregnancy which
proceeded beyond 20 weeks’ gestation and resulted in
a live birth, stillbirth, or neonatal death). Some investigators include within the definition of recurrent pregnancy loss, women who have had ≥ 3 non-consecutive
pregnancy losses. To add further confusion, because
reproductive practice has changed and many women
nowadays do not embark on their first pregnancy
until they are in their mid to late 30s, there has
been an increasing tendency to consider intervention
for women who have a history of only two consecutive miscarriages and to “label” these women as having a history of recurrent pregnancy loss.

Pathogenesis
Thrombophilia
By definition, thrombophilias are disorders of
hemostasis which predispose to thrombosis. Included
are heritable deficiencies of the natural anticoagulants
antithrombin, protein C, and protein S and common
mutations in the genes encoding clotting factor V,
factor V Leiden which results in increased resistance
to activated protein C and clotting factor II, the
prothrombin G20210A mutation. Also included are
acquired abnormalities such as antiphospholipid
antibodies and some disorders of mixed genetic and

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

141

Section 4. Thrombophilia and fetal loss

Table 12.1 Incidence of pregnancy loss
Spontaneous abortion of fertilized ova

30%–50%

Spontaneous loss of clinically recognized
pregnancy before 24 weeks

15%

Spontaneous loss of two or more consecutive
pregnancies

Three or more consecutive pregnancy losses
before 20 weeks

Table 12.2 The prevalences of heritable thrombophilias in
European populations
Thrombophilia

Prevalence (%)

Antithrombin deficiency

0.25–0.55

Protein C deficiency

0.20–0.33

Protein S deficiency

0.03–0.13

Factor V Leiden (heterozygous)

2–7

Prothrombin G20210A (heterozygous).

environmental etiology such as hyperhomocysteinemia. Around 10% of Caucasians carry an identifiable
heritable thrombophilia (Table 12.2).

Placentation in normal pregnancy

142

Successful pregnancy requires trophoblast invasion
into the maternal uterine spiral arteries converting
them into large dilated vessels, which lack a functioning contractile smooth muscle wall. Prior to this
remodeling, the spiral arterioles are occluded by
endovascular trophoblasts. It is postulated that this
plugging protects the early intervillous spaces from
maternal systemic arterial pressure and protects the
developing intervillous trophoblasts from high oxygen tension and oxidative damage. Although the blood
flowing through the spiral arteries and the placental
intervillous spaces is maternal, the cells lining these
spaces are embryonic trophoblasts. Thus the hemostatic balance within the placenta may be disturbed as
a result of hypercoagulability of the maternal blood or
as a result of abnormality of cellular regulatory mechanisms of fetal origin operating at the feto-maternal
interface. Since the fetal blood is separated from the
trophoblasts by fetal endothelial cells, the fetal hemostatic balance is influenced only by components of fetal
origin.

Placental pathology in
pregnancy loss
Thrombi in the spiral arteries or fibrin deposition in
the intervillous spaces on the maternal side of the placenta may result in inadequate placental perfusion.
Microthrombi are frequently found in the vessels of the
placentas from women who have experienced pregnancy loss and placental infarction has been described
in the placentas of some, but not all, women who have a
pregnancy loss and who have thrombophilia. Placental
thrombosis and infarction are, however, not uncommon in fetal loss cases in the absence of any identifiable thrombophilia and no placental lesion is specific for thrombophilia. Most of the studies that have
reported on the placental pathology in women with
thrombophilia and a history of pregnancy loss have
concentrated on women with antiphospholipid syndrome. There is limited information about the placental pathology in women with pregnancy loss and
an underlying heritable thrombophilia. Furthermore,
there are methodological problems with many of the
published studies. Some studies have compared the
placentas from women with thrombophilia and pregnancy loss with the placentas from non-thrombophilic
women with normal gestations – others with placentas
from non-thrombophilic women with pregnancy loss.
Others have included no control group at all.

Is heritable thrombophilia associated
with pregnancy loss?
Observational studies
It has long been accepted that antiphospholipids in
maternal plasma increase the risk of both early and late
pregnancy loss. More recently, attention has turned to
the question of the potential role of heritable thrombophilias in the causation of pregnancy loss. Associations between heritable thrombophilias and pregnancy loss were first noted in families in which the
probands had presented with venous thrombosis. One
such study of family members, the European Prospective Cohort on Thrombophilia (EPCOT), reported the
incidence of pregnancy loss in a cohort of 571 women
with heritable thrombophilia who had experienced
1524 pregnancies and 395 age-matched controls who
had had 1019 pregnancies and reported a significantly
greater percentage of women with thrombophilia had
a history of pregnancy loss (29.4% vs. 23.5%).1 The

Chapter 12. Thrombophilia and pregnancy loss

Table 12.3 Pregnancy loss in women with heritable
thrombophilia (identified because of a family history of venous
thrombosis). Data from EPCOT Study1

Pregnancy loss Pregnancy loss
at <28 weeks at ≥ 28 weeks

Antithrombin
deficiency

Odds Ratio
(95% CI)

1.7 (1.0–2.8)

5.2 (1.5–18.1)

Protein C deficiency 1.4 (0.9–2.2)

2.3 (0.6–8.3)

Protein S deficiency 1.2 (0.7–1.9)

3.3 (1.0–11.3)

Factor V Leiden

0.9 (0.5–1.5)

2.0 (0.5–7.7)

Combined defects

0.8 (0.2–3.6)

14.3 (2.4–86.0)

Odds Ratio (OR) for fetal loss associated with thrombophilia was 1.35, 95% Confidence Interval (CI) 1.01–
1.82. When the data for all thrombophilic women and
the controls were stratified according to the stage of
gestation at which the pregnancy losses occurred, the
odds ratio was statistically significant only for fetal
losses after 28 weeks’ gestation (OR 3.6, 95% CI 1.4–
9.4). For pregnancy loss at, or before, 28 weeks the
odds ratio was 1.27 (95% CI 0.94–1.71). When the data
for fetal losses were stratified according to the specific
thrombophilic defects, the odds ratios for individual
thrombophilias were significant only for antithrombin
deficiency and protein S deficiency for pregnancy loss
after 28 weeks’ gestation (Table 12.3). The odds ratio
for pregnancy loss after 28 weeks’ gestation in women
with more than a single identifiable heritable thrombophilia was 14.3 (95% CI 2.4–86.0), suggesting a possible dose–response effect.

Meta-analyses
Following the early studies of families, numerous studies have examined possible associations between heritable thrombophilias and pregnancy loss. Within the

past 5 years, two meta-analyses have been published.
Both noted significant heterogeneity between studies.
Rey et al., following analysis of 31 case-control, cohort,
and cross-sectional studies, reported that factor V Leiden and the prothrombin G20210A mutation are significantly associated with recurrent early fetal loss and
with non-recurrent late fetal loss, and that protein S
deficiency was associated with late non-recurrent fetal
loss but not with recurrent fetal loss (Table 12.4).2 Rey
also found that maternal factor V Leiden was associated with recurrent late pregnancy loss – odds ratio
7.83 (95% CI 2.83–21.7). The meta-analysis by Robertson et al. confirmed3 that factor V Leiden and the
prothrombin G20210A mutation are associated with
recurrent early fetal loss and non-recurrent late fetal
loss, and reported that both factor V Leiden and the
prothrombin G20210A mutation are associated with
non-recurrent second trimester loss (Table 12.4). As
in the Rey meta-analysis, Robertson et al. found that
protein S deficiency is associated with an increased
risk of late pregnancy loss. This meta-analysis also
reported an association between hyperhomocysteinemia and early pregnancy loss (odds ratio 6.25, 95% CI
1.37–28.42).

Very early pregnancy loss – embryo loss
Most reports do not separate very early pregnancy
losses (before 10 weeks’ gestation) from later first
trimester losses. In a cohort study, 491 patients with
a history of adverse pregnancy outcome, maternal
thrombophilia was associated with an increased risk
of pregnancy loss after 10 weeks’ gestation (odds
ratio 1.76, 95% CI 1.05–2.94 for women with one
thrombophilia and odds ratio 1.66, 95% CI 1.03–2.68
for women with more than one identifiable thrombophilia) (Table 12.5).4 Paradoxically, the presence of

Table 12.4 Heritable thrombophilia and pregnancy loss. Data from two meta analyses∗2 and∗∗3

Factor V Leiden

Prothrombin G20210A

Odds Ratio (95% CI)

∗

∗∗

∗

Protein S deficiency
Odds Ratio (95% CI)
∗∗

Rey

Robertson

Rey

Robertson

Rey∗

Robertson∗∗

Recurrent 1st
trimester loss

2.01
(1.13–3.58)

1.91
(1.01–3.61)

2.56
(1.04–6.29)

2.70
(1.37–5.34)

14.7
(0.99–218)

Non recurrent
2nd trimester loss

4.12
(1.93–8.81)

8.60
(2.18–33.95)

Late pregnancy
loss

3.26
(1.82–5.83)

2.06
(1.10–3.86)

2.30
(1.09–4.87)

2.66
(1.28–5.53)

7.39
(1.28–42.6)

20.09
(3.70–109.15)

143

Section 4. Thrombophilia and fetal loss

Table 12.5 Heritable thrombophilia and embryo or fetal loss4

1 thrombophilia >1 thrombophilia
Odds Ratio (95% CI) Odds ratio (95% CI)
Embryo loss
⬍10 weeks’
gestation
Fetal loss
⬎10 weeks’
gestation
⬎ 14 weeks’
gestation

0.55 (0.33–0.92)

0.48 (0.29–0.78)

1.76 (1.05–2.94)

1.66 (1.03–2.68)

3.41 (1.9–6.1)

3.86 (2.26–6.59)

one or more maternal thrombophilias seemed to be
protective of recurrent very early (less than 10 weeks’
gestation) pregnancy loss (odds ratio 0.55, 95% CI
0.33–0.92 for one and odds ratio 0.48, 95% CI 0.29–
0.78 for multiple thrombophilias).

Fetal thrombophilia
It has been suggested that fetal carriage of thrombophilic mutations may have adverse clinical consequences. In one case control study a twofold increase
in factor V Leiden carrier frequency was noted in abortuses compared with unselected pregnant women, but
most studies have not shown a significant association between fetal carriage of the most prevalent heritable thrombophilias (factor V Leiden and prothrombin G20210A) and feto-placental thrombosis.

Does maternal heritable
thrombophilia cause pregnancy loss?

144

The majority of published studies have been too small
and therefore inadequately powered to detect odds
ratios of 2 or more for heritable thrombophilias which
usually have prevalences of less than 5% in the general
population. Meta-analyses support the hypothesis that
at least some heritable thrombophilias are associated
with pregnancy loss but, where the data are mature
and the confidence intervals are narrow (i.e. for factor
V Leiden and prothrombin G20210A), the point estimates of the odds ratios are small, suggesting that the
associations, if they truly exist, are weak.
Even if an association exists, it may not be a causal
association. However, a few studies have shown that,
compared with heterozygotes carrying a single thrombophilic variant, homozygous patients or patients
with combinations of thrombophilic variants have
increased odds ratios for pregnancy loss. This apparent dose effect would support the hypothesis of causal-

ity but needs further evidence. Evidence that pregnancy outcome could be improved in thrombophilic
women with a history of pregnancy loss by reducing
the hypercoagulability with anticoagulant treatment
would offer indirect support to the hypothesis that
maternal thrombophilia may cause pregnancy loss.
Many factors including chromosomal abnormalities, endocrine disorders, anatomical aberrations,
and infections have been shown to cause pregnancy loss, but around 40% of cases of recurrent
pregnancy losses are unexplained after gynecological, hormonal, immunological, microbiological, and
karyotypic investigations. There is an increasing risk
of pregnancy loss as the number of previous losses
increases. Recurrent pregnancy loss is recognized to
be a multi-causal disorder. Whilst heritable thrombophilias may not alone cause pregnancy loss, it is possible that carriage of a thrombophilic variant may contribute to a complex network of factors, which together
result in pregnancy failure.
It is generally assumed that the mechanism of pregnancy failure associated with maternal thrombophilia
involves fibrin deposition or thrombosis secondary to
hypercoagulability but, although it is biologically plausible that placental thrombosis may have a role in
the causation of fetal loss after 10 weeks’ gestation,
it is not plausible that this mechanism would cause
embryo loss (before 10 weeks) prior to development of
the placental vasculature. The vast majority of recurrent pregnancy losses occur early in pregnancy. In
women with antiphospholipid syndrome in addition
to hypercoagulability, a non-prothrombotic mechanism has been postulated. Antiphospholipid antibodies have been shown to inhibit extravillous trophoblast
differentiation and subsequent placentation. Studies of
trophoblast differentiation and early placental development are lacking in heritable thrombophilias, but
experiments in mice have shown that maternal protein C is activated following binding to thrombomodulin on the trophoblast surface. Activated protein C
then binds to endothelial protein C receptor also on
the trophoblast surface and with protein S as a cofactor down-regulates local coagulation activation. Tight
regulation of thrombin generation is essential for the
regulation of trophoblast cell growth and limits the
production of fibrin degradation products, which trigger trophoblast apoptosis. It is therefore possible that,
in early pregnancy in humans, some maternal heritable thrombophilias may exert an adverse effect on
normal trophoblast development.

Chapter 12. Thrombophilia and pregnancy loss

Diagnosis
Thrombophilia testing
Who should be tested?
Routine testing for thrombophilias in unselected populations is not recommended. There are important
issues relevant to the clinical utility and cost effectiveness of testing that must be addressed in considering who should be tested. Positive tests are not sensitive predictors of poor pregnancy outcome in women
with no history of pregnancy complications. There
are a number of published guidelines which suggest
that women with a history of recurrent pregnancy loss
and women with a history of unexplained late pregnancy loss be tested for antiphospholipids, testing for
both lupus anticoagulant activity and elevated anticardiolipins.5 Some authors have extended this guidance
and have suggested that women with a history of recurrent early pregnancy loss or an unexplained late pregnancy loss should, in addition, be tested for heritable
thrombophilia.5 Others take an opposing view suggesting testing for heritable thrombophilia is at present
not indicated, since there is insufficient evidence on
which to base any intervention in women with a history of pregnancy loss with no other identified abnormality apart from a heritable thrombophilia.6

What tests?
Currently, not only is there a lack of consensus about
which individuals (if any) merit thrombophilia testing, but there is also no universal agreement regarding
which tests should be included in the “thrombophilia
screen.” Most diagnostic laboratories would include
functional assays of antithrombin, protein C and an
immunologic assay of free protein S along with tests to
detect factor V Leiden and the prothrombin G20210A
mutation (Table 12.6). A few centers include an assay
of homocysteine in the panel of tests they offer for
women with a history of pregnancy loss.

Pitfalls
If testing for heritable thrombophilia is pursued, managing clinicians should be aware that there are numerous potential pitfalls in the interpretation of “thrombophilia screens” particularly in pregnant or recently
pregnant women. Antithrombin activity falls slightly
towards the end of a normal pregnancy, but usually

Table 12.6 Thrombophilia screening tests

Heritable
thrombophilias

Preferred test method

Antithrombin deficiency

Heparin co-factor activity

Protein C deficiency

Chromogenic activity

Protein S deficiency

Free protein S antigen

Factor V Leiden

Activated protein C resistance after
predilution of test plasma in factor V
depleted plasma or DNA based

Prothrombin G20210A

DNA based

Acquired
thrombophilia

Preferred test method

Lupus anticoagulant

Clotting based tests∗

Anticardiolipins

Immunologic assays of IgG and IgM
anticardiolipin

∗

See Chapter 11

levels remain within the reference range for non pregnant subjects. Protein C activity is unaffected by gestation, although an elevation of protein C activity
occurs in the early puerperium. Even in non-pregnant
women there is considerable overlap of protein S levels between “normals” and subjects with heritable protein S variants. The levels of both free and total protein S are reduced by 60%–70% in uncomplicated pregnancy. A diagnosis of possible protein S deficiency
made on a sample collected during pregnancy or the
puerperium requires confirmation when the woman
is no longer pregnant, puerperal or using hormonal
contraception. Pregnancy is also associated with a progressive increase in resistance to activated protein C
(APC) due to the physiological rise in clotting factor VIII levels and fall in protein S levels. Using the
original APC resistance test, around 40% of pregnant
women in their third trimester have an APC sensitivity ratio below the general population reference range.
Testing for factor V Leiden therefore requires the use of
a modified APC resistance test with predilution of the
test sample in factor V deficient plasma or genetic testing. Thrombophilia test results should always be interpreted by staff experienced in the reporting of thrombophilia tests and in the light of clear clinical information about each particular patient.

Management
General measures
First, primary prevention of vascular placental complications using antithrombotics is not indicated, so

145

Section 4. Thrombophilia and fetal loss

routine screening of asymptomatic women cannot be
justified on this basis. Second, the prognosis for future
pregnancies in women with a heritable thrombophilia
who have a history of recurrent pregnancy may be better than generally expected. The EPCOT investigators
reported that the prognosis in subsequent pregnancies
of women with recurrent pregnancy loss and underlying heritable thrombophilia was a live birth rate of
63%.1
In women with a history of previous pregnancy
loss, ante-natal surveillance to assess placental function is useful. Given the possible association between
hyperhomocysteinemia and pregnancy loss, it would
seem prudent to suggest that women with a history of
pregnancy loss take prophylactic doses of folic acid in
subsequent pregnancies

Are antithrombotics useful?
Studies published in the 1990s reported improved
pregnancy outcome in women with antiphospholipid
syndrome and a history of recurrent pregnancy loss,
given prophylactic doses of heparin combined with
low dose aspirin compared with those given aspirin
alone, but there is a paucity of data to indicate whether
antithrombotic therapy is beneficial in women with
heritable thrombophilia and pregnancy loss.
In a prospective study, 131 women with heritable
thrombophilia (identified because of a family history
of venous thrombosis) were followed through their
first pregnancy. Only 7 of 83 (8%) given thromboprophylaxis to prevent venous thrombosis had a pregnancy loss compared with 10 of 48 (21%) who received
no thromboprophylaxis (relative risk 0.3; 95% CI 0.1–
1.0).7

Open, non-controlled studies

146

In women with heritable thrombophilia and a previous history of pregnancy loss, low molecular weight
heparin has been tested in open, non-controlled studies in which outcomes were compared with the outcome of the subjects’ previous pregnancies or with outcomes in controls who were either untreated or treated
differently. In studies in which pregnancy outcome in
women with thrombophilia and a history of pregnancy
loss treated with once daily prophylactic doses of a low
molecular weight heparin was compared with their
past obstetric history, low molecular weight heparin
use was associated with an increase in the live birth rate
from 20% to 75%.8

Randomized studies
Although some randomized controlled studies have
been reported, they lack a no treatment or placebo
group. In one, women with factor V Leiden, prothrombin G20210A or protein S deficiency and a history of
unexplained pregnancy loss after 10 weeks’ gestation,
live births were recorded in 86% of 80 women given
daily prophylactic doses of a low molecular weight
heparin from 8 to 37 weeks’ gestation and in only 29%
of the 80 women given daily low dose aspirin (odds
ratio 15.5, 95% CI 7–34; P⬍0.0001).9 The randomization and “blinding” in this trial have been criticized.
There is no evidence that aspirin improves fetal outcome in women with heritable thrombophilias and a
history of fetal loss.8

Purist vs. pragmatic management
From a purists’ standpoint there is currently insufficient evidence on which to base antithrombotic intervention in women with a history of pregnancy loss
with no other identified abnormality apart from a heritable thrombophilia. This is the position adopted by
many authors, by the British Committee for Standards
in Haematology6 and by the authors of a recently published Cochrane Review.10 In support of this position
it has to be reiterated that the use of antithrombotic
drugs during pregnancy is not without risk for mother
and fetus and, in general, empirical intervention during pregnancy should be discouraged.
Pragmatists, on the other hand, argue that there is
at least some observational evidence that women with
heritable thrombophilia who have suffered pregnancy
loss may benefit from intervention with antithrombotic drugs in future pregnancies and they point out
that low molecular weight heparins are used increasingly in pregnant women and are, in general, considered safe. Based on extrapolation from the evidence of
benefit from intervention with heparin and low dose
aspirin in women with antiphospholipid syndrome
and recurrent pregnancy loss, the limited evidence in
women with heritable thrombophilia, and the relative
safety of prophylactic doses of low molecular weight
heparin in pregnancy, an increasing number of clinicians are willing to prescribe antithrombotic agents
to women with heritable thrombophilia and a history
of two or more otherwise unexplained miscarriages
or one unexplained later intra-uterine fetal death.
The American Consensus of Chest Physicians suggested for women with heritable thrombophilia and

Chapter 12. Thrombophilia and pregnancy loss

recurrent miscarriage or a second-trimester or later
loss, prophylactic doses of low molecular weight heparin (or minidose unfractionated heparin) with low
dose aspirin therapy during pregnancy and following
delivery.5

Dilemmas
Women who have suffered pregnancy loss have many
questions and will seek information about the possible
cause, the likelihood of recurrence, and the possibility of intervention to try to reduce the chances of further pregnancy loss. At present, however, there is a lack
of solid evidence on which to base advice about the
appropriateness or otherwise of testing these women
for heritable thrombophilias or on the management of
those who may be found to have a heritable thrombophilia.

Lack of evidence
Studies on the management of pregnancy loss are frequently flawed. The subjects included are often poorly
selected and form a heterogeneous group lacking stratification for important factors such as maternal age,
past obstetric history, and stage of gestation. Some
studies have compared pregnancy outcome in patients
subjected to a new intervention with the outcome in
their own previous pregnancies. This strategy ignores
the fact that some of these women will have suffered previous pregnancy loss merely by chance and
will, as a result of the phenomenon of “regression to
mean,” have a high chance of having a successful pregnancy outcome without any additional intervention.
Reports of studies in which the pregnancy outcome
in women with a history of pregnancy loss subjected
to some experimental intervention is compared with
pregnancy outcome in historical controls who were
either untreated or treated differently are subject not
only to the phenomenon of “regression to mean” in the
treated patient group but also to problems ascertaining
the control information.

Randomized trials are needed urgently
Proper evaluation of interventions in women with
a history of pregnancy loss requires randomized,
double-blind, controlled trials in which patients are
carefully selected to ensure that the treated and con-

trol groups are similar with respect to all of the important determinants of pregnancy outcome. Randomized, controlled trials have proven difficult to complete
because many women do not wish to risk being
randomized to the control group. A number of randomized double-blind studies are, however, nearing
completion. It has to be hoped that these studies will
provide more solid evidence on which to base information and advice for women with a history of unexplained late pregnancy loss or recurrent early loss.
In the meantime, many clinicians choose to treat
patients on an individual and pragmatic basis with
prophylactic daily doses of low molecular weight
heparin (e.g. enoxaparin 40 mg daily or dalteparin
5000 units daily) throughout pregnancy and the puerperium. Some also advocate the addition of low dose
aspirin (⬍150 mg) daily. The pros and cons of intervention should be discussed with the patient and the
lack of proof of efficacy of thromboprophylaxis made
clear. Ideally, this discussion should take place during
preconception counseling.

Summary

r Many studies have examined the association
between heritable thrombophilias and fetal loss,
but the results are frequently contradictory,
populations heterogeneous, and the absolute risk
(if any) small.
r Published meta-analyses suggest that factor V
Leiden, prothrombin G20210A, and protein S
deficiency are associated with an increased risk of
recurrent early pregnancy loss and non-recurrent
late pregnancy loss.
r Women with a history of pregnancy loss merit
increased surveillance in subsequent pregnancies
and should be given folic acid during pregnancy.
r Currently, there is a lack of evidence on which to
base any pharmacologic intervention in women
with a history of pregnancy loss with no other
identified abnormality apart from a heritable
thrombophilia.
r Despite the lack of evidence from randomized,
double-blind, placebo-controlled trials, many
clinicians are offering women with a history of
pregnancy loss found to have a heritable
thrombophilia self-administered prophylactic
doses of low molecular weight heparin +/− daily
low dose aspirin in subsequent pregnancies.

147

Section 4. Thrombophilia and fetal loss

References
1. Preston FE, Rosendaal FR, Walker ID et al.
Increased fetal loss in women with heritable
thrombophilia.[comment]. The Lancet 1996; 348:
913–916.
2. Rey E, Kahn SR, David M, Shrier I. Thrombophilic
disorders and fetal loss: a meta-analysis. The Lancet
2003; 361: 901–908.
3. Robertson L, Wu O, Langhorne P et al. Thrombophilia
in pregnancy: a systematic review. British Journal of
Haematology 2006; 132: 171–196.
4. Roque H, Paidas MJ, Funai EF et al. Maternal
thrombophilias are not associated with early
pregnancy loss. Thrombosis and Haemostasis 2004; 91:
290–295.
5. Bates SM, Greer IA, Hirsh J, Ginsberg JS. Use of
antithrombotic agents during pregnancy: the Seventh
ACCP Conference on Antithrombotic and
Thrombolytic Therapy. Chest 2004; 126: 627S–
644S.

148

6. Walker ID, Greaves M, Preston FE. Investigation and
management of heritable thrombophilia. British
Journal of Haematology 2001; 114: 512–528.
7. Vossen CY, Preston FE, Conard J et al. Hereditary
thrombophilia and fetal loss: a prospective follow-up
study. Journal of Thrombosis and Haemostasis 2004; 2:
592–596.
8. Brenner B, Hoffman R, Blumenfeld Z, Weiner Z,
Younis JS. Gestational outcome in thrombophilic
women with recurrent pregnancy loss treated by
enoxaparin. Thrombosis and Haemostasis 2000; 83:
693–697.
9. Gris JC, Mercier E, Quere I et al.
Low-molecular-weight heparin versus low-dose
aspirin in women with one fetal loss and a
constitutional thrombophilic disorder. Blood 2004;
103: 3695–3699.
10. Di Nisio M, Peters L, Middeldorp S. Anticoagulants
for the treatment of recurrent pregnancy loss in
women without antiphospholipid syndrome. Cochrane
Database of Systematic Reviews 2005;(2):CD004734.

Section

Hemorrhagic disorders

Section 5
Chapter

13a

Hemorrhagic disorders

Management of obstetric hemorrhage:
obstetric management
Annette Briley and Susan Bewley

Introduction and epidemiology
Obstetric hemorrhage (OH) is the leading cause of
maternal mortality worldwide. In the UK, mortality
rates are relatively low, with 17 deaths per 100 000
maternities recorded in the Confidential Enquiry into
Maternal and Child Health (CEMACH) report 2000–
2002.1 However, morbidity remains high, and timely
recognition and management is of the utmost importance.
Antepartum hemorrhage (APH) is defined as
bleeding from the genital tract after 24 weeks’ gestation
and affects approximately 3%–4% of all pregnancies.
The most common cause of APH is due to the presence of placenta previa, where the placenta is abnormally located in the lower uterine segment, covering
or partially covering the internal os. As pregnancy progresses, especially as the lower segment forms or the
cervix dilates, the woman is prone to episodes of bleeding that may be profuse. Another common cause of
APH is placental abruption, when the placenta prematurely separates either partially or totally. It may be a
single episode or recurrent, small or large, and the features may be typical and multiple (bleeding and pain,
tender and woody Couvelaire uterus with stillbirth)
or atypical and isolated (bleeding, premature labor,
fetal growth restriction, abnormal CTG). However, the
cause of many cases of APH is often unknown.
Primary postpartum hemorrhage (PPH) is the
most common obstetric hemorrhage and is defined
by the World Health Organization (WHO) as the loss
of blood estimated to be ⬎500 ml from the genital tract within 24 hours of delivery. After this,
and until 6 weeks’ postpartum, abnormal bleeding
from the genital tract is defined as secondary PPH.
Hemorrhage is considered severe when blood loss
exceeds 1000 ml.2 The major cause of postpartum

hemorrhage is uterine atony, when the uterus fails to
contract fully after delivery of the placenta. PPH complicates 11% of deliveries worldwide, and is annually
responsible for 132 000 maternal deaths.3 . Even with
appropriate active management, around 3% of women
will experience a PPH2 following vaginal delivery and
a recent study in low risk Australian women suggested
it was as high as 12%.4 Hemorrhage is a direct cause
of around 30% of all maternal deaths worldwide, the
majority occurring in the poorest countries.5 Substandard care has been highlighted as a factor in 60% of
maternal deaths in the UK Confidential Enquiry into
Maternal Deaths 2003–2005 report.6
Table 13a.1 shows the estimated time to death for
obstetric emergencies, highlighting APH and PPH, in
particular, and revealing obstetric hemorrhage as the
most dangerous complication of pregnancy for the
mother.7

Prevention
Prevention of PPH is via the recognition of any risk
factors present either ante-natally or during the intrapartum period, and the subsequent implementation of
preventative management/strategies.
Although there are a host of risk factors (see Table
13a.2 below), postpartum hemorrhage often occurs in
women with no identifiable predictors and therefore
clinicians must be prepared for this eventuality at each
and every delivery.8
The degree of risk will influence the management
of these women from place of birth to mode of delivery and post-natal care. Women at higher risk of hemorrhage should be advised to have their babies in
consultant-led units with an on-site blood bank. It is
important to involve the woman and her family in the
multi-disciplinary plan for her delivery, which must

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

151

Section 5. Hemorrhagic disorders

Table 13a.1 Estimated time to death for
obstetric emergencies

Cause

Time to death

Postpartum hemorrhage

2 hours

Antepartum hemorrhage

12 hours

Uterine rupture

1 day

Eclampsia/severe PET

2 days

Obstructed labor

3 days

Infection

6 days

be well documented and reviewed as the pregnancy
progresses and risk factors change. CEMACH recommends planned management, particularly in cases of
placenta percreta.6
Ultrasound localization of the placenta in all
women, especially those who have had previous
Cesarean section,9 should be reported and documented clearly in the handheld notes. Ante-natal assessment of full blood count and treatment of anemia is
essential.

The importance of communication with all members of the multi-disciplinary team, and early involvement of senior medical and midwifery staff, have been
highlighted in successive Confidential Enquiries and
CEMACH Reports to improve prognosis. The Scottish Audit has found that reporting morbidities and
the resultant review of management has consistently
reduced the incidence of substandard care.10
If a woman is at risk of PPH, there are preventative
and predictive measures which can be implemented
in the intrapartum period. Such interventions include
giving oral ranitidine (150 mg), gaining intravenous
access with two large bore cannulae and taking blood
to send for a full blood count, group, and save.
Active management of the third stage of labor is
recommended for any woman at increased risk of
PPH. This shortens the time between delivery of the
baby and the placenta and membranes with no significant increase in retained placenta.11,12 Active management involves the administration of oxytocin (or other
uterotonic drug) with, or shortly following, delivery

Table 13a.2 Risk factors for postpartum hemorrhage4,7,20

Factors
Pre-pregnancy

Pregnancy acquired

Delivery acquired

Third stage

152

Maternal age ⬎35 years
Nulliparity
Grand multiparity
Asian ethnicity
Obesity
Previous Cesarean section
Previous PPH
Uterine fibroids
Factor VIII deficiency – hemophilia A carrier
Factor IX deficiency – hemophilia B carrier

Risk (if known)
×3
×2
×2
×3

Multiple pregnancy
Placenta praevia
Abnormal placental implantation – accreta, increta, and percreta
APH in current pregnancy
Polyhydramnios
Pre-eclampsia or pregnancy induced hypertension
Sepsis (including chorioamnionitis and/or endometritis)

×5
× 15

Elective cesarean section
Precipitate labor
Maternal pyrexia in labor
Oxytocin administration for induction or augmentation of labor
Labor lasting ⬎12 hours
Operative vaginal delivery
Emergency Cesarean section
Fetal macrosomia (Baby weight ⬎4 kg)
Ruptured uterus

×4

Tissue – Retained placenta (causes 10% of PPH)
Tone – Uterine atony (causes 70% of PPH)
Trauma – laceration to perineum, vagina, or cervix (causes 20% of PPH)
Thrombin – coagulopathies (causes 1% of PPH)
Infection

×4

×2
×2
×2
×9
×2

Chapter 13a. Obstetric management

Fig. 13a.1 Abnormal uteroplacental
implantation.

Increta:
placenta invades
myometrium

Normal
Implantation:
note cleavage plane

Percreta:
placenta
penetrates
through
myometrium and
serosa

Accreta:
placenta
adherent

of the anterior shoulder of the baby. Controlled cord
traction may reduce the risk of retained placenta and
subsequent need for medical intervention. There have
been no reported adverse effects of controlled cord
traction.13–15 More recently, international guidelines
include uterine massage following delivery of the placenta as the last part of active management of the third
stage,16 although there is little evidence of the effectiveness of this.17

Pathogenesis of PPH
The most common cause of PPH, accounting for
approximately 70% of occurrences is uterine atony.
There are numerous reasons for the uterus failing
to contract effectively; including exhaustion, sepsis,
and retained products. Other causes of PPH include
perineal trauma, uterine inversion, clotting disorders,
pelvic hematomas, and cervical tears. An abnormally
implanted placenta (see Fig 13a.1) (placenta accreta,
increta or percreta) can remain in situ and hence prevent the uterus from contracting properly. Placenta
previa and accreta are becoming an increasing problem, attributed to abnormal adherence of the placenta in subsequent pregnancies following Cesarean
section.
If obstetric hemorrhage is not managed efficiently
and effectively, this will lead to shock, hemostatic
failure from disseminated intravascular coagulation
(DIC), and ultimately death.

Diagnosis
Diagnosis of obstetric hemorrhage is typically by the
visualization of blood loss from the genital tract. In the
case of APH, bleeding can be concealed, and the only

sign may be evidence of maternal compromise and/or
fetal distress. With PPH, the volume of blood loss
is usually estimated visually, although this is notoriously inaccurate. In the acute situation before hemodilution, hemoglobin will not represent the amount of
blood lost and fit, young women may appear to compensate and maintain vital signs until a late stage.
Alertness and attention to clinical symptoms and signs
are vital. Some units attempt to measure blood loss
by weighing blood-soaked items, for example, sanitary pads and sheets. CEMACH suggest that an Early
Warning Score chart be used to assess maternal compromise and so give a more accurate representation
of maternal condition compared with visual estimation of blood loss, to prevent delay in emergency
management.6

Obstetric management
In the case of APH, management will depend on
the amount of bleeding, maternal compromise and/or
degree of fetal distress. A concealed APH large enough
to cause intra-uterine death is probably at least
1.5 liters and DIC and PPH should both be assumed
and anticipated.

Immediate management of PPH
Once PPH has been diagnosed, action must be rapid.
Figure 13a.2 is an effective tool in identifying what
needs to be done by whom, as often in the case of a
PPH several actions need to be taken simultaneously.
Whilst constantly assessing maternal resuscitation
requirements (pulse, blood pressure, respiration, temperature), the uterus should be massaged to stimulate a contraction, which may assist stemming of the

153

Section 5. Hemorrhagic disorders

HEAD
•
•
•
•
•

Check airway
Check breathing
Administer O2
Lie flat
Record time of relevant events

ARMS
Check pulse and BP
Establish LARGE BORE IV access x 2
Check FBC, clotting and Xmatch 4–6 units
Start FLUID RESUSCITATION
2 liters crystalloid
Drugs:
Ergometrine 0.5 mg IV (IM)
Syntocinon infusion (10 u/hr)
Prostaglandin F2α 0.25 mg IM
Consider moving to theater if >2 doses required
Consider misoprostol 800 mcg PR

UTERUS
START HERE- CALL FOR HELP
•
Massage uterus to stimulate contraction
•
Deliver placenta if still in situ
•
CO-ORDINATE:
o Assistant 1 at “HEAD”
o Assistants 2 and 3 at “ARMS”
•
Empty bladder- insert catheter
•
If atony persists apply bimanual compression
•
Review other causes; 4 T’s (Tone, Trauma, Tissue, Thrombin)
•
Move to theater early if bleeding persists

Fig. 13a.2 Management of PPH: organizing the team (adapted from PersePhone/Pingirl).2

bleeding should the cause be atonic. This massaging action also helps expel retained products
or blood clots. A full bladder could prevent the
uterus from contracting properly by impeding on the
space, and therefore catheterization is recommended
(Table 13a.3).
Other uterotonics have been suggested with limited anecdotal evidence. These include:
r Dinoprostone; however, this is not suitable in
r
r

154

hypovolemic situations.
Gemeprost (cervogem)
Sulproston – a Prostaglandin E2 widely used in
France as a second line drug after oxytocin (before
ergometrine). Can cause coronary spasm,
hypertension, pyrexia, nausea, and vomiting.
R 5 iu in 19 ml normal saline given
Vasopressin
by subendothelial infiltration. Avoid intravenous
administration as it causes severe hypertension.
Tranexamic acid – a lysine derivative, which
appears well tolerated. There is no evidence of

increased thrombosis and this drug is probably
underused. I g intravenously with, if necessary,
repeat dose 4 hours later.
r Methotrexate – prevents DNA replication and
may be useful in conservative management of
placenta accreta.

Volume maintenance in PPH
Initially, while blood is being cross-matched,
volume replacement with crystalloid should
be instituted. Close attention to fluid balance is
required to avoid the perils of hypoxia-hypovolemia,
on the one hand, and cardio-pulmonary overload
on the other. In massive hemorrhage, fluid replacement can be controlled with central venous and
arterial lines and anesthetic and hematology input
is vital both during the event and subsequently
on high-dependency or intensive care units (see
Chapter 13b).

Chapter 13a. Obstetric management

Table 13a.3 Drug Management of PPH

Name of drug

How it works

Administration

Side effects

Oxytocin
Prevention

Stimulates rhythmic upper uterine
segment contractions

IM as part of syntometrine (acts in 2–3
minutes, lasts up to 60 mins)
IV bolus 5 i.u. (acts in 1 min, has half life of
3 mins)
IV 5 i.u. bolus can be repeated
Infusion- 40 units/500 mls over 4 hours

Hypotension, due to vasodilatation, especially in cardiac
patients
r Administer slowly
r In cardiac patients infuse
10 units over 30 mins
Anti-diuretic hormone effect
r Fluid overload – can lead
to pulmonary edema
r Hyponatremia

Ergometrine
Recognition

Sustains uterine contraction via alpha
receptors in the upper and lower
uterine segment of the uterus
Combined with oxytocin as
Syntometrine
Acts in 2–5 mins lasts up to 3 hours
If atony persists give 0.5mg IV

Carboprost
R
)
(Hemobate
Treatment

This is a prostaglandin F2 alpha

Not for IV administration
Intramyometrially into the fundus of the
uterus to avoid blood vessels (but there
are very large vessels in the uterus)
Intramuscularly, 250 ␮g (microgram) every
15 mins (maximum of 2mg)
In practice most women are not given
more ⬎ 2 doses
85% women respond to the first dose

Has predelection for smooth
muscle of the bronchi and
therefore caution is
required with asthmatics
Bronchospasm
Significant intrapulmonary
shunt
Hypoxia

Microprostol
R
)
(Cytotec
Treatment

This is a prostaglandin E1 analog
It is thermostable and does not
require refrigeration.
This drug is cheap and therefore
useful in resource limited
countries.

Multiple routes of administration
Orally, sublingually, rectally
800–1000 mcg (effective within 3 mins)

Shivering
Pyrexia

Potent agonist causes blood
vessels to constrict
Vomiting+ + + most women
WILL vomit
Hypertension- do not give to
women with pre-existing
high BP, Pre-eclampsia or PIH
or cardiac disease

Fig. 13a.3 Bimanual uterine compression.

Surgical management of PPH

r Manual removal of placenta – This is an
emergency procedure to separate the placenta

manually that should be considered if it has not
delivered within an hour of birth. Partial
separation and delays can be associated with very
heavy bleeding.
r Bimanual uterine compression This is an effective
way of stemming bleeding by compressing the
uterus with both hands (Fig 13a.3)
r Examination under anesthesia (EUA) and
evacuation of retained products of conception
(ERPC) This should be performed in theater with
appropriate conditions, personnel, and
instruments and in preparation for further
procedures. It is important to explore the whole of
the uterus, cervix, vagina, and perineum in a
rigorous way even if one cause is found or
excluded. The aim of EUA is to assess the cause of
bleeding and take action accordingly. The cause
may be found, for example, retained cotyledon,
pelvic hematoma (common after a normal vaginal

155

Section 5. Hemorrhagic disorders

delivery and often requiring surgery) or cervical
tears. ERPC for secondary PPH, especially if
associated with sepsis, must be performed with
great care as it can lead to perforation of the
uterus.
r Balloon tamponade can be used. This involves
placing a balloon in the uterus and inflating it.
Most commonly around 800–1000 ml are used to
ensure the balloon does not fall out after a vaginal
delivery, but less is required if only compressing
the lower segment after elective Cesarean; the
balloon is then left in situ for 24 hours after which
time it is gradually deflated. This is a cost-effective
method. In an emergency a condom could be
filled with fluid and inserted into the uterus to
apply pressure to stop the bleeding.
r Packing with surgical gauze. This is a traditional
and effective way to stop surgical bleeding and
ooze from a raw or sutured surface, although the
disadvantage is that a second procedure may be
required for removal. Packs have to be placed
under pressure, and can be left in the uterus, the
vagina, or in the abdomen.
r The B-Lynch brace suture – Although not popular
in some units, this has revolutionized practice in
recent years. The brace suture has not been
evaluated in a RCT, but the case history evidence
is compelling, and any speculated effect on
fertility has to be compared with hysterectomy,
which would otherwise be the next surgical
option. It has been suggested that the B-Lynch
brace suture may indent the uterus or cause
necrotic uterus. Prophylactic brace sutures can be
advocated in Jehovah’s Witnesses and in others
who will refuse transfusion who require Cesarean
section and are assessed as at increased risk of
PPH (although risk assessment is difficult).

156

r Internal iliac vessel ligation – this is an
old-fashioned technique and less familiar to
obstetricians nowadays, though more often used
by gynecological oncologists. The principle has
evolved into an interventional radiology
technique of uterine artery embolization to stop
bleeding. The collateral circulation is adequate to
protect the uterus, but the equipment is not
available in all units, and there is a risk to future
fertility.
r Hysterectomy – Women may die if the decision to
do a hysterectomy is not made or made too late,
but practitioners must be prepared to defend their
decision making in the legal process, as
unnecessary loss of fertility is devastating. This
must be the treatment of last resort, having
attempted conservative measures first. The
UKOSS study of hysterectomy did show a failure
rate both for brace suture and interventional
radiology embolization.17
All women, and their companions, deserve a contemporaneous explanation, and a discussion afterwards about what happened, so that any questions
can be answered. After a PPH, women can make a
remarkable physical recovery. However, unrecognized
or undocumented PPH may lead to dizziness, fainting,
or collapse in the immediate postpartum period. All
women with symptoms or a recognized PPH should
have a postpartum or day 2–3 hemoglobin in case iron
supplements should be prescribed. A prolonged recovery may be associated with fatigue, exhaustion, and
interference with breast feeding and bonding. Massive
hemorrhage can be very traumatic, for women, their
families, and for staff, and the need for explanation and
reflection must not be underestimated. Staff skills must
be constantly updated.

Chapter 13a. Obstetric management

References
1. Lewis G. The Confidential Enquiry into Maternal and
Child Health (CEMACH). Why mothers die
(2000–2002), 2005; London: CEMACH.
2. American Academy of Family Practitioners (AAFP)
Advanced Life Support in Obstetrics. Syllabus Updates
2008. www.aafp.org/online/en/home/cme/
aafpcourses/clinicalcourses/also/syllabus.html
#Parsys0003 downloaded 21/05/2008
3. World Health Report. Make every mother and child
count. http://who.int/whr/2005 en.pdf 2005; accessed
19th May 2008.
4. Ford JB, Roberts CL, Simpson JM, Vaughan J,
Cameron CA. Increased postpartum hemorrhage rates
in Australia. International Journal of Gynaecology and
Obstetrics (the official organ of the International
Federation of Gynaecology and Obstetrics) 2007; 98,
237–243.
5. Khan KS, Wojdyla D, Say L, Gulmezoglu AM, van
Look PF. WHO analysis of causes of maternal death: a
systematic review. The Lancet 2006; 367: 1066–1074.

11. Bais JM, Eskes M, Pel M, Bonsel GJ, Bleker OP.
Postpartum haemorrhage in nulliparous women:
incidence and risk factors in low and high risk women.
A Dutch population-based cohort study on standard
(⬎ or = 500 ml) and severe (⬎ or = 1000 ml)
postpartum haemorrhage. European Journal of
Obstetrics Gynecology Reproduction Biology 2004; 115:
166–172.
12. Prendiville WJ, Elbourne D, McDonald S. Active
versus expectant management of the third stage of
labour. Cochrane Database System Review 2000(2):
CD00007.
13. Khan GQ, John IS, Wani S et al. Controlled cord
traction versus minimal intervention techniques in
delivery of the placenta: a randomised controlled trial.
American Journal of Obstetrics and Gynecology 1997;
177: 770–774.
14. Zhao S, Xiaofeng S. Clinical study in curing
postpartum haemorrhage in the third stage of labor.
Journal of Practical Obstetrics and Gynecology 2003; 19:
278–280.

6. CEMACH. Saving Mother’s Lives: Reviewing maternal
deaths to make motherhood safer – 2003–2005. The
Seventh Report of the Confidential Enquiries into
Maternal Deaths in the United Kingdom DoH 2007.

15. Giacalone PL, Vignal J, Daures JP et al. A randomised
evaluation of two techniques of management of the
third stage of labour in women at low risk of
postpartum haemorrhage. British Journal of Obstetrics
and Gynaecology 2000; 107: 396–400.

7. Magann EF, Evans S, Chauhan SP, Lanneua G, Fisk
AD, Morrison JC. The length of the third stage of labor
and risk of postpartum hemorrhage. Obstetrics and
Gynecology 2005; 105(2): 290–293.

16. Lalonde A, Daviss BA, Acosta A, Herschderfer K.
Postpartum hemorrhage today: ICM/FIGO initiative
2004–2006. International Journal of Gynaecology and
Obstetrics 2006; 94: 243–253.

8. DoH. Why Mothers Die, 1997–1999. The Fifth Annual
Report of the Confidential Enquiries in to Maternal
Deaths in the United Kingdom. DoH RCOG Press
London.

17. Chelmow D. Postpartum haemorrhage: prevention.
British Medical Journal Clinical Evidence 2007; 2:
1410.

9. Magann EF, Evans S, Hutchinson M et al. Postpartum
hemorrhage after cesarean delivery; an analysis of risk
factors. South Medical Journal 2005; 98: 681–685.
10. Penney G, Adamson L. Scottish Confidential Audit of
Severe Maternal Morbidity 4th Annual Report 2006
SPCERH.

18. Knight M, UKOSS. Peripartum hysterectomy in the
UK: management and outcomes of the associated
haemorrhage. British Journal of Obstetrics and
Gynecology: an International Journal of Obstetrics and
Gynaecology, 2007; 114, 1380–1387.

157

Section 5
Chapter

13b

Hemorrhagic disorders

Management of obstetric hemorrhage:
anesthetic management
Vivek Kakar and Geraldine O’Sullivan

Introduction

Monitoring

Once the diagnosis of obstetric hemorrhage has been
made, early senior anesthetic involvement (experienced registrar or a consultant) is vital. The UK Confidential Enquiry into Maternal and Child Health 2000–
2002 (Why Mothers Die1 ) and 2003–2005 (Saving
Mothers Lives2 ) show that hemorrhage is still one of
the commonest causes of direct maternal deaths. In
Why Mothers Die 1999–2002, 17 maternal deaths were
caused by hemorrhage; care was considered suboptimal in 5 of these 17 cases. In the Saving Mothers Lives
2003–2005, hemorrhage caused 14 deaths and was a
complicating factor in 9 others. In as many as 10 of
these 14 deaths, the patient received suboptimal care.

Blood pressure, oxygen saturation, and electrocardiogram (ECG) should be continuously monitored. There
should be a relatively low threshold to insert an arterial line in bleeding patients. A central venous pressure (CVP) monitor may be required in cases of
massive hemorrhage, although not as a part of the
initial resuscitation. A urinary catheter should be
inserted in all cases to measure the hourly urine
output. Resuscitation should be guided by clinical
parameters, arterial blood gases, lactates, CVP, and
urinary output. In cases with cardiovascular and/or
renal problems, some form of cardiac output monitoring (LiDCO, PiCCO, or esophageal Doppler in
intubated patients) can provide useful information.
Accurate assessment of blood loss is essential and
can be achieved by weighing the surgical swabs
and measuring the volume of blood in the surgical suction. The Association of Anaesthetists of Great
Britain and Ireland (AAGBI) guidelines4 for monitoring should be followed during surgery, recovery
and transfer of these patients, should that be required.
Other recommendations2,5–6 are that some form of
track and trigger scoring system (such as Modified Early Warning Scores (Table 13b.1)) should be
used in high risk patients monitored on the labor ward
to facilitate early identification of patients with ongoing hemorrhage.

Communication
Early and clear communication between obstetricians,
midwives, anesthetists, hematologists, and the porters
is essential. A “leader” should coordinate the ongoing
management of the hemorrhage. Arguably, this leader
should be the senior anesthetist. Extra help, surgical,
and/or anesthetic, should also be summoned.

Access
Several large bore intravenous (IV) cannulae
(14G/16G) should be sited. Central venous catheterization may be needed at a later stage, but should
not delay resuscitation in emergent situations, in
otherwise healthy patients. Ultrasound guidance is
recommended by NICE3 for the insertion of a central
venous catheter in the internal jugular vein. Subclavian vein cannulation should specifically be avoided
in established or suspected coagulopathy, as occurs in
concealed abruption, sepsis, severe pre-eclampsia and
massive transfusion.

158

Oxytocics (see also Chapter 13a)
(a) Oxytocin: Five to ten units should be administered
as a slow IV bolus. Rapid IV administration can
cause profound hypotension and tachycardia.
Cardiac arrest has also been reported. It works
within 2–3 minutes but due to its short half-life it
needs to be administered as an infusion, e.g.
40 IU in 40ml of normal saline over 4 hours.

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

Chapter 13b. Anesthetic management

Table 13b.1 Modified early warning scoring system5

3
HR
(bpm)
BP

⬎45%↓

⬍40

40–50

51–100

101–110

111–129

≥130

30%↓

15%↓

Normal

15%↑

30%↑

⬎45%↑

15–20

21–29

≥30

RR
(/min)

≤8

9–14

Temp
(◦ C)

⬍35.0

35.0–38.4

CNS
Urine
Output

A
Nil

⬍1 ml/kg/2 h

⬍ 1 ml/kg/h

⬎38.5
V

⬎3 ml/kg/2 h

A = Alert V = Responds to Verbal commands P = Responds to Pain U = Unresponsive.

(b) Ergometrine: It is an extremely effective
second-line drug for an atonic uterus. The IV dose
is 100–300 mcg. Uterine contraction occurs
within 5 min of an intramuscular (IM) injection
and 1 minute after an IV injection. Its effects last
at least 1 hour.
(c) Carboprost (methyl prostaglandin F2␣ ): It is an
extremely potent drug and is administered by IM
or intramyometrial injection. It should be
administered in 250 ␮g increments, repeated at
15-minute intervals up to a maximum of 2 mg.
The majority (85%) of patients will usually
respond to the first or second dose, and in practice
the full 2 mg dose will rarely be employed as
ongoing severe hemorrhage usually necessitates
further surgical/radiological intervention.
(d) Misoprostol: This is a prostaglandin E1 analog. It
is supplied as 200 ␮g tablets and 800–1000 ␮g
should be administered rectally.

Fluids
Immediate resuscitation should begin with crystalloids and colloids. There is no evidence of superiority of one over the other in non-septic obstetric
patients.7 In an emergency, the choice of fluid is immaterial. Hartmann’s solution is the most physiologically
balanced solution; normal saline can also be used,
although it can itself cause metabolic acidosis after several liters have been used.
Commonly available colloids include starches
(Voluven), gelatins (Haemaccel, Gelofusine), and albumin. There have been concerns about the effect of
starches on platelet function and renal function. A

recent large study8 found significantly increased renal
failure and blood transfusion requirements in septic patients who required more than 22 ml/kg of 10%
hetastarch (0.5/200). This study did not include the
obstetric population. Gelatins can interfere with blood
grouping, cross-matching, and cause allergic reactions. A recent Cochrane review failed to establish any
difference between different colloids in terms of outcome.9
Blood pressure, heart rate, and urine output are
good endpoints in assessing adequate resuscitation. In
fit and healthy young patients, tachycardia will usually represent uncorrected hypovolemia. Blood pressure will generally not fall until 30% of the blood volume (∼1500 ml) has been lost (Table 13b.2).

Preventing the “lethal triad” of
hypothermia, acidosis and
coagulopathy
It has been demonstrated in trauma patients with
massive bleeding, that if they are allowed to become
hypothermic and acidotic, their coagulopathy worsens or is refractory to correction.10 This has also been
shown to be true in other cases of hemorrhagic shock
(Fig. 13b.1).
Evidence suggests that a fall in temperature from
37 ◦ C to 33 ◦ C reduces rFVIIa activity by 20%, whilst
a fall in pH from 7.4 to 7.0 reduces rFVIIa activity by 90%. Platelet function is also inhibited to a
varying degree. It seems that acidosis alone does not
seem to affect the clotting, but increases the effect of
hypothermia on clotting.11 Therefore, the prevention

159

Section 5. Hemorrhagic disorders

Table 13b.2 Classification of hemorrhagic shock

Class 1
(Compensated)

Class 2
(Mild)

Class 3
(Moderate)

Class 4
(Severe)

Blood loss
(% Circulating blood
volume)

750 ml
(⬍15%)

800 – 1500 ml
(15%–30%)

1500 – 2000 ml
(30%–40%)

⬎2000 ml
(⬎40%)

Systolic blood pressure

No change

Orthostatic Fall

Low

Very low

Diastolic blood pressure

No change

Raised

Reduced

Very low

Pulse rate

⬍100

⬎100

⬎120 (weak)

⬎140 (very weak)

Capillary refill

Normal

Slow (⬎2 s)

Slow (⬎2s)

Prolonged (⬎5 s)

Respiratory rate

Normal

Raised (⬎20/min)

Urine output

⬎30 ml/h

20 – 30 ml/h

10 – 20 ml/h

0 – 10 ml/h

Extremities

Normal

Pale

Pale and cold

Complexion

Normal

Pale

Ashen

Mental state

Alert, thirsty

Anxious, thirsty

Anxious, aggressive or
drowsy

Drowsy, confused
or unconscious

Adapted from ATLS manual.

Hypothermia
and
acidosis
Hemorrhage

The lethal triad

Coagulopathy
Fig. 13b.1 The lethal triad
Fig. 13b.2 Fluid warmer.

160

of hypothermia and acidosis is an essential component
in the successful management of massive hemorrhage.
Patients should be kept warm using forced air
warming blankets. All fluids should be warmed using
any of the commonly available fluid warmers (Fig.
13b.2). When the fluid needs to be infused rapidly, normal fluid warming devices may not be able to perform
efficiently, when pressurized fluid is run through them.
Therefore, every obstetric unit should have access to a
Level 1 infusor (Fig. 13b.3) which can both pressurize
and warm blood and fluids rapidly.

Blood and blood component therapy12
(see Chapter 13c)
Blood replacement should be guided by bedside
and/or laboratory hemoglobin testing. When hemorrhage is first diagnosed, blood should be sent to the
laboratory for a group and cross match, full blood
count (FBC) and coagulation screen (INR, aPTT, fibrinogen). The hemoglobin, coagulation screen, and
platelet count will need to be repeated at regular

Chapter 13b. Anesthetic management

The cells are separated by hemoconcentration and differential centrifugation in 0.9% saline, and washed in
1–2 L of 0.9% saline. This process removes circulating
fibrin, debris, plasma, micro-aggregates, complement,
platelets, free hemoglobin, circulating pro-coagulants,
and most of the heparin. At the end of the salvaged
process, the hematocrit of the salvaged blood is usually between 55% and 60%.

Problems
Cell salvage was until recently considered to be contraindicated in obstetrics because of the following two
theoretical risks:
r Amniotic fluid embolism: In the literature, only

Fig. 13b.3 Level 1 Infusor.

intervals in order to facilitate the logical use of coagulation factors. Alternatively, if available, a thromboelastogram (TEG) can be used to guide the replacement of coagulation factors.

Cell salvage13–18
Intraoperative cell salvage and auto transfusion has
been available for many years and has been very useful
in cardiac surgery, major vascular surgery, orthopedic
surgery, and trauma. Cell salvage is now increasingly
considered in massive obstetric hemorrhage as recent
research has shown it to be safe in obstetrics.

Indications
Cell salvage is a technique for re-cycling operative
blood loss. It is particularly appropriate for elective
surgery where massive blood loss is anticipated, e.g.
placenta previa/acreta/percreta and for mothers who
refuse blood and blood products, e.g. Jehovah’s witnesses. Once skill has been acquired with the technique, it can be rapidly set up, even in an emergency.

one death has so far been reported after the
use of salvaged blood. However, the patient was
a Jehovah’s Witness with severe pre-eclampsia
and HELLP syndrome (Hemolysis, elevated liver
enzymes and low platelet count), and a leukocyte depletion filter was not used. It has now
been shown that use of a 40 ␮ leukocyte depletion filter (Fig. 13b.6), on the return limb to the
patient, effectively depletes or entirely removes
fetal squames, white blood cells, and platelets from
the salvaged blood. To date, no case of amniotic fluid embolism has been reported in patients
who received salvaged blood during Cesarean section where a leukocyte filter was used. In addition, amniotic fluid embolism is now considered to
be more of an immunological phenomenon rather
than actual physical embolism. Nevertheless, any
contamination of the salvaged blood with amniotic
fluid should be avoided as far as possible.
r Allo-immunization: Despite the use of several
wash cycles and filters, it is still not possible to
avoid contamination of the salvaged blood with
fetal red blood cells. This is because the machine
cannot distinguish between fetal and maternal red
blood cells. The amount can vary between 2 ml and
19 ml and Kleihauer counts should be routinely
performed in the postpartum period. All rhesus
negative mothers should be immunized with antiD. A second dose may be required if the Kleihauer
suggests heavy contamination with fetal cells.

Principles of cell salvage (Figs 13b.4
and 13b.5)

Cost

Blood is aspirated from the surgical site through heparinized tubing and a filter into a collecting reservoir.

Although the machines can be very expensive, most
hospitals lease them from the manufacturer. Typically,

161

Section 5. Hemorrhagic disorders

Saline +
anticoagulant

Suction from
operative slte

Double lumen aspiration
+ anticoagulent assembly
Reservoir

Retransfusion
bag
Saline
wash

Return to patient

Pump
Waste bag

Centrifuge
bowl

Fig. 13b.4 Cell salvage schematic diagram.17

it costs approximately £100–170 per patient (towards
disposables) to setup and use the cell salvage machine.
So, the cost of disposables is covered as soon as you
need to transfuse more than one unit. A systematic
review of over 600 studies comparing various transfusion strategies to reduce allogenic blood transfusion
found that the relative risk of requiring allogenic blood
transfusion with cell salvage was 0.59 and it was more
cost effective than all other strategies except acute normovolemic hemodilution. Every unit with cell salvage
facilities should have a protocol for the use of cell salvage in obstetrics.

Investigations

162

FBC and clotting should be checked frequently. In
cases of ongoing blood loss, resuscitation should
be guided by bedside hemoglobin estimation (e.g.
HemoCue, Fig. 13b.7) and/or arterial blood gas
estimation. Liver and renal function should also
be assessed at baseline and, once the patient has
been stabilized, especially in patients with complex co-morbidities, multiple medications, massive
transfusion, or prolonged period of intraoperative
hypotension.

Regional Vs general anesthesia
In an elective situation, where significant blood loss
is anticipated, such as with anterior placenta praevia,
regional anesthesia can still be considered, although
the patients should be warned of the occasional need to
convert to general anesthesia (GA) intra-operatively.
Baseline hemoglobin, venous access, invasive monitoring and a cell salvage unit should be established
prior to starting such cases. Two to four units of blood
should also be cross-matched.
In an emergency situation, anesthetic management will be determined by both fetal and maternal considerations. GA is usually considered in cases
of severe hemodynamic instability, sepsis, and suspected or confirmed coagulopathy. If GA is used in
severely hypovolemic patients, the anesthetic induction agents, ketamine or etomidate should be used
instead of thiopentone or propofol, as they do not
cause the profound hypotension commonly seen when
the latter two agents are used in hypovolemic patients.
If a bleeding patient needs to be transferred to the
interventional radiology suite, it might be worth securing the airway prior to transfer, especially if the radiology department is not very close to the obstetric unit.
In addition, adequate anesthetic facilities and assistance should be available in the radiology suite.

Chapter 13b. Anesthetic management

Fig. 13b.6 Leukocyte filters.

Fig. 13b.5 Cell salvage machine.

Post-hemorrhage care
Post-operative care will usually be on a high dependency unit, but transfer to an intensive care unit
may be necessary particularly if the patient requires
mechanical ventilation. If possible, cardiovascular and
metabolic parameters should be stabilized prior to
transfer. Acceptable standards of monitoring should be
maintained during the transfer as mentioned earlier.
Once the bleeding has been controlled and the
patient is stable, regular thromboprophylaxis should
be commenced.

Documentation
Accurate and complete documentation of the sequence
of events is very important. In cases of poor outcome,
poor documentation is indefensible even if excellent
care was provided. One person can be assigned the
job of keeping a record of all the drugs and fluids

Fig. 13b.7 Hemocue.TM

163

Section 5. Hemorrhagic disorders

164

administered and of the personnel involved in the
resuscitation.

ventional radiology, and plans for their management
should be made in advance.

Drills/protocols

Debriefing and counseling

A multidisciplinary massive hemorrhage protocol
must be available in all units and should be updated
and rehearsed regularly. Women known to be at high
risk of bleeding should be seen by a consultant anesthetist in the ante-natal period. These patients should
ideally be delivered in centers with facilities for blood
transfusion, cell salvage, intensive care, and inter-

Supportive counseling of all the team members
involved is vital, should the hemorrhage result in
maternal death. Such an event represents a tragedy
not only for the woman’s family, but also for the
carers. Debriefing after such episodes can be a very
good opportunity to reinforce learning points and seek
improvements for future.

Chapter 13b. Anesthetic management

References
1. CEMACH Report 2000–2002(Why Mothers Die).
(www.cemach.org.uk)
2. CEMACH Report 2003–2005(Saving Mothers’ Lives).
(www.cemach.org.uk)
3. National Institute for Clinical Excellence. Ultrasound
Locating Devices for Placing Central Venous Catheters.
Guideline number 49. London: NICE; 2002.
4. Association of Anaesthetists of Great Britain and
Ireland. Recommendations for Standards of Monitoring
During Anaesthesia and Recovery. London: Association
of Anaesthetists; 2007.
5. Intensive Care Society. Guidelines for the Introduction of
Outreach Services. Intensive Care Society; 2002.
6. NICE clinical guideline 50. Acutely ill patients in
hospital. July 2007.
7. Perel P, Roberts I. Colloids versus crystalloids for fluid
resuscitation in critically ill patients. Cochrane
Database Systematics Reviews 2007; 17: CD000567.
8. Brunkhorst FM, Engel C, Bloos F et al. Intensive
insulin therapy and pentastarch resuscitation in severe
sepsis. New England Journal of Medicine 2008; 358:
125–139.
9. Bunn F, Trivedi D, Ashraf S. Colloid solutions for fluid
resuscitation. Cochrane Database System Review 2008
4: CD001319.
10. Tsuei BJ, Kearney PA. Hypothermia in the trauma
patient. Injury 2004; 35: 7–15.
11. Dirkmann D, Hanke AA, Gorlinger K, Peters J.
Hypothermia and acidosis synergistically impair

coagulation in human whole blood. Anesthesia and
Analgesia 2008; 106: 1627–1632.
12. Association of Anaesthetists of Great Britain and
Ireland. Blood Transfusion and the Anaesthetist.
London: Association of Anaesthetists; 2008.
13. Allam J, Cox M, Yentis S M. Cell salvage in obstetrics.
International Journal of Obstetric Anesthesia 2008; 17:
37–45.
14. Sullivan I, Faulds J, Ralph C. Contamination of
salvaged maternal blood by amniotic fluid and fetal
red cells during elective Caesarean section. British
Journal of Anaesthesia 2008; 101: 225–229.
15. Catling SJ, Williams S, Fielding A. Cell salvage in
obstetrics: an evaluation of ability of cell salvage
combined with leucocyte depletion filter to remove
amniotic fluid from operative blood loss at caesarean
section. International Journal of Obstetric Anesthesia
1999; 8: 79–84.
16. Waters JH, Biscotti C, Potter PS, Phillipson E.
Amniotic fluid removal during cell salvage in
caesarean section patients. Anesthesiology 2000; 92:
1531–1536
17. UK National Institute for Health and Clinical
Excellence. Intraoperative blood cell salvage in
obstetrics. IP Guidance Number: IPG144. Available
from URL: http://www.nice.org.uk/guidance
18. Davies L, Brown TJ, Haynes S, Payne K, Elliott RA,
McCollum C. Cost effectiveness of cell salvage and
alternative methods of minimising perioperative
allogeneic blood transfusion: a systematic review and
economic model. Health Technology Assessment 2006;
10: 1–228.

165

Section 5
Chapter

13c

Hemorrhagic disorders

Management of obstetric hemorrhage:
hemostatic management
Eleftheria Lefkou and Beverley Hunt

Blood loss
Obstetric hemorrhage (OH) is defined by the World
Health Organization (WHO)1 as a blood loss of more
than 500 ml in the first 24 hours after birth, or of
more than 1000 ml when Cesarean section has been
performed. A more comprehensive definition could
be any blood loss which can provoke a physiological change threatening the woman’s life. According to
the American College of Obstetrics and Gynecology,
OH is defined as either a 10% change in hematocrit
between admission and postpartum, or the need for a
blood transfusion.2
Current best practice for the hematological management of obstetric hemorrhage (OH) emphasizes
the need for speedy and appropriate use of blood components with close monitoring of blood loss. However,
best practice is not always followed. This seems, in part,
to be due to poor understanding in the appropriate use
of blood components and pharmacological agents to
reduce bleeding.
In this chapter we give practical guidelines for the
hematological management of OH. Table 13c.1 shows
the available blood components and their derivatives
used in hemostatic replacement therapy.

Hemostatic replacement therapy
Red cell products
Red blood cell (RBC) transfusion is a first-line intervention to treat the inadequate oxygen delivery (but
not the volume loss) seen in OH. In the UK whole
blood is not usually available.
One unit of packed red cells increases the
hemoglobin by approximately one g/dl and the
hematocrit by 3%. There are no evidence-based guidelines for transfusion of RBC into hemodynamically

166

unstable women with OH. According to the British
Committee for Standards in Hematology guidelines
on the management of massive blood loss, red cell
transfusion is likely to be required when 30%–40%
blood volume is lost; when 40% blood volume loss
is immediately life-threatening.3 As a general rule,
the target hemoglobin levels should be greater than
8 g/dL.
Ideally, all pregnant women should be transfused
with red cells of the same ABO and Rhesus group.4 In
an urgent situation where blood is required immediately, with unknown patient’s blood group, all women
under 50 years must be given group O Rhesus negative
red cells, in order to avoid sensitization and hemolytic
disease of the newborn in subsequent pregnancies.
Every obstetric unit should have 2 units of O-Negative
blood in the fridge for emergency use. However, antenatal ABO and rhesus grouping and sending a sample to the blood transfusion laboratory to confirm
ABO grouping allows the release of matched blood.
The physicians should be mindful that blood grouping takes less than 10 min and so ABO group-specific
red cells should be administrated as soon as possible.

Platelet transfusion
In the UK platelets are usually obtained by plateletpheresis from one donor- single donor plateletpheresis and stored in polyolefin packs with a viability of
about 5 days; or they are removed from a unit of blood
and bagged togethers known as pooled random donor
platelets.
In massive OH after a 1.5–2 × blood volume
replacement, a platelet count ⬍50 × 109 /l should be
anticipated. The target of platelet transfusion is to
maintain platelet count ⬎ 50 × 109 /l (70–110 × 109 /l).
In cases with qualitative platelet abnormalities, as in

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

Chapter 13c. Hemostatic management

Table 13c.1 Blood components and their derivatives used in
hemostatic replacement therapy
r
r
r
r
r
r
r
r
r

Red cells
Platelet pools
Fresh frozen plasma
Cryoprecipitate or fibrinogen concentrates
Recombinant Factor VIIa
Antithrombin, Protein C and activated Protein C
concentrates
Prothrombinase concentrates (II, VII, IX and X)
Plasma-derived and Recombinant Factor VIII and IX
von Willebrand Factor concentrates

some inherited diseases such as Glanzmann’s thrombasthenia or Bernard–Soulier syndrome, or acquired
disorders such as liver or kidney disease, or druginduced platelet dysfunction, the trigger for platelet
transfusion should be higher, depending not on the
number but on the function of platelets. In the UK one
platelet apheresis concentrate will increase the platelet
count by 50 × 109 /L in most adult patients. Ideally, the
platelet count should be checked 10–15 minutes after
platelet infusion to ensure the adequacy of therapy. A
poor increment of less than 20 × 109 /L after 15 minutes in a patient without ongoing bleeding to suggest
the presence of antiplatelet antibodies, usually human
leukocyte antigen (HLA) antibodies.

Fresh frozen plasma (FFP)
Fresh frozen plasma (FFP) is separated within 6–8
hours of whole blood collection, frozen at −18 ◦ C and
stored for up to 1 year. The volume of a typical unit is
200 to 250 ml. FFP contains normal levels of all coagulation factors, except FVIII, which rapidly decays, leaving around 60% levels. The indications for use of FFP
in massive transfusion and disseminated intravascular
coagulation with significant bleeding is PT or APTT
ratio ⬎1.5. There is no evidence base for the dose that
should be used, however 15 mL/kg is widely accepted.
Solvent detergent prepared FFP has a lower risk of
transfusion transmitted infection but has reduced levels of macromolecular von Willebrand factor (VWF),
which is of little concern in the management of
bleeding.

Fibrinogen
There are two sources of fibrinogen available in
the UK: cryoprecipitate and fibrinogen concentrate.
Cryoprecipitate is made from donor plasma by placing the plasma in a fridge at 4 ◦ C. This allows all the
large molecules such as fibrinogen, von Willebrand

factor, and Factor VIII to precipitate out. These are separated off and they are redissolved in a small residue
of plasma, as it is warmed. A typical adult dose is two
five-donor pools (equivalent to 10 single donor units)
containing 3–6 g fibrinogen in a volume of 200 to 500
ml. As a rule of thumb, 10 bags of cryoprecipitate will
increase a normal adult’s fibrinogen by 1 g/L.
Fibrinogen concentrate is available but not licensed
for use in massive transfusion in the UK. Its potential side effects include hypertension, anaphylaxis, and
arterial thrombosis. It is licensed for use in patients
with congenital a- or hypo-fibrinogenemia.
In normal pregnancy fibrinogen levels are elevated
as part of the hemostatic response to pregnancy, with
levels at term between 5 and 7 g/L. So even normal
non-pregnant levels (range: 1.5–4.0 g/L) of fibrinogen
mean that significant consumption of fibrinogen has
occurred.
In OH fibrinogen levels are often very low. With a
poorly contractile uterus or intra-abdominal bleeding,
large volumes of clot form and rapidly consume all the
available fibrinogen. Often, fibrinogen levels are as low
as 0.1 g/L. In this situation there is not enough fibrinogen for normal coagulation to occur. Infusion of fresh
frozen plasma is not enough to replete the deficiency
and so 20 bags of cryoprecipitate should be used as
soon as possible to elevate fibrinogen by approximately
2 g/dL.

Pharmacological agents that
reduce bleeding
Antifibrinolytics
The two agents which have been used in the UK are
tranexamic acid and aprotinin. Aprotinin has been
suspended from marketing in the UK with concerns
about its safety.5
Tranexamic acid binds to plasminogen and thus
inhibits its binding to fibrin. It has a plasma half-life
of 2 hours. It is contraindicated in renal tract bleeding
and in renal failure.
Tranexamic acid has been used extensively to
reduce perioperative bleeding, whether or not there
is evidence of hyperfibrinolysis. A recent Cochrane
review6 shows it is safe in that it is not associated
with an increased risk of venous thromboembolism
with short-term use. However, there are no studies
of the efficacy and use of antifibrinolytics in OH,
but theoretically the hemostatic changes of obstetric

167

Section 5. Hemorrhagic disorders

hemorrhage should be little different from surgical
bleeding and trauma. We know that massive bleeding will stimulate epinephrine, which will cause release
of fibrinolytic activators. If there is a low fibrinogen,
clot formation will be defective and the clot is open
and more liable to being penetrated by fibrinolytic
activators. The use of tranexamic acid in traumatic
bleeding is being investigated in the CRASH-2 (clinical randomization of antifibrinolytic therapy in significant hemorrhage), which aims to randomize 20 000
patients to tranexamic acid vs. placebo and will report
in 2010. The same group plan to do a similar randomized controlled trial of tranexamic acid in obstetric
hemorrhage (the WOMAN study).
The authors suggest in the interim that tranexamic
acid in a 1–2 g bolus should be strongly considered in
the management of OH in view of its safety and efficacy
in other settings.

Recombinant VIIa

168

Recombinant FVIIa is not adequately studied in OH.
It is licensed in Europe for treatment of hemophilia
patients with inhibitors to factors VIII and IX, and
for patients with Glanzmann’s thrombasthenia, and
FVII deficiency. It has no other licensed indication for
any other group of patients but it has been used “off
license” in the management of bleeding.
There are no prospective, randomized placebocontrolled studies in the use of rFVIIa in OH, but many
case reports. Unfortunately, case reports can lead to
considerable reporting bias with a tendency towards
reporting only positive outcomes. There are, however,
three major studies reporting data in the use of rFVIIa
in this setting. The first study comes from the Northern European Registry 2000–2004, and gives data from
the use of rFVIIa in primary postpartum hemorrhage
(PH), from nine European countries.7 A total of 113
individual cases are presented and the authors conclude that there was some improvement in more than
80% of women and few adverse effects. But it is not
clear that best practice for blood components was
applied prior to use of rFVIIa, i.e that the use of blood
components appropriately would not have resulted in
the same improved outcome. The second study, from
Finland, reports retrospectively the one-center experience on the administration of rFVIIa to 38 parturients.8 The authors conclude that there is no evidence
that the use of rFVIIa was better than standard management with blood components. The last study from

Ireland, reports massive OH in 28 cases, with rFVIIa
use in six patients, in a 3-year period at one institution.9 The authors concluded that there is a need
for resuscitation, surgical intervention and appropriate use of blood products and no place for the routine
use of rFVIIa. Haynes et al.10 summarizes 44 reported
cases with rFVIIa in OH and added four cases from
their experience. Data from this study showed that,
despite the administration of rFVIIa, invasive surgery
or procedures, such as hysterectomy or embolization, remained necessary. From the 48 patients, seven
responded only partially to treatment and three died
despite treatment. A relatively recent systematic review
on the efficacy and safety of rFVIIa for treatment of
severe bleeding conclude that more randomized controlled trials are required to assess the use of rFVIIa for
patients without a pre-existent coagulation disorder
and with severe bleeding.11 A recent Cochrane review
did not find real evidence of its off license use but there
was a trend towards reduced mortality and increased
thromboembolic events.12 The review did not include
any studies of obstetric hemorrhage.
There is current concern about the safety of rFVIIa
in “off licence” indications. A recent meta-analysis
showed an arterial thrombosis rate of 5.6% in those
receiving rVIIa compared with 3% in the placebotreated patients.
Thus the use of rVIIa in OH should ideally be limited to clinical trials or in intractable hemorrhage in
carefully selected patients, where there are adequate
levels of platelets and coagulation factors and bleeding has not resolved despite optimal management and
good transfusion practice.
The current recommended dose is 90 ␮g/kg
repeated up to every 2 hours. Currently, no monitoring is available for rFVIIa therapy. It is important to
remember that the success of rFVIIa is dependent on
several pre-conditions that include:
(a) the presence of adequate platelets (⬎50 × 109 /L)
and coagulation factors (fibrinogen levels ⬎1g/L),
and
(b) the absence of acidemia and hypothermia.

Other products
Prothrombinase complexes contain Factors II, VII, IX,
and X isolated from thousands of units of blood and
stored as a powder that requires rehydration for use.
They are used for the emergency reversal of vitamin K
antagonists, an unlikely prospect in pregnant women.

Chapter 13c. Hemostatic management

Their use in OH has been restricted to cases with inherited or acquired deficiency of coagulation factors.
DDAVP (1-deamino-8-d-arginine vasopressin,
desmopressin) is a non-blood-derived alternative
(a synthetic analog of vasopressin) that retains the
antidiuretic action of the natural hormone and also
stimulates the release of tissue plasminogen factor
(tPA). These effects are used to elevate the plasma
factor VIII and vWF level two- to fourfold above
the baseline, by its release from storage sites. It can
correct the hemostatic defect in mild hemophilia
A or von Willebrand disease (VWD) sufficiently
to cover minor surgery or at a minor bleeding
episode. DDAVP should be used with caution in
women with pre-eclampsia, due to the antidiuretic
effect and to the potential risk of hyponatremia
that can lead to convulsions. Therefore, restriction
of fluid intake is required to accompany its use.
Other side effects comprise mild facial flushing and
headache.
There are insufficient data about the efficacy and
safety of DDAVP for prophylaxis and treatment of OH.
It has been used safely during pregnancy in women for
other indications (see Chapter 14). DDAVP does not
pass into breast milk in significant amounts and so it
may be used in labor and in the postpartum period.

Hematological monitoring
and management
Regular full blood counts (FBCs) and coagulation
screens should be used to guide therapy, with regular (up to hourly) requests with massive loss. The
turnaround time in an average hospital makes near–
patient testing an attractive option. A thromboelastogram (TEG) is an alternative, but is poorly validated
in this setting.
In general, if the bleeding continues and no bleeding point can be found:
r Keep hemostatic monitoring going.
r Consider an antifibrinolytic agent.
r Consider the bleeding history and the possibility
of undiagnosed von Willebrands disease or other
bleeding disorder.
Speedy responses are required to prevent a cycle of
failure to catch up e.g. excessive hemorrhage → depletion of hemostatic factors →further bleeding →, etc.

Table 13c.2 Use of Blood components in OH
Red blood cells:
r Maintain Hb ⬎ 8 g/dL
r One unit of packed red cells increases the hemoglobulin by
1 g/dL and the hematocrit by 3%.
Platelet transfusion:
r Maintain platelet count ⬎ 50 × 109 /L (70 × 109 –110 ×
109 /L).
r If platelet count ⬍50 × 109 /L give one adult therapeutic
dose of platelets
FFP:
r If INR/APTT ratios ⬎ 1.5 give FFP 15ml/kg
Cryoprecipitate:
r Maintain fibrinogen ⬎1.0 g/L
r 10 bags of cryoprecipitate will increase a normal adult’s
fibrinogen by 1.0 g/L

Ideally, a blood warmer should be used to help prevent
hypothermia, and regular blood gases should be performed (see Chapter 13b).

Disseminated intravascular coagulation
DIC is not a usual direct consequence of OH but
rather a complication of appropriate, delayed or inadequate treatment. Delayed and inappropriate treatment will lead to prolonged hypoxia, hypovolemia,
hypothermia or extensive muscle damage and thus
to a DIC-like syndrome. A prolonged PT and aPTT,
thrombocytopenia and low fibrinogen levels (⬍1.0
g/l), are highly suggestive of a developing DIC-like
syndrome.

Post-hemorrhage care
Once bleeding has stopped, it should be remembered
that, in the UK, the most common cause of maternal
direct deaths in UK is thromboembolism.
Mothers that hemorrhage and have operative intervention will have excellent acute phase responses postpartum that will make the blood more prothrombotic and the patients at high risk of venous thromboembolism. Therefore, thromboprophylaxis with a
low molecular weight heparin should be started as
soon as possible postpartum, according to current
guidelines.
Table 13c.2 summarizes the suggested guidelines
for the management of OH. Current best practice
emphasizes the need for speedy and appropriate use of
blood components with monitoring if bleeding continues.

169

Section 5. Hemorrhagic disorders

References
1. Ronsmans C, Graham WJ; Maternal mortality: who,
when, where, and why? Lancet Maternal Survival
Series steering group. Lancet 2006; 368: 1189–1200.
2. ACOG Practice Bulletin: Clinical Management
Guidelines for Obstetrician-Gynecologists Number 76,
October 2006: postpartum hemorrhage. Obstetrics and
Gynecology 2006; 108: 1039–1047.
3. Stainsby D, MacLennan S, Thomas D et al. Guidelines
on the management of massive blood loss. British
Journal of Haematology 2006; 135: 634–641.
4. Handbook of Transfusion Medicine, ed DBL
McClelland, United Kingdom Blood Services, 4th edn.
5. Ferguson A.D., Hebert C.P.A, Mazer C.D et al. A
comparison of aprotinin and lysine analogues in
high-risk cardiac surgery. The BART study. New
England Journal of Medicine 2008; 358: 2319–2331.
6. Henry DA, Carless PA, Moxey AJ et al.
Anti-fibrinolytic use for minimising perioperative
allogeneic blood transfusion. Cochrane Database of
Systematic Reviews 2007; 17: CD001886.
7. Alfirevic Z, Elbourne D, Pavord S et al. Use of
recombinant activated factor VII in primary

170

postpartum hemorrhage: the Northern European
registry 2000–2004. Obstetrics and Gynecology 2007;
110: 1270–1278.
8. Ahonen J, Jokela R, Kortila K. An open
non-randomized study of recombinant activated factor
VIIa in major postpartum haemorrhage. Acta
Anaesthesiologica Scandinavica 2007; 51: 929–936.
9. McMorrow RC, Ryan SM, Blunnie WP et al. Use of
recombinant factor VIIa in massive post-partum
haemorrhage. European Journal of Anaesthesiology
2008; 7: 1–6.
10. Haynes J, Laffan M, Plaat F. Use of recombinant
activated factor VII in massive obstetric haemorrhage.
International Journal of Obstetrics and Anesthesia 2007;
16: 40–49.
11. Levi M, Peters M, Buller HR. Efficacy and safety of
recombinant factor VIIa for treatment of severe
bleeding: a systematic review. Critical Care Medicine
2005; 33: 883–890.
12. Stanworth S, Birchall J, Doree C, Hyde C. Recombinant
factor VIIa for the prevention and treatment of
bleeding in patients without haemophilia. Cochrane
Database of Systematic Reviews 2007; 18: CD005011.

Section 5
Chapter

13d

Hemorrhagic disorders

Management of obstetric hemorrhage:
radiological management
Ash Saini and John F. Reidy

Introduction
Those rare cases of obstetric hemorrhage that do
not respond to conservative measures such as the
use of uterotonics, correction of coagulopathies, laceration repair, evacuation of retained products, and
uterine packing have traditionally been treated by
more invasive surgical methods. These include: application of a uterine compression suture, ligation of
the arterial supply to the uterus and, if all else fails,
hysterectomy.
The use of interventional radiology (IR) to treat
postpartum hemorrhage (PPH) was initially described
in 1979.1 Since then, uterine artery embolization
(UAE) to treat fibroids has become common practice and interventional radiologists have been able
to apply the technique to the treatment of obstetric
hemorrhage. UAE plays a vital role once conservative treatments have failed, avoiding the trauma of
major surgery, reducing transfusion requirements and
importantly preserving fertility. Although there are no
randomized controlled trials, there are a large number of case series which have established arteriography and embolization for PPH to be safe and effective,
with a success rate of 90% and very low complications.2
In the United Kingdom recommendations have been
made that all hospitals with delivery units should have
access to an emergency IR service and to consider early
or prophylactic embolization as an important tool in
the prevention and management of obstetric hemorrhage.3,4
The main indications for embolization are in
the emergency treatment of PPH and electively, in
high risk cases. Embolization has also successfully
been used to treat hemorrhage secondary to ectopic
pregnancy, gestational trophoblastic disease, and
acquired uterine arterio-venous malformations.5–9

Contraindications are all relative and include
coagulopathy, renal failure, and severe contrast
allergy.

The role of interventional
radiology in the management
of emergency postpartum
hemorrhage
There are a variety of indications for the use of
embolization, but it is most commonly used following a vaginal delivery to treat primary PPH secondary
to uterine atony (Table 13d.1). Obstetric units should
have protocols to guide referral (Fig. 13d.1).
Although embolization should ideally be performed on a hemodynamically stable patient, some
leeway exists depending on anesthetic support and the
speed and experience of the local IR service. Early
liaison with interventional radiology and referral for
embolization is critical. In this respect obstetricians
should be aware of the potential delay in treatment
whilst the patient is transferred to an angiographic
suite and plan accordingly. It is often the case that, if
conservative treatments are unsuccessfully employed
for such prolonged periods, the patient is then too
unstable to be transferred to the radiology department
for embolization.
It is essential that the procedure is performed
by an experienced interventional radiologist. As with
other more invasive surgical procedures, it is imperative that the patient is resuscitated by another practitioner, preferably an anesthetist. The procedure is
minimally invasive, carried out under local anesthetic.
Standard transfemoral arterial access is obtained
and specialized catheters inserted into the anterior

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

171

Section 5. Hemorrhagic disorders

Table 13d.1 Indications for the use of interventional radiology
in the treatment of emergency postpartum hemorrhage
Primary postpartum hemorrhage due to:
– atonic uterus post-vaginal delivery (most common, usually
following prolonged labor)
– uterine, cervical, or vaginal tears
– hemorrhage post-Cesarean delivery
– pelvic bleeding in a surgically challenging location, e.g.
broad ligament, pelvic, or vulval hematoma
– undiagnosed placenta previa or accreta
Other indications for the use of emergent embolization:
– hemorrhage following therapeutic or accidental abortion or
interstitial ectopic pregnancy
– secondary post-operative PPH
– acquired uterine arterio-venous malformations caused by
instrumentation after delivery

division of each internal iliac artery. Arteriography
is performed, but it is not essential that bleeding
is demonstrated. If the source of bleeding is identified (usually the uterine or vaginal arteries), selective
catheterization is performed and the vessel embolized
with small particles (1–2 mm) of Gelfoam (Upjohn)
until the antegrade flow of contrast stops (Fig. 13d.2).
Gelfoam is the embolic agent of choice as it creates a
temporary occlusion and is small enough to stop flow
in distal branches but unlikely to reach capillary level
and cause uterine necrosis. If the source of bleeding
remains undetected, or if time and patient anatomy
BEST MEDICAL THERAPY

does not allow selective catheterization, then empiric
embolization of the anterior division of the internal
iliac artery is performed.

The role of interventional
radiology in the elective or
prophylactic management of
postpartum hemorrhage
Even rarer than those cases of unexpected emergency
obstetric hemorrhage treated by embolization, are a
high risk group of patients with abnormalities of placentation in whom massive bleeding at the time of
delivery may occur. In these patients the elective use
of embolization has been advocated.
Placenta accreta occurs when there is abnormally
firm attachment of the placental villi to the uterine wall and remains a formidable clinical challenge,
with patients most at risk of emergency hysterectomy.
Although previously extremely rare, there has been
a tenfold increase in cases, thought to be secondary
to the rise in Cesarean deliveries, in which the uterine incision acts as the nidus for abnormal placentation.10,11 Placenta accreta most commonly occurs
in patients with placenta previa. Three variants are
Hemorrhage stopped

Continued hemorrhage

UTERINE
PACKING/TAMPONADE

CRITICAL
EARLY LIAISON WITH
INTERVENTIONAL
RADIOLOGY

Hemorrhage stopped

Continued hemorrhage
Obstetric and anesthetic review

Unstable patient – unfit
for transfer
Stable patient – fit
for transfer to
angiographic suite

172

UTERINE ARTERY
EMBOLIZATION

SURGICAL THERAPY
• Uterine compression suture
• Hysterectomy

Fig. 13d.1 Example of a protocol for the
management of emergency postpartum
hemorrhage.

Chapter 13d. Radiological management

Fig. 13d.3

Fig. 13d.2 Persistent massive primary PPH following Cesarean
section and uterine packing (∗ ). Right transfemoral arterial access
has been obtained and a catheter placed in the origin of the right
internal iliac artery. Angiography demonstrates active pelvic
hemorrhage (arrow), however the precise source is not
demonstrated (a). Selective arteriography identifies hemorrhage
arising from a vasoconstricted right uterine artery (b), (c) which was
successfully embolized with particles of Gelfoam.

recognized, depending on the depth to which the placenta extends through the myometrium (Fig 13a.1).
In its most severe form the placenta extends beyond
the serosal surface of the uterus to invade neighboring
structures, usually the bladder (placenta percreta).
Effective management starts with a high index
of clinical suspicion and accurate pre-natal diagnosis. Pelvic ultrasound by a skilled ultrasonographer is
reliable in excluding a diagnosis of placenta accreta.
In cases with suspicious but inconclusive ultrasonographic findings, magnetic resonance imaging (MRI)
may be used to optimize diagnostic accuracy.12 Once
diagnosed, it is vital that a multi-disciplinary approach
is adopted, with appropriate anesthetic, hematologi-

cal, IR, and urological support as necessary. Patients
are treated either by elective Cesarean hysterectomy
or, if possible, by Cesarean delivery, for example, in
cases of placenta percreta where attempts to perform a hysterectomy would lead to further blood
loss. Although based on very small numbers of cases,
IR has an important role in helping to minimize
hemorrhage.5,13–16
Bilateral transfemoral arterial access is obtained
and pre-delivery catheterization of the anterior division of both internal iliac arteries with occlusion balloons performed (Fig. 13d.3). The balloons are deflated
throughout the delivery and then inflated immediately afterward, thus allowing time to better control the hemorrhage surgically. Alternatively, Gelfoam
embolization can then be performed non-selectively or
selectively through a micro-catheter passed co-axially
through the lumen of the occlusion balloon catheter
and into the uterine arteries. Although the efficacy of
this technique remains to be fully determined, it is
likely to represent the safest form of combined management in a group of patients that have a higher risk of
hemorrhage and are more likely to undergo an emergency hysterectomy.17

173

Section 5. Hemorrhagic disorders

174

Complications of
embolization and effects
on fertility

series have reported a return to normal menses with
no significant adverse effects on future fertility.18–20

Minor complications related to the puncture site
include hematoma or pseudoaneurysm formation, but
are uncommon since patients are usually young and
free of vascular disease. Major complications are even
rarer with anecdotal reports of pelvic sepsis, uterine necrosis, bladder necrosis, and transient buttock
ischemia.2
A prime advantage of embolization is that it
avoids the need for hysterectomy. Although no large
prospective studies have been completed, several case

The techniques of selective arterial embolization to
treat emergency obstetric hemorrhage, and balloon
occlusion used electively in high risk patients are
proven safe and effective methods of treatment for
obstetric hemorrhage. They can reduce transfusion
requirements, preserve fertility and thus have the
potential to reduce maternal morbidity and mortality.
Obstetric departments should have protocols for management that include early referral to IR and consideration of embolization prior to surgery.

Summary

Chapter 13d. Radiological management

References

delivery. Obstetrics and Gynaecology 2006; 107:
1226–1236.

1. Brown BJ, Heaston DK, Poulson AM et al.
Uncontrollable postpartum bleeding: a new approach
to hemostasis through angiographic arterial
embolisation. Obstetrics and Gynaecology 1979; 54:
361–365.

12. Warshak CR, Eskander R, Hull AD et al. Accuracy of
ultrasonography and magnetic resonance imaging in
the diagnosis of placenta accreta. Obstetrics and
Gynaecology 2006; 108: 573–582.

2. Doumouchtis SK, Papageorgiou AT, Arulkumaran S.
Systematic review of conservative management of
postpartum haemorrhage: what to do when medical
treatment fails. Obstetrical and Gynaecological Survey
2007; 68: 540–547.

13. Weeks SM, Stroud TH, Sandhu J et al. Temporary
balloon occlusion of the internal iliac arteries for
control of hemorrhage during caesarean hysterectomy
in a patient with placenta previa and placenta increta.
Journal of Vascular and Interventional Radiology 2000;
11: 622–624.

3. Investigation into 10 maternal deaths at, or following
delivery at, Northwick Park Hospital, North West
London NHS Trust, between April 2002 and April
2005. London: Healthcare Commission, 2006.
4. The role of emergency and elective interventional
radiology in postpartum haemorrhage – good practice
No. 6 Royal College of Obstetricians and
Gynaecologists, June 2007.
5. Mitty HA, Sterling KM, Alvarez M et al. Obstetric
hemorrhage: prophylactic and emergency arterial
catheterisation and embolotherapy. Radiology 1993;
188: 183–187.
6. Kerr A, Trambert J, Mikhail M et al. Preoperative
transcatheter embolization of abdominal pregnancy:
report of three cases. Journal of Vascular and
Interventional Radiology 1993; 4: 733–735.
7. Lobel SM, Meyetovitz MF, Benson CC et al.
Preoperative angiographic uterine artery embolization
in the management of cervical pregnancy. Obstetrics
and Gynaecology 1990; 76: 938–941.
8. Pearl ML, Braga CA. Percutaneous transcatheter
embolization for control of life-threatening pelvic
hemorrhage from gestational trophoblastic disease.
Obstetrics and Gynaecology 1992; 80: 571–574.
9. Badawy SZ, Etman A, Singh M et al. Uterine artery
embolization: the role in obstetrics and gynaecology.
Clinical Imaging 2001; 25: 288–295.
10. Wu S, Kocherginsky M, Hibbard JU. Abnormal
placentation: twenty-year analysis. American Journal of
Obstetrics and Gynaecology 2005; 192: 1458–1461.
11. Silver RM, Landon MB, Rouse DT et al. Maternal
morbidity associated with multiple repeat cesarean

14. Levine AB, Agarwal R, Seckl MJ et al. Placenta accreta:
comparison of cases managed with and without pelvic
artery balloon catheters. Journal of Maternal and Fetal
Medicine 1999; 8: 173–176.
15. Dubois J, Garel L, Grignon A et al. Placenta percreta:
balloon occlusion and embolization of the internal
iliac arteries to reduce intra-operative blood losses.
American Journal of Obstetrics and Gynaecology 1997;
176: 723–726.
16. Hansch E, Chitkara U, McAlpine J et al. Pelvic arterial
embolization for control of obstetric hemorrhage: a
five year experience. American Journal of Obstetrics
and Gynaecology 1999; 180:1454–1460.
17. Zaki ZM, Bahar AM, Ali ME et al. Risk factors and
morbidity in patients with placenta previa accreta
compared to placenta previa non-accreta. Acta
Obstetrics and Gynaecologica Scandinavica 1998; 77:
391–394.
18. Ornan D, White R, Pollak J et al. Pelvic embolization
for intractable postpartum hemorrhage: long-term
follow-up and implications for fertility. Obstetrics and
Gynecology 2003; 102: 904–910.
19. Salomon LJ, deTayrac R, Castaigne-Meary V et al.
Fertility and pregnancy outcome following pelvic
arterial embolization for severe post-partum
haemorrhage. A cohort study. Human Reproduction
2003; 18: 849–852.
20. Descargues G, Mauger Tinlot F et al. Menses, fertility
and pregnancy after arterial embolization for the
control of postpartum haemorrhage. Human
Reproduction 2004; 19: 339–343.

175

Section 5
Chapter

Hemorrhagic disorders

Inherited disorders of primary hemostasis
Sue Pavord

Introduction
The management of inherited bleeding disorders during pregnancy, delivery, and the postpartum period
is particularly challenging. Consideration should be
given to the inheritance risk to the fetus and the bleeding risk to the mother, with appropriate multidisciplinary management plans to minimize complications
for both. Good communication among the haematologists, obstetricians, anesthetists, neonatologists, and
labor ward staff is required, as well as full information
for the patient. This should begin prior to conception
and be reviewed as pregnancy advances. Guidelines for
management have been provided by a task force of the
UK Haemophilia Centre Doctors’ Organization.1

Von Willebrand disease
Von Willebrand disease (VWD) is the most common of the inherited bleeding disorders. It is characterized by a deficiency of Von Willebrand factor
(VWF), a large multimeric glycoprotein, which plays
a crucial role in the first steps of thrombus formation.
Pregnancy, delivery, and the postpartum period pose
significant challenges to the hemostatic system, and
women with VWD need to be carefully managed during these at-risk times.

VWF activity
The functions of VWF are twofold:
r It mediates platelet adhesion and aggregation at
sites of vascular damage, initially by forming a
bridge between the platelet Gp1b receptor and the
subendothelial collagen fibers, exposed by the
injured vessel.

176

r It acts as a carrier for FVIII, protecting it from
proteolysis and facilitating its cofactor activity by
transportation to the site of vascular injury.
The synthesis by endothelial cells and subsequent
secretion from its storage site in the Wiebel Palade
bodies determines the plasma concentration of VWF.
A separate supply is synthesized by megakaryocytes
and stored in the alpha granules of platelets, from
where it is released during platelet activation, providing a rapid and local increase to levels during vessel repair. These actions are particularly important in
sites of fast flowing blood and high shear forces, as
occurs in the arterial circulation and the microvasculature. In these conditions, the globular VWF molecule
is dragged into an elongated shape, exposing its platelet
binding sites, whilst in the venous system where blood
flows more slowly, fibrinogen-dependent clot formation predominates.

Clinical features
The principal clinical manifestations of VWD reflect
the dual function of VWF. Reduced activity of VWF,
leading to impaired platelet plug formation gives rise
to bleeding from mucosal surfaces, typical of the
thrombocytopathies, whilst the rapid clearance of the
unprotected factor VIII impedes fibrin clot formation, causing symptoms characteristic of the coagulopathies, such as prolonged bleeding after surgery.
Both these effects have potentially serious implications
for women in pregnancy.

Disease prevalence
VWF is encoded by a gene spanning 178 kb of genomic
DNA on the short arm of chromosome 12. Numerous
genetic mutations, affecting VWF production, have
been described2 and VWD has an estimated frequency

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

Chapter 14. Inherited disorders of primary hemostasis

Table 14.1 Subtypes of von Willebrand disease

Subclassification

Multimeric pattern

VWF function

Specific characteristics

Type 1

Normal

Quantitative deficiency of VWF.
Accounts for about 70% of all cases.

Type 2a

Absent HMW multimers

Abnormal

Impaired platelet adhesion

Loss of HMW multimers through
increased platelet binding

Abnormal

Increased affinity for platelets causing
thrombocytopenia

Normal

Abnormal

Impaired platelet adhesion

Normal

Abnormal FVIII binding

Reduced affinity for FVIII with low FVIII levels due
to short half-life

Type 3

Absent

Severe bleeding phenotype

HMW = high molecular weight.

in the population of around 1%, based on the number of people with bleeding symptoms, low VWF, +/−
a positive family history. Clinical penetrance of the
genetic abnormalities is variable;3 in some cases they
are fully penetrant, accounting for the low VWF levels and bleeding phenotype, but in others the VWF
mutation may simply act as a risk factor for bleeding
in combination with other modifying factors, such as
platelet dysfunction or the presence of blood group O,
which is typically associated with 25% lower levels than
the other blood groups.4 In other cases, classic VWD
mutations have not been identified but VWF may still
play a role. Thus, those with a clinically significant
bleeding phenotype amount to only about 0.02% of
the population. These patients are usually diagnosed in
childhood, whereas the milder forms may not present
until after significant hemostatic challenges, such as
menstruation and childbirth. This probably explains
the misconception in the original reports by Erik von
Willebrand, describing women as twice as likely to be
affected than men.

Classification of VWD
VWF is present in the plasma as a series of multimers,
assembled from varying amounts of identical subunits.
The composition of multimers, which range from 150
000 to 20 000 000 Daltons, has been used to classify
VWF into its different subtypes (Table 14.1). Its adhesive function is largely dependent on the high molecular weight (HMW) multimers, which are released during platelet and endothelial cell activation.
Type 3 disease is characterized by unmeasurable
VWF levels and consequently, severely low FVIII levels, with median FVIII levels being around 4%. Thus,

in addition to clinical features of impaired primary
hemostasis, these patients behave like those with moderate hemophilia, with potential for spontaneous joint
and muscle bleeds. Transmission is autosomal recessive with patients being homozygous or double heterozygous for the abnormal VWF gene, inherited from
asymptomatic parents. Prevalence in the UK is around
1 per million, being most frequent in communities
where consanguinous marriages are common.

Laboratory evaluation
Routine coagulation screening tests including the prothrombin time (PT) and activated partial thromboplastin time (APTT) do not detect VWD unless the
factor VIII level is below normal, prolonging the
APTT. Specific assays for FVIII activity and VWF antigen and activity are available, the latter including Ristocetin cofactor activity and collagen binding assay.
Platelet function can be assessed by PFA-100, which
measures the time taken for platelets to close over
a hole in a collagen membrane coated with ADP or
epinephrine. However, all assays are limited in their
specificity and sensitivity and none show good correlation with the severity of bleeding. Furthermore, levels can be influenced by external factors such as physical and mental stress. Thus, despite the numerous tests
available, the diagnosis of VWD and its subclassification is often difficult.

Hormonal influences on levels in pregnancy
Levels of von Willebrand factor and factor VIII start to
rise from 6 weeks’ gestation, increasing progressively
throughout pregnancy to three to five times baseline
levels by delivery.5 This is due to increased synthesis

177

Section 5. Hemorrhagic disorders

of VWD, although the cause for the increase in FVIII
is not entirely clear but in part reflects improved stabilization by VWF. This rise is beneficial to many patients
with type 1 VWD, in whom normal levels are often
reached by delivery. However, those with starting baseline levels of ⬍15 IU/dL may fail to reach normal values6 and in patients with type 2 disease, where the
molecule functions abnormally, the condition may not
improve and may even deteriorate.7 This is particularly evident in type 2b where the rise in dysfunctional
VWF protein enhances abnormal platelet binding and
exacerbates thrombocytopenia. In patients with type
2N disease, factor VIII levels tend to remain low
because of impaired binding by the abnormal VWF
and patients with type 3 disease show little or no rise
in VWF.8
A study of VWF levels in 248 healthy women,
showed that they remained elevated for 1–3 days postpartum and then returned to baseline by day 7–21.9
Factor VIII and VWF levels are also influenced
by thyroxine, which shows a physiological rise in its
bound form during pregnancy. Hypothyroidism may
be associated with clinical and laboratory features of
VWD which corrects with thyroid replacement. 10

Obstetric complications
Maternal bleeding
Women with VWD have an increased risk of bleeding
events and even death during childbirth.11 Although
the physiological rise in VWF and FVIII protects many
women with mild type 1 disease during delivery, they
remain vulnerable in early pregnancy and in the postpartum period. Studies have found:

178

r One-third of women with VWD have bleeding
during their first trimester.
r 15%–30% of women with VWD have primary
postpartum hemorrhage.
r Delayed postpartum bleeding occurs in 20%–25%
of women with VWD.
r There is a relatively high frequency of perineal
hematoma, a normally rare complication of
vaginal birth.
r The risk of receiving a blood transfusion is
increased fivefold
r Maternal mortality rate is ten times higher than
that for women without the condition.

Table 14.2 Pre-pregnancy management of VWD
r

r
r
r
r

r
r
r

r
r

Reassess severity of clinical bleeding tendency
including previous responses to hemostatic
challenges.
Check baseline investigations if not already
known
Establish response to DDAVP
Obtain consent for use of plasma products
after full counseling of risks
Where plasma-derived products have been
received in the past, the presence of
transfusion transmitted infection should be
excluded.
Vaccinate against hepatitis A and B if not
already immune
Check hemoglobin and serum ferritin and give
oral supplements as necessary
All should receive counseling about risks of
increased bleeding, particularly in the
postpartum period and particularly for women
with type 2 or 3 VWD
A management plan should be discussed with
all patients.
All patients should be offered genetic
counseling as they are at risk of delivering an
affected child
All should receive explanation regarding
evaluation of the infant after delivery

Pregnancy outcomes
Women with VWD are no more likely to experience
premature labor, placental abruption, fetal growth
restriction or intrauterine fetal death.11 Early miscarriage has not been shown to be any more frequent than
in the general population, but can be complicated by
significant bleeding.8,12

Pregnancy management for women
with VWD
The safe management of women with VWD requires
good communication between the hematologist,
obstetricians, anesthetists, neonatologists, and labor
ward staff. The patient should be fully informed of
potential bleeding risks and the plan for management
of pregnancy, delivery, and the postpartum period.
This should begin prior to conception and should be
reviewed as pregnancy advances (Table 14.2).

Chapter 14. Inherited disorders of primary hemostasis

Ante-natal management
In view of the physiological rise in factor VIII and
VWF during pregnancy, most women with mild type
1 VWD achieve levels above 50 IU/dL, the lower limit
of the normal range outside of pregnancy, and can be
safely managed in standard obstetric units in collaboration with hemophilia center staff. Women with types
2 and 3 VWD, or moderate or severe type 1, or a history of severe bleeding, should be referred for pre-natal
care and delivery to a center where there are specialists
in high risk obstetrics, as well as a Hemophilia Center.
Laboratory, pharmacy, and blood bank support is also
essential.
For all types of VWD, levels should be checked routinely at booking, 28 weeks and if still abnormal, 34
weeks’ gestation. If an adequate rise is demonstrated,
only a third trimester sample may be necessary for subsequent pregnancies, unless earlier interventions are
required.
Levels are always needed prior to invasive procedures such as chorionic villus sampling, amniocentesis, or cervical cerclage. If VWF activity or FVIII
levels are ⬍50 IU/dL, women should receive prophylaxis. DDAVP should be used in preference to plasma
derived products in type 1 VWD, to avoid potential for
transfusion transmitted infections (Table 14.3).

DDAVP in pregnancy
DDAVP (1-deamino-8-D-arginine vasopressin) a synthetic derivative of antidiuretic hormone that acts
specifically through type 2 vasopressin receptors,
stimulates release of ultralarge VWF multimers from
storage in the Wiebel Palade bodies of the endothelial
cells. This is not a direct stimulatory effect, but is mediated through intracellular calcium mobilization and
cyclic adenosine monophosphate. Administration has
traditionally been by slow intravenous infusion, over
20 minutes, of 0.3 ␮g per kilogram of body weight,
although a subcutaneous preparations are now commonly used, with similar efficacy and fewer side effects.
Intranasal preparations are also available. Administration results in a three to fivefold increase in both
VWF and factor VIII, within 30–60 minutes, lasting
for 8–10 hours. To assess the response to DDAVP,
VWF activity levels and factor VIII should be measured before administration and again at 1 and 4
hours after, to determine peak levels and clearance rate,
respectively.

Table 14.3 Ante-natal management of VWD
r

Check VWF antigen and activity and FVIII levels
at booking, 28 and if still abnormal, 34 weeks’
gestation and prior to any invasive procedure.
In patients with type 2B VWD, the platelet
count should also be monitored. Platelet
transfusions, as well as VWF factor replacement
may sometimes be required for bleeding or to
cover surgical procedures and spontaneous
miscarriage.
Aim for FVIII and VWF:RCo activity levels of ≥
50% to cover surgical procedures or
spontaneous miscarriage.
Treat with desmopressin in preference to
coagulation factor concentrates whenever
possible, checking pre- and post-treatment
VWF activity levels and factor VIII.
Distribute action plan for acute bleeding
events, to hematology and obstetric staff and
ensure patient is given an emergency number
for contact.

DDAVP is generally thought to be safe for mother
and fetus13 and previous concerns regarding the
potential risk of uterine contractions or neonatal
hyponatremia have diminished, in view of its selective effect on V2R receptors. Fluid intake should
be restricted to 1 liter for the following 24 hours,
to prevent maternal hyponatremia caused by water
retention from the antidiuretic effect. An in vitro
placenta model showed that DDAVP, at therapeutic dose, did not cross the placenta in detectable
amounts.14
The advantages of DDAVP are its low cost, unlimited availability, and most importantly, the avoidance
of blood products. However, there are many situations
where DDAVP will be contraindicated or ineffective
and plasma products necessary (Table 14.4). In these
patients, prophylactic treatment with a clotting concentrate containing Factor VIII and von Willebrand
Factor should be considered to raise levels ⬎50 IU/dL
for ante-natal procedures and childbirth. Patients with
type 2B disease may also require platelet transfusion if
thrombocytopenia is severe.

Coagulation factor replacement
There are several licensed plasma-derived VWF/FVIII
products available. The spectrum of VWF HMW

179

Section 5. Hemorrhagic disorders

Table 14.4 Situations where DDAVP may not be suitable

180

Patients with insufficient baseline
levels

Patients with baseline VWF or FVIII levels of less than 15 IU/dl may not
achieve post-infusion levels, which are sufficient to control or
prevent bleeding.

Some subtypes of Type 1, including
Vicenza subtype

Some subtypes of type 1 VWD show decreased survival of
endogenously produced VWF following DDAVP compared with
normal survival of exogenously administered VWF.

Previous intolerance or severe
adverse effects

During intravenous infusion, hypotension, headache and facial
flushing are common but generally mild. Blood pressure should be
monitored during and after infusion.

Known cardiovascular disease,
pre-eclampsia or unstable blood
pressure

There are anecdotal reports of myocardial and cerebral infarction and
DDAVP should be avoided in patients known to have arterial disease.

Tachyphylaxis after repeated doses

The response to DDAVP diminishes after repeated doses.

Type 2 VWD (except some cases of
2M)

DDAVP is less likely to correct the abnormality in type 2 VWD,
although there may be a transient shortening of the bleeding time.
Because the molecule is abnormal, the usual increase in the amount
of VWF protein following DDAVP is unlikely to improve function.

Type 2B

The heightened and spontaneous binding of the abnormal VWF
molecule to normal platelets may be aggravated by the rise in VWF
levels after DDAVP, increasing platelet clearance from the circulation
and exacerbating thrombocytopenia.

Type 2N

DDAVP also causes release of Factor VIII from stores. However, due to
the abnormal VWF : FVIII binding, the response to DDAVP is
shortened to a median of 3 hours.

Type 3 VWD

These patients lack releasable stores of VWF and do not respond to
DDAVP.

multimers and ratio of VWF:RCo/FVIII activity differs between them, but this does not appear to
cause a difference in efficacy.15 They are available as
lyophilized powders and, after reconstitution in water,
can be administered by slow bolus intravenous injection. Treatment is usually given 1 hour pre-operatively.
Pre-and post-levels should be checked and therapeutic
levels of FVIII and VWF:RCo ⬎50 IU/dL maintained
until hemostasis is secure. For most ante-natal procedures, a single pre-operative treatment is sufficient, but
in some cases a second dose may be required at 12–24
hours, depending on the nature of the procedure and
the measured levels.

Intrapartum management
Although there are no large prospective studies that
correlate VWF:RCo and FVIII levels with the risk
of bleeding at the time of childbirth, the opinion of experts is that levels above 50 IU/dL should
be achieved before vaginal delivery or Cesarean
section.1,16
Neonates are at risk of intracranial hemorrhage
and cephalhematomas during labor and delivery. The
increase in FVIII and VWF, induced by the stress of
labor, provides some protection for babies with mild
type 1 disease but in more severe types, trauma to the
baby should be minimized by avoiding extracephalic

Chapter 14. Inherited disorders of primary hemostasis

Table 14.5 Intrapartum management

Table 14.6 Postpartum management

r
r

Allow spontaneous labor and normal vaginal
delivery, if no other obstetric concerns, to
minimize risk of intervention.
If FVIII or VWF:RCo activity levels ⬍50 IU/dl at
the last check, the test needs to be repeated.
If levels ⬍50 IU/dL, treat with DDAVP if known
responder, otherwise plasma-derived factor
concentrate. Treatment should be given at the
onset of established labor and pre- and
post-treatment levels should be obtained.
Avoid prolonged second stage of labor, with
early recourse to Cesarean section if necessary,
to reduce risk of trauma to mother and baby
and risk of uterine atony.
For fetuses at risk of having type 2 or 3 disease
or moderately severe type 1, avoid fetal blood
sampling, fetal scalp monitoring, Ventouse
delivery and mid-cavity or rotational forceps.
Avoid aspirin and consider alternatives for
NSAIDs. Intramuscular injections may be
suitable if FVIII and VWF activity and PFA-100
are in the normal range.
Active management of the third stage of labor
and early suturing of episiotomy and
lacerations.

version, Ventouse delivery, fetal blood sampling, scalp
electrodes, and rotational forceps.

Analgesia
There is no consensus on the levels that are safe for
regional anesthesia but if FVIII and VWF activity levels are ⬎ 50 IU/dL, regional anesthesia may be undertaken. Consideration should also be given to the levels
at the time of catheter removal and repeat treatment
given beforehand if necessary. Intramuscular injections are not contraindicated if FVIII and Von Willebrand activity are shown to be normal, but attention
should be given to the prolonged antiplatelet effect
of non-steroidal anti-inflammatory drugs, being 2–3
days for indomethacin and diclofenac. Aspirin has an
irreversible effect on platelets, but following ibuprofen
normalization of platelet aggregation occurs within 24
hours. Before delivery, all women with VWD should
have the opportunity to discuss analgesia with an anesthetist.

r
r
r

Ensure careful surgical hemostasis and effective
uterine contraction in all cases.
Repeat VWF activity levels and Hb prior to
discharge.
Give prophylactic tranexamic acid.
For patients with significantly low
pre-pregnancy levels, consider DDAVP if known
responder.
For types 2 and 3 disease or severe type 1,
ensure VWF activity levels are maintained at
⬎50 IU/dL for 3 days following vaginal delivery
or 5 days if Cesarean section has been
performed.
Maintain regular contact with the patient after
discharge and encourage them to report
excessive blood loss.
Consider use of combined oral contraceptive
pill if excessive bleeding is ongoing despite
prophylaxis. This is particularly beneficial to
patients with type 1 disease due to the
associated increase in functional VWF protein.

Postpartum management (Table 14.6)
The postpartum fall in FVIII and VWF levels is variable, occurring between 24 hours and 2 weeks after
birth. In normal pregnancies, the median duration of
bleeding after childbirth is 21 to 27 days, with delayed
or secondary postpartum hemorrhage occurring in
fewer than 1% of cases. In women with VWD this is
much more common, affecting 20%–25% of cases. In
addition, there are multiple cases of postpartum hemorrhage that have occurred despite prophylaxis. The
average time of presentation of postpartum hemorrhage in women with VWD is 10–20 days after delivery.17 Women with mild type 1 disease should be
encouraged to report excessive bleeding but, for more
severe cases, hemoglobin should be monitored and
regular contact with the patient maintained for several
weeks.

Tranexamic acid
Patients with mild type 1 disease can usually be
safely managed with tranexamic acid alone. This is
a lysine analog, which saturates lysine binding sites
on plasminogen and prevents them from interacting
with the fibrin surface, thus inhibiting fibrinolysis.

181

Section 5. Hemorrhagic disorders

It has proven efficacy in reducing blood loss, without increasing thrombotic risk. It is contraindicated in
patients with hematuria and doses should be reduced
in renal failure.
Tranexamic acid crosses the placenta and should
generally be avoided during pregnancy, although it
has been used to treat ante-natal bleeding in a limited
number of cases without adverse fetal effects reported.
Traces have been found in breast milk but this has not
been associated with changes to neonatal fibrinolytic
activity.

DDAVP and plasma products
For patients with significantly low pre-pregnancy levels, DDAVP can be given at the time of cord clamping, although as the peak effect is 40 to 60 minutes after
administration, it may be more beneficial if administered during the second stage of labor or immediately
before Cesarean section. DDAVP may be used to raise
factor levels in responders, but care must be taken in
its administration at the time of childbirth and extra
fluids should be avoided. Tranexamic acid is a useful
adjunct to desmopressin, particularly as it counteracts
the mild fibrinolytic effect of DDAVP related to the
associated rise in tissue plasminogen activator.
All patients with type 3 and most with type 2
disease require plasma derived VWF concentrates, to
maintain levels ⬎50 IU/dl for at least 3 days after vaginal delivery and 5 days following Cesarean section.1
These patients also usually require prolonged administration of tranexamic acid and close monitoring
(Table 14.6).

Potential complications
of factor replacement
Transfusion transmitted infections

182

To minimize risk of viral transmission, two independent and effective steps which complement each other
in their mode of action, are incorporated into the
plasma product manufacturing process. These include
dry heat treatment at 80 ◦ C for 72 h, pasteurization
at 60 ◦ C for 10 h, or solvent detergent (SD) treatment with tri(n-butyl) phosphate and Tween-80 or Triton X. A third step of nanofiltration has been introduced for some products. No cases of HIV, hepatitis
B, and hepatitis C have occurred with products inactivated by the currently used processes; however, some
viruses, such as parvovirus B19, are relatively resist-

ant to all these inactivation techniques. Parvovirus
infection can have serious consequences in pregnancy,
being associated with hydrops fetalis and intrauterine
fetal death. In addition, new emerging infections as
well as those such as vCJD, capable of crossing between
species, will remain potential infective risks.

Inhibitor formation
Alloantibodies to exogenous VWF are a rare complication of treatment and more likely to occur in patients
with type 3 disease, associated with large or complete
VWF gene deletions or stop codons. The prevalence
in these patients is around 8%. The antibodies render
replacement therapy ineffective and can cause severe
anaphylactic reactions.

Thrombosis
Excessive accumulation of FVIII may arise after
repeated administration of Von Willebrand factor containing concentrates. Resulting thrombosis has been
reported but mostly these cases were peri-operative
without use of monitoring.
It is advised that when using VWF containing concentrates peri-operatively, monitoring of FVIII:C and
VWF:RCo should be used in deciding dosing of therapy and excessive FVIII levels avoided. Mobilization
and hydration should be encouraged and anti-embolic
or stockings considered. Pharmacological thromboprophylaxis should generally be avoided, particularly
with type 3 disease.

Neonatal management
Being an autosomal dominant condition in most cases,
the risk of transmission is 50%. However, the variable
penetrance of type 1 VWD results in only around one
third being clinically affected. Type 3 disease is autosomal recessive giving a 25% risk if a previous sibling
has been affected. The risk of peri-natal intracranial
hemorrhage is low, even in neonates with VWD type 3.
Nevertheless newborns at risk of moderate and severe
types need to be tested for VWD using cord blood and
assessed to exclude intracranial hemorrhage. For diagnostic purposes, however, levels are unreliable in most
cases, being artificially low due to the gestational age
or increased from the stress of labor and delivery and
need repeating at 6–12 months when adult values are
reached (Table 14.7).

Chapter 14. Inherited disorders of primary hemostasis

Table 14.7 Neonatal management
r

If severe disease phenotype is expected, a cord
sample should be tested for FVIII level and VWF
activity. The limitations of testing at this stage
should be understood.
r Babies with type 2B disease may require
platelet transfusion if there is severe
thrombocytopenia or bruising/ bleeding
manifestations.
r Intramuscular vitamin K should be avoided until
results are known and given orally if necessary.
Any heel prick tests should have pressure
applied afterwards for 5 minutes.

Inherited disorders of platelet
function
There are a number of platelet function disorders and
for most cases management requires assessment of
maternal bleeding phenotype with consideration given
to use of platelets to cover ante-natal procedures and
delivery, DDAVP may be used if response has been
previously demonstrated. Patients with bleeding histories should be given tranexamic acid for 5–14 days
postpartum. If there is a neonatal risk of platelet dysfunction, traumatic delivery should be avoided and if
thrombocytopenia is a feature of the condition, a cord
sample should be taken at birth. Special mention is
given to Glanzmann’s thrombasthenia and Bernard–
Soulier syndrome.

Glanzmann’s thrombasthenia
Glanzmann’s thrombasthenia is a congenitally
acquired platelet disorder with an autosomal recessive
mode of inheritance. Platelets are normal in number,
but their ability to aggregate is reduced due to loss of
the surface receptor glycoprotein IIbIIIa. Pregnancy
and delivery are rare in these patients but have been
associated with a high risk of severe postpartum
hemorrhage.
Recombinant activated factor VIIa is licensed for
use in patients with this disorder. In pharmacological concentrations, FVIIa is capable of binding to the
surface of activated platelets and improving thrombin generation to enhance adhesion and aggregation
of platelets lacking GPIIb/IIIa. The usual dose given
is 90 mcg/kg 2–3 hourly. Bleeding can also be suc-

cessfully prevented by transfusion of platelets before
and after delivery. However, platelet transfusion can
stimulate isoantibody formation against glycoprotein
IIb–IIIa, resulting in a decreased efficacy of subsequent transfusions. A single donor platelet preparation
should be used in preference to pooled platelet transfusion to reduce this risk and where possible should be
HLA matched.
Delayed bleeding up to 2–3 weeks postpartum has
been reported and in these circumstances, DDAVP
and tranexamic acid are useful to reduce platelet transfusion requirements.

Neonatal management
Unless the father has the same condition, the fetus is
heterozygous, with platelets carrying specific paternal
antigens that are not present on the maternal platelets
and thus are capable of causing maternal alloimmunization. Transplacental transfer of the maternal
antiplatelet immunoglobulin G antibodies can lead to
severe isoimmune neonatal thrombocytopenia and a
risk of intracranial hemorrhage in the fetus. Women
should be monitored for the development of platelet
specific antibodies.

Bernard–Soulier Syndrome
The Bernard–Soulier syndrome (BSS) is a rare autosomal recessive bleeding disorder, characterized by
impaired platelet aggregation with ristocetin and
a normal to decreased number of unusually large
platelets whose membranes lack glycoprotein complex
GP Ib/IX/V. In some patients the disease can go unrecognized until the third or fourth decade.
Four different features of BSS may contribute to the
hemorrhagic diathesis: thrombocytopenia, abnormal
platelet interaction with vWF, impaired platelet interaction with thrombin, and abnormal platelet coagulant
activity. BSS is caused by genetic defects in the genes
of GPIb␣, GPIb␤, GPIX or GPV. This variety of mutations could explain the heterogeneity of the syndrome;
however, the clinical manifestation may even differ in
consecutive pregnancies of the same patient.
The main complications encountered in reported
cases have been antepartum hemorrhage excessive
intra-operative bleeding and immediate and delayed
postpartum hemorrhage, development of maternal
antiplatelet antibodies leading to fetal intracranial
hemorrhage and neonatal alloimmune thrombocytopenia.

183

Section 5. Hemorrhagic disorders

184

Management

Neonatal management

Management is similar to that for Glanzmanns
thrombasthenia and includes the judicious and
timely use of platelet transfusions to prevent bleeding
whilst minimizing the risk of platelet refractoriness.
Regional anesthesia should be avoided and postpartum tranexamic acid and DDAVP prescribed as necessary.

The risk to the fetus is unpredictable but thrombocytopenia can occur due to heterozygosity of the platelet
function disorder and, more significantly, fetomaternal alloimmunization, which may be encountered
even in the absence of demonstrable antibodies. Thus
the management may follow that for fetal/neonatal
alloimmune thrombocytopenia. (See Chapter 5.)

Chapter 14. Inherited disorders of primary hemostasis

References
1. Lee CA, Chi C, Pavord SR et al. UK haemophilia
Center Doctors’ Organization. The obstetric and
gynaecological management of women with inherited
bleeding disorders – review with guidelines produced
by a taskforce of UK Haemophilia Center Doctors’
Organisation. Haemophilia 2006; 12: 301–336.
2. ISTH. SSC VWF Database [Database on the
Internet]. Sheffield, UK: University of Sheffield.
http://www.vwf.group.shef.ac.uk/index.html
(Accessed 23.01.2008).

9. Sanchez-Luceros A, Meschengieser SS, Marchese C.
et al. Factor VIII and von Willebrand factor changes
during normal pregnancy and puerperium. Blood
Coagulation and Fibrinolysis 2003; 14: 647–651.
10. Manfredi E, van Zaane B, Gerdes VE et al.
Hypothyroidism and acquired von Willebrand’s
syndrome: a systematic review. Haemophilia 2008; 14:
423–433.
11. James AH, Jamison MG. Bleeding events and other
complications during pregnancy and childbirth in
women with von Willebrand disease. Journal of
Thrombosis and Haemostasis 2007; 5: 1165–1169.

3. Goodeve A, Eikenboom J, Castaman G et al.
Phenotype and genotype of a cohort of families
historically diagnosed with type 1 von Willebrand
disease in the European study, Molecular and Clinical
Markers for the Diagnosis and Management of Type 1
von Willebrand Disease (MCMDM-1VWD). Blood
2007; 109: 112–121.

12. Foster PA. The reproductive health of women with von
Willebrand Disease unresponsive to DDAVP: results of
an international survey. On behalf of the
Subcommittee on von Willebrand Factor of the
Scientific and Standardization Committee of the ISTH.
Thrombosis and Haemostasis 1995; 74: 784–790.

4. Gill JC, Endres-Brooks J, Bauer PJ et al. The effect of
ABO blood group on the diagnosis of von Willebrand
disease. Blood 1987; 69: 1691–1695.

13. Mannucci PM. Use of desmopressin (DDAVP) during
early pregnancy in factor VIII-deficient women. Blood
2005; 105: 3382.

5. Stirling Y, Woolf L, North WR, Seghatchian et al.
Haemostasis in normal pregnancy. Thrombosis and
Haemostasis 1984; 52: 176–182.

14. JG Ray, R Boskovic, B Knie et al. In vitro analysis of
human transplacental transport of desmopressin.
Clinical Biochemistry 2004; 37: 10–13.

6. Ramsahoye BH, Davies SV, Dasani H, Person JF.
Obstetric management in von Willebrand’s disease: a
report of 24 pregnancies and a review of the literature.
Haemophilia 1885; 1: 140–144.

15. Mannucci PM, Tenconi PM, Castaman G, Rodeghiero
F. Comparison of four virus-inactivated plasma
concentrates for treatment of severe von Willebrand
disease: a cross-over randomized trial. Blood 1992; 79:
3130–3137.

7. Conti M, Mari D, Conti E et al. Pregnancy in
women with different types of von Willebrand
disease. Obstetrics and Gynaecology 1986; 68:
282–285.
8. Kadir RA, Lee CA, Sabin CA et al. Pregnancy in
women with von Willebrand’s disease or factor XI
deficiency. British Journal of Obstetrics and
Gynaecology 1998; 105: 314–321.

16. Pasi KJ, Collins PW, Keeling DM et al. Management of
von Willebrand disease: a guideline from the UK
Haemophilia Center Doctors’ Organization.
Haemophilia 2004; 10: 218–231.
17. Roque H, Funai E, Lockwood CJ. von Willebrand
disease and pregnancy. The Journal of Maternal–Fetal
Medicine. 2000; 9: 257–266.

185

Section 5
Chapter

Hemorrhagic disorders

Inherited coagulopathies
Sue Pavord

Hemophilia
Introduction
Hemophilia is characterized by a deficiency of factor
VIII (hemophilia A) or factor IX (hemophilia B), both
key components of the intrinsic pathway of the coagulation cascade. The gene is carried on the long arm
of the X chromosome, so males are clinically affected
and females are carriers. Female carriers may also have
low factor levels due to skewed X chromosome inactivation, giving rise to an increased tendency to bleed.
Thus management of pregnancy requires the assessment of bleeding risk for both mother and baby, with
particular attention given to multidisciplinary planning and co-ordination of healthcare professionals at
the time of, and after, delivery.

Disease incidence
Hemophilia A and B occur with an incidence of
around 1:5000 and 1:10 000 male births, respectively.
The severity of the disease runs true in families and, if
the family history is known, the bleeding risk to male
offspring can be largely predicted (Chapter 16). However, 40%–50% of cases are sporadic and unexpected
with no family history of the condition.

Clinical features
The hallmark of the condition is hemarthrosis, resulting in progressive arthropathies requiring joint fusions
or joint replacement to alleviate pain. The bleeding risk
correlates with the level of coagulation factor (Table
15.1). Patients with severe hemophilia or those with
recurrent joint bleeds require prophylaxis with factor
concentrate twice (factor IX) or thrice (factor VIII)
weekly with an aim to keep trough levels at, or above,
5% and avoid spontaneous bleeds. Acute bleeds require

186

immediate treatment to minimize joint and soft tissue damage. After a period of training and assessment of competency, factor concentrate can be selfadministered using home stocks, although many children on prophylaxis have difficulty with venous access
and require insertion of portacaths, which are often
complicated by recurrent infections.

Hormonal influences on levels
in pregnancy
Levels of factor VIII increase from 6 weeks’ gestation,
to two to three times baseline by term. Factor IX levels
are relatively unaltered.1

Obstetric complications
Maternal bleeding
Female carriers of hemophilia typically have half levels
of factor VIII/IX. Unbalanced lyonization, where there
is uneven X chromosome inactivation, may result in
significantly lower levels. For carriers of hemophilia
A, the pregnancy-induced rise in factor VIII level alleviates any potential problems for childbirth, although
they remain vulnerable in early pregnancy and those
with baseline levels ⬍15 IU/dL may not achieve normal levels by delivery. Women with low factor IX levels
remain at risk of bleeding throughout pregnancy.

Pregnancy outcomes
Miscarriage and placental insufficiency syndromes are
not increased. The main risk is to the neonate at the
time of delivery, as well as the maternal bleeding risk,
particularly postpartum, for those mothers with low
factor levels.

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

Chapter 15. Inherited coagulopathies

Table 15.1 Severity of hemophilia according to factor level

Table 15.2 Ante-natal management of hemophilia carriers

Factor level Severity of
(% activity) clinical condition Bleeding risk

⬍1

Severe

Spontaneous joint and
muscle bleeds

1–5

Moderate

Joint and muscle bleeds
mainly after trauma.
Occasional spontaneous
bleeds

⬎5

Mild

Trauma/surgery induced
bleeding

Neonatal risk
The most significant potential complication for the
neonate is intracranial hemorrhage (ICH), particularly
following instrumental or traumatic birth. The risk is
approximately 50 times greater than for the general
population and affects around 4% of all hemophilia
boys,2 although it is clearly highest in those with severe
hemophilia or where the disease is unexpected and no
preventative strategies, neonatal surveillance, or considered management plan are in place.
ICH is most often associated with extracranial
hemorrhage (ECH) after trauma and any significant
ECH in a newborn should raise the suspicion of underlying coagulopathy and ICH. Common complications
are cephalhematomas and abnormal bleeding after
injection or venepuncture. Other reported events are
umbilical bleeding, hematuria and retro-orbital bleeding.

Pre-pregnancy management
All women with a family history of hemophilia should
be assessed for carrier status, including pedigree profile and calculation of statistical risk, baseline factor
levels and genetic mutation analysis where possible.
(See Chapter 16.)
r Carriers should receive effective counseling
regarding their risk of
(a) bleeding, particularly in the postpartum period;
(b) delivering an affected male.
These risks need to be determined and fully discussed with the patient, including options for pre-natal
diagnosis. Appropriate multidisciplinary management
plans should be agreed to minimize complications for
both mother and baby.

r
r

Check factor levels at booking, 28 and if still abnormal, 34
weeks’ gestation or prior to any invasive procedure.
Aim for FVIII /FIX levels of ≥ 50 U/dL to cover surgical
procedures or spontaneous miscarriage.
For carriers with low factor VIII levels DDAVP may be used
but recombinant factor concentrate is required to raise
factor IX levels.

Ante-natal management (Table 15.2)

r All women should be offered pre-natal diagnosis
(Chapter 16), but women who do not wish for this
should to have the fetal sex determined by
ultrasound when the anomaly scan is
performed.
r Factor levels should be checked at booking, 28 and
if still abnormal, at 34 weeks’ gestation. Factor
VIII levels usually rise in pregnancy, but factor IX
tends to remain constant. If an adequate rise in
Factor VIII is demonstrated, only a third trimester
sample may be necessary for subsequent
pregnancies, unless earlier interventions are
required.
r Factor levels should also be checked prior to
potentially hemorrhagic events such as invasive
diagnostic procedures, spontaneous abortion, or
termination of pregnancy. If levels are ⬍50 IU/dL,
women should receive prophylaxis.
r DDAVP can be used to raise factor VIII levels by
around three times, but recombinant clotting
factor concentrate is needed for factor IX deficient
women and may be required for those with factor
VIII levels below 15 IU/dL, as the response to
DDAVP may be insufficient. Pre- and
post-treatment levels should be checked and
therapeutic levels maintained for a suitable time
period depending on the procedure.

Intrapartum management (Table 15.3)
Although there are no large prospective studies that
correlate FVIII or IX levels with the risk of bleeding at
the time of childbirth, the opinion of experts is that levels should be above 50 IU/dL. If treatment is required,
the level should be brought to 100 U/dL pre-delivery
and maintained at ⬎50 U/dL for at least 3–5 days.
Excessive treatment should be avoided due to the risk
of thrombosis and thus careful titration and monitoring of levels is required.

187

Section 5. Hemorrhagic disorders

Table 15.3 Intrapartum management of hemophilia carriers
r
r

188

If FVIII/IX levels ⬍50 IU/dL at the last check, the
test needs to be repeated on arrival in labor
Recombinant factor concentrate is required to
raise factor IX levels. Treatment should be given
at the onset of established labor and pre and
post treatment levels should be obtained.
Allow spontaneous labor and normal vaginal
delivery, if no other obstetric concerns, to
minimize risk of intervention.
Avoid prolonged second stage of labor, with
early recourse to Cesarean section if necessary,
to reduce risk of trauma to the baby.
Avoid fetal blood sampling, fetal scalp
monitoring, Ventouse delivery, and mid-cavity
forceps, or forceps involving rotation of the
head.
Active management of the third stage of labor
and early suturing of episiotomy and
lacerations for patients with low factor levels.
Regional anesthesia has been shown to be safe
if the coagulation screen is normal and factor
levels are ⬎50 IU/dL treatment is required but
levels must be checked prior to removal of the
catheter as they may fall rapidly in the
postpartum period.
If maternal FVIII/IX levels ⬍ 50 IU/dL, caution
with non-steroidal anti-inflammatory drugs and
intramuscular injections.

Neonates are at risk of intracranial hemorrhage
and cephalhematomas during labor and delivery. The
risk is not increased by vaginal delivery but the second stage of labor should not be prolonged and early
recourse to Cesarean section may be required. Trauma
should be minimized by avoiding extra cephalic version, Ventouse delivery, fetal blood sampling, scalp
electrodes, and rotational forceps. The cut-off value
for predicted factor VIII or IX, above which no
birth restrictions are necessary, has not been defined,
although mild hemophilia is unlikely to be associated with severe bleeding at birth. Furthermore, the
increase in FVIII, induced by the stress of labor, provides some protection for babies with mild hemophilia
A. Female carriers have a small risk of extreme lyonization and low factor levels, but this needs to be weighed
up against the possibly greater risks of withholding
instrumental delivery and invasive fetal monitoring.

As Factor IX levels are lower at birth and are not
increased by the stress of delivery, female carriers of
severe Hemophilia B are theoretically at higher risk.

Analgesia
There is no consensus on the levels required for
regional anesthesia but this is generally considered to
be safe if FVIII /IX levels are ⬎ 50 IU/dL.3,4 Consideration should also be given to the levels at the time of
catheter removal and repeat treatment given beforehand if necessary. Intramuscular injections and nonsteroidal anti-inflammatory drugs are not contraindicated if factor levels are normal. All women with low
factor levels should have the opportunity to discuss
analgesia with an anesthetist prior to delivery.

Postpartum
Postpartum blood loss should be assessed as factor
levels may fall rapidly after delivery. Levels should be
maintained at ⬎50 IU/dL for at least 3 days, or for 5
days if Cesarean section has been performed.5
DDAVP and/or tranexamic acid may be useful to
prevent excessive postpartum bleeding. These agents
are described in Chapter 14.

Neonatal management
Affected babies may suffer bruising and bleeding at
venepuncture and heel prick sites and even spontaneous organ or joint bleeding. To identify neonates at
risk, a cord sample should be taken for coagulation
factor assay and the result must be known before the
patient leaves hospital. Female babies at risk of being
carriers for severe hemophilia may also require a cord
sample, as very low levels may occur due to severely
unbalanced lyonization. However factors VIII and IX
from the newborn do not reflect the true baseline level
and may need repeating at 6 months of age, when adult
values are reached.
Venepunctures and intramuscular injections,
including vitamin K should be avoided until the cord
factor level is known. Vitamin K could be given orally
if the results are delayed. Severely affected babies
should receive an ultrasound scan of the head to
assess for signs of intracranial hemorrhage (ICH),
particularly if delivery was traumatic or labor prolonged. This investigation is non-invasive, but lacks
sensitivity, particularly for subdural bleeds, which is
the commonest site of ICH in the neonate.

Chapter 15. Inherited coagulopathies

The mean age for occurrence of ICH is 4.5 days,2
when the baby is likely to be at home. Hence parents
and midwives should be informed of the early signs of
ICH; poor feeding, listlessness, vomiting, and seizures,
so that treatment can be administered without delay.
To date, there is no evidence for the benefit of prophylactic factor concentrate, about which the risk of
inhibitor development is debated. However, it may be
justified in selected cases, such as prematurity or traumatic delivery, where the risk of ICH is greater.

Factor XI deficiency

Table 15.4 Pre-pregnancy management of women with
Factor X1 deficiency
r

Assess clinical bleeding tendency and
coexistence of confounding factors such as
VWD or platelet dysfunction.
r Offer pre-natal diagnosis where there is a risk of
severe deficiency and the mutation is known.
r Discuss potential maternal bleeding risk and
options for management.
r Consent for use of blood products if necessary
and ensure hepatitis A and B immunity.

Introduction

Clinical features

Factor XI is an important component of the intrinsic
coagulation pathway, playing a key role in the amplification of initial thrombin production, via activation of factor IX. The additional amount of thrombin
activates thrombin-activatable fibrinolysis inhibitor
(TAFI), which consolidates the fibrin clot and protects it from degradation by fibrinolysis. Thus deficiency of Factor XI is manifest mostly by injury or
surgery-related bleeding at sites which are prone to
local fibrinolysis, such as the nose and genitourinary tract. Women with factor XI deficiency are
at risk of menorrhagia and bleeding in relation to
childbirth.

The bleeding tendency in FXI-deficient individuals is
highly variable.8 Factor XI activities ⬍ 15 U/dL have
been designated as severe deficiency, although bleeding is not closely correlated with factor levels9 as it is
with hemophilia A and B. Neither does the abnormal
genotype causing the condition seem to bear any relationship to bleeding tendency, which is inconsistent
amongst family members. Indeed, in most patients
spontaneous bleeding, as well as bleeding after hemostatic challenge, does not occur and phenotype may
depend on other associated factors, such as coexistence
of mild von Willebrand disease.

Hormonal influences on factor levels
Disease incidence
The inheritance of Factor XI deficiency is autosomal. It
is most common amongst Ashkenazi Jews, where the
estimated heterozygosity rate is as high as 8%.6 The
incidence in the non-Jewish population is reported to
be around 1:100 000, although this is likely to be an
underestimate, as it may frequently remain undetected
as routine coagulation assays may be normal in heterozygotes and there may be no bleeding history.
The predominant mutations in Ashkenazi Jews are
a Glu117stop codon in exon 5 designated type II, and
a Phe283Leu mutation in exon 9 designated type III.
Homozygotes for type II and type III mutation have
factor XI activities ⬍ 1 U/dL and 8–15 U/dL, respectively, with compound heterozygotes for type II and
III having factor XI levels between these values.7 In
non-Jewish populations, rapidly increasing numbers
of mutations and polymorphisms have been reported,
now reaching over 80. For the majority of these, the
level of FXI antigen has not been reported.

Factor XI levels usually remain constant in pregnancy,
but studies have shown inconsistencies in levels with
increases or decreases as pregnancy advances.10

Obstetric complications
The main risk in pregnancy is of uterine hemorrhage
during invasive procedures, miscarriage, or postpartum. Patients with FXI levels ⬍15 IU dL-1 have a 16%–
30% risk of peripartum bleeding11 and this has been
confirmed to almost exclusively affect those with a predetermined bleeding phenotype.10 Thus it is important
to attempt to ascertain, by thorough history taking,
which patients are at risk of bleeding, so that ante-natal
procedures, childbirth, and the postpartum period can
be managed appropriately (Table 15.4).

Ante-natal management
As it is often not feasible to check levels in an acute
situation, routine monitoring should be carried out

189

Section 5. Hemorrhagic disorders

Table 15.5 Ante-natal management of women with Factor X1
deficiency

Table 15.6 Intrapartum management of women with Factor
X1 deficiency

Check levels at booking, 28 and 34 weeks’
gestation and prior to invasive procedures.
r Patients with severely low levels or a positive
bleeding history should be given prophylaxis to
cover invasive procedures.
r Other patients can be managed expectantly
with close observation and treatment available
on standby should bleeding occur.

at booking, 28, and 34 weeks. Whilst low factor levels cause prolongation of the APTT, test reagents vary
in their sensitivity to factor XI levels and a normal
APTT does not exclude mild deficiency. This is particularly so in pregnancy where the increase in factor
VIII may normalize the APTT even when factor XI
is reduced. Thus a specific coagulation factor assay is
required (Table 15.5).

Treatment options to cover delivery
and ante-natal procedures

190

Most patients can be managed expectantly,12 but those
with severely reduced levels or a positive bleeding history require prophylaxis for invasive ante-natal procedures, miscarriage, and delivery.
Factor XI concentrate provides effective cover and
has a mean half-life of 52 hours, so a single dose is usually sufficient. However, it is associated with a potential risk of transfusion transmitted infections, common to all plasma products, as well as an increased
risk of thrombosis due to coagulation activation.13
The increased thrombotic risk may be further exaggerated in pregnancy where there is already activation of
coagulation and increased thrombin generation.
Fresh frozen plasma contains variable amounts
of factor XI and patients with severe deficiency are
unlikely to achieve levels above 30 IU dL−1 . However,
it is helpful for milder cases and involves less donor
exposure than factor XI concentrate. A dose of 15–
20 mL/kg is effective, but the risk of fluid overload
must be considered.
Monitoring of the response to FFP or factor XI concentrate is important and, due to the thrombogenicity of the latter, levels should not be allowed to exceed
70 IU dL−1 . A recent study found inhibitor development, after transfusion of plasma derived factor XI, in

An on-demand policy can be advocated,
including for those with severely low levels,
during and after vaginal delivery.
r Most patients undergoing Cesarean Section
can be managed expectantly, but those with
severe deficiency should be given prophylaxis.
r Measures should be taken to avoid unnecessary
trauma to the baby during delivery.
r The third stage of labor should be actively
managed.
33% of patients with severe factor XI deficiency due
to homozygous type II mutation (which accounts for
approximately 25% of Jewish patients with severe factor XI deficiency).
R
,
Recombinant factor VIIa (rFVIIa, NovoSeven
Novo Nordisk Ltd, Bagsvaerd, Denmark) is currently
being assessed as a possible alternative to plasmaderived FXI replacement and avoids the risk of bacterial or viral infections, transfusion-related lung injury
and development of inhibitors to factor XI. It is as
yet unlicensed for use in this setting and the optimal dose has not been ascertained. A suggested dose
for minor procedures is 90 ␮g/kg administered intravenously before surgery and 4 h later. For major
surgery, 2 hourly infusions are necessary due to the
short half-life of the product.

Intrapartum management
Around 70% of patients do not experience bleeding
problems at delivery. This may be due to increased levels of coagulation factors, including factor VIII and
fibrinogen, at term. Also, the pregnancy-associated
reduction in fibrinolytic activity, due to decreased levels of tissue plasminogen activator and urokinase and
an increased level of plasminogen activator inhibitor2, contributes to hemostasis. Thus, even for those with
severe factor XI deficiency an on-demand policy can
usually be adopted for vaginal delivery.14 However, it
is important that the patient is closely observed and
that all relevant staff is aware of the management plan.
It may be that a similar policy can be adopted for
patients with severe deficiency undergoing Cesarean
section, but until further studies are done, these
patients should probably receive prophylaxis with one
of the agents described above (Table 15.6).

Chapter 15. Inherited coagulopathies

Regional anesthesia

Hormonal influences on factor levels

Epidural anesthesia should be avoided in patients with
low factor XI levels. If the procedure is necessary, it
should be covered with factor XI concentrate and an
adequate response demonstrated. FFP is not recommended due to the variable levels of FXI. Recombinant factor VIIa may provide effective cover, but further evaluation is required in this area.

Factors II, V, and XIII tend to remain constant
throughout pregnancy or show a slight increase but
there is a progressive rise in factors VII, X, and fibrinogen, particularly in the third trimester. This is beneficial to heterozygous women with mild or moderate factor deficiency, but in homozygous women, with
severe deficiency, levels remain low.

Postpartum management

Pre-pregnancy management

The incidence of primary and secondary postpartum
hemorrhage, in patients with untreated factor XI deficiency, has been reported to be 16% and 24%, respectively. Tranexamic acid is effective, although its use with
factor XI concentrate should be avoided. The standard
dose is 1g 6–8 hourly for 3–5 days, with the first dose
being administered in labor.

The clinical bleeding tendency and response to hemostatic challenges should be ascertained. Women should
be counseled about their potential bleeding risk in
relation to pregnancy, ante-natal procedures, delivery,
and the postpartum period. Consent for use of blood
products should be obtained and immunity to hepatitis A and B ensured. Genetic counseling should be
given and pre-natal diagnosis offered where possible.

Neonatal management

Ante-natal management

Neonatal hemorrhage due to peripartum events is
rare but nevertheless care should be taken during
delivery to avoid unnecessary trauma to the baby,
including avoidance of Ventouse extraction, rotational
forceps, and invasive monitoring techniques. Spontaneous bleeding or intracranial hemorrhage has not yet
been reported in neonates, but a cord blood sample
should be taken to determine the potential for bleeding during high risk procedures such as circumcision.
Neonatal levels are approximately half that of adults
and repeat testing after 6 months of age is required to
provide an accurate baseline level.

Rare coagulation factor deficiences
The rare coagulation disorders include deficiencies of
coagulation factors II, V, VII, X, V, and VIII, combined
vitamin K-dependent factors, FXIII, and disorders of
fibrinogen. The spectrum of bleeding manifestations
in individuals with these disorders is variable, but
some may present with severe bleeds, including
intracranial hemorrhage and hemarthroses. With the
exception of dysfibrinogenemia, these disorders have
autosomal recessive inheritance and their prevalence,
in the severe form, varies between 1:500 000 and
1:2,000 000. Pregnancy in women with these disorders
or couples at risk of having an affected child, should
be managed in an obstetric unit with close links to a
Hemophilia Center.

Levels should be checked at booking and repeated at
28 and 34 weeks’ gestation. Depending on the factor level and clinical bleeding tendency, prophylaxis
may be required for ante-natal procedures and delivery. There is little evidence to guide therapeutic decisions but in general, relatively low levels of factors
II, V, VII, and X, of around 20 IU/dL, are sufficient
for normal hemostasis.12,15 Therefore, patients with
partial deficiencies and no history of bleeding can be
managed expectantly. Otherwise, replacement therapy
should start at the onset of established labor and the
factor half-life should be considered to determine the
need for, and timing of, repeat doses. Pre- and posttreatment factor levels should be obtained and effective
levels maintained for 3–5 days after delivery.

Treatment options
Prothrombin complex concentrates can be used for
patients with factors II or X deficiency. These are
pooled plasma-derived products containing known
quantities of factors II, IX, and X, with or without factor VII. The strength of the concentrate is expressed in
terms of units of FIX, but this is approximately equal to
the units of prothrombin. Concomitant use of tranexamic acid should be avoided because of the risk of
thrombosis.
FFP is the only available product for FV deficiency
and may also be used for patients with prothrombin

191

Section 5. Hemorrhagic disorders

and FX deficiency. A virally inactivated product should
be used. An initial dose of 15 mL/kg should be given,
with repeat doses dictated by factor levels and clinical response. Women with factor V deficiency failing
to respond to FFP may benefit from platelet transfusions, which provide a concentrated supply of platelet
factor V.
Recombinant FVIIa is the treatment of choice for
surgery or childbirth in women with FVII deficiency,
at a dose of 20–25 mcg/kg administered every 4–6
hours.
Tranexamic acid is useful in preventing postpartum bleeding, although should not be used in conjunction with prothrombin complex concentrates.
Patients with combined vitamin K-dependent factors can be treated with daily vitamin K, although FFP
may be needed in the event of bleeding.

Early pregnancy failure
Maternal FXIII plays a critical role in uterine hemostasis and maintenance of the placenta during gestation.
The risk of miscarriage in women with severe factor
XIII deficiency is around 50%, depending on the subtype. These women should receive prophylactic infusions of FXIII at monthly intervals, aiming for a trough
level of ⬎3 U/dL, although higher Factor XIII levels
may be needed for delivery. 16
Fibrinogen is important for implantation and
patients with afibrinogenemia or hypofibrinogenemia
have a high rate of early miscarriage occurring at 6–8
weeks’ gestation. Regular infusions of fibrinogen concentrate, to maintain trough levels ⬎0.6 g/L, should
be started as soon as pregnancy is confirmed and
continued throughout pregnancy and the peripartum
period.17 Fibrinogen consumption tends to increase as

192

pregnancy advances. Repeated ultrasounds should be
carried out to detect concealed placental bleeding and
monitor fetal growth.
Dysfibrinogenemia has been associated with a high
incidence of miscarriage and stillbirth18 but clinical
phenotypes vary and management should be individualized, depending on the fibrinogen level and
the clinical presentation of the disorder in the family. Thromboprophylaxis with low molecular weight
heparin is required for those with personal or family history of thrombosis and fibrinogen replacement
for bleeding phenotypes, but many cases are asymptomatic without the need for specific treatment.

Thrombosis
The potential for thrombosis following factor replacement must be considered and attention given to simple thromboprophylactic measures such as adequate
hydration, compression stockings, and early mobilization. Patients with afibrinogenemia- or dysfibrinogenemia are at particular risk of thrombosis due to
impaired regulation of thrombin generation. Loose
platelet thrombi form and are susceptible to embolization, therefore careful consideration should be given
to the balance of bleeding and thrombotic risk. For
these patients, a continual infusion of fibrinogen concentrate to maintain levels above 1.5 g/L during the
peripartum period allows for fine control.

Neonatal management
Perinatal trauma such as Ventouse delivery, rotational
forceps, and fetal blood sampling should be avoided.
Severe and moderate deficiencies can be diagnosed on
a cord blood sample. Severely affected babies require
cranial ultrasound to detect any ICH.

Chapter 15. Inherited coagulopathies

References
1. Chi C, Lee CA, Shitagh N et al. Pregnancy in carriers
of haemophilia. Haemophilia 2008; 14: 56–64.
2. Kulkarni, Roshni and Lusher, Jeanne M. Intracranial
and extracranial hemorrhages in newborns with
hemophilia: a review of the literature. Journal of
Pediatric Hematology/Oncology 1999; 21: 289–
295.
3. Kadir RA, Economides DL, Braithwaite J et al. The
obstetric experience of carriers of haemophilia. British
Journal of Obstetrcs and Gynaecology 1997; 104:
803–810.
4. EA., Letsky. Haemostasis and epidural anaesthesia.
International Journal of Obstetric Anesthesia. 1991; 1:
51–54.
5. Lee CA, Chi C, Pavord SR et al. UK haemophilia
Centre Doctors’ Organization. The obstetric and
gynaecological management of women with inherited
bleeding disorders – review with guidelines produced
by a taskforce of UK Haemophilia Centre Doctors’
organization. Haemophilia 2006; 12: 301–336.
6. Asakai R, Chung DW, Davie EW, Seligsohn U. Factor
XI deficiency in Ashkenazi Jews in Israel. New England
Journal of Medicine 1991; 325: 153–158.

10. Myers B, Pavord S, Kean L et al. Pregnancy outcome in
Factor XI deficiency: incidence of miscarriage,
antenatal and postnatal haemorrhage in 33 women
with Factor XI deficiency. British Journal of Obstetrics
and Gynaecology: an International Journal of Obstetrics
and Gynaecology 2007; 114: 643–646.
11. Kadir RA, Lee CA, Sabin CA, Pollard D, Economides
DL. Pregnancy in women with von Willebrand’s
disease or factor XI deficiency. British Journal of
Obstetrics and Gynaecology 1998; 105: 314–321.
12. Bolton-Maggs PH, Perry DJ, Chalmers EA et al. The
rare coagulation disorders–review with guidelines for
management from the United Kingdom Haemophilia
Centre Doctors’ Organisation. Haemophilia 2004; 10:
593–628.
13. Bolton-Maggs PH, Colvin BT, Satchi BT et al.
Thrombogenic potential of factor XI concentrate.
Lancet 1994; 344: 748–749.
14. Salomon O, Steinberg DM, Tamarin I et al. Plasma
replacement therapy during labor is not mandatory for
women with severe factor XI deficiency. Blood
Coagulation and Fibrinolysis 2005; 16: 37–41.
15. Kadir R, Chi C, Bolton-Maggs P. Pregnancy and rare
bleeding disorders. Haemophilia 2009; 15: 990–1005.

7. Hancock JF, Wieland K, Pugh RE et al. A molecular
genetic study of factor XI deficiency. Blood 1991; 77:
1942–1948.

16. Asahina T, Kobayashi T, Takeuchi K et al. Blood
coagulation factor XIII deficiency and successful
deliveries: a review of the literature. Obstetrical and
Gynecological Survey 2007; 62: 255–260.

8. Bolton-Maggs PH, Patterson DA, Wensley RT,
Tuddenham EG. Definition of the bleeding tendency in
factor XI-deficient kindred – a clinical and laboratory
study. Thrombosis and Haemostasis 1995; 73: 194–
202.

17. Kobayashi T, Kanayama N, Tokunaga N et al. Prenatal
and peripartum management of congenital
afibrinogenaemia. British Journal of Haematology
2000; 109: 364–366.

9. Bolton-Maggs PH, Wan-Yin BY, McCraw AH et al.
Inheritance and bleeding in factor XI deficiency.
British Journal of Haematology 2008; 69: 521–528.

18. Haverkate F, Samama M. Familial dysfibrinogenemia
and thrombophilia. Report on a study of the SSC
Subcommittee on Fibrinogen. Thrombosis and
Haemostasis 1995; 73: 151–161.

193

Section 5
Chapter

Hemorrhagic disorders

Genetic counseling and pre-natal diagnosis
in hemophilia
Andrew Mumford

Introduction
Hemophilia is the most common severe genetic bleeding disorder and presents significant risk to the fetus
at delivery. Hemophilia is also associated with significant morbidity later in life and may require intensive long-term treatment, which may be a considerable
burden to affected families. High quality obstetric care
of women with a family history of hemophilia is therefore paramount but presents particular management
challenges.

Genetic counseling in hemophilia
Genetic counseling refers to the process of communication of information to women and families to enable
informed decision making about the consequences of
carrying a fetus with hemophilia. Genetic counseling
for hemophilia should encompass the issues of carrier
testing and pre-natal diagnosis.
Successful genetic counseling should be supportive and requires careful two-way discussion between
families and healthcare professionals who are familiar with hemophilia management and with the techniques available for carrier testing and pre-natal diagnosis. Since genetic counseling often raises complex
ethical and moral issues, this process may require multiple face-to-face consultations supported by clear and
objective written information. Ideally, genetic counseling should be initiated before pregnancy is planned.1
Genetic counseling is a step-wise process and may
require discussion about the following issues:
r family diagnosis of hemophilia and clinical
severity;
r inheritance pattern of hemophilia within the
family to exclude carriership or to identify
“possible” and “obligate” carriers;

194

r pattern of transmission and consequences of
hemophilia in future offspring;
r benefits and hazards of carrier detection
techniques;
r Options available for management of pregnancy,
including pre-natal diagnosis.

Heritability of hemophilia
As hemophilia A and B are sex-linked disorders,
affected families may contain males with hemophilia
and female hemophilia carriers who usually do not
have abnormal bleeding, but may transmit hemophilia
to males in the next generation (Fig. 16.1). Analysis
of an accurate family pedigree is essential to establish
the probability of hemophilia carriership and transmission risk.
r Sons of female hemophilia carriers have a 50%
chance of having hemophilia.
r Daughters of female hemophilia carriers have a
50% chance of being carriers.
r Sons of males with hemophilia will not inherit
hemophilia.
r Daughters of males with hemophilia will always
inherit hemophilia and will therefore be obligate
hemophilia carriers.
r Approximately 50% of individuals newly
diagnosed with hemophilia have no family history
of hemophilia.
Some women can be excluded as being hemophilia
carriers by analysis of the family pedigree. Although
these women do not have the gene change responsible
for hemophilia elsewhere in their families, they retain
a small risk, as in the general population, of being carriers of a different hemophilia gene change that has
arisen by spontaneous mutation. For hemophilia A,

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

Chapter 16. Genetic counseling and pre-natal diagnosis in hemophilia

I
1

II
2

Obligate hemophilia
carrier

II
1

III
1

Fig. 16.1 Pedigree of a family with hemophilia showing possible
patterns of transmission of hemophilia over three generations. The
offspring of a female hemophilia carrier (I.2) can include males with
hemophilia (II.1), males without hemophilia (II.2), female hemophilia
carriers (II.3), and females who are not hemophilia carriers (II.4). The
offspring of males with hemophilia (II.1) can either be males without
hemophilia (III.1) or female obligate hemophilia carriers (III.2).

50% chance of being
hemophilia carrier

III
2

12.5% chance of being
male with hemophilia

IV
1

this background risk of carriership is approximately 1
in 20 000 women.

Prediction of carrier status by
pedigree analysis
For women from families with hemophilia, the probability that a pregnancy will yield a fetus that is a male
with hemophilia can be calculated from the family
pedigree using simple rules of Mendelian inheritance
(Fig. 16.2).
For families in which there is hemophilia in one
individual but no antecedent history of hemophilia
(sporadic hemophilia), calculating the risk of
hemophilia in subsequent members of the same
generation is more difficult (Fig. 16.3). Sporadic
hemophilia usually arises because of new mutations
in the F8 or F9 genes (hemophilia A and B, respectively) occurring during gametogenesis in either the
mother or a maternal ancestor of an affected male.
However, spontaneous mutations occur more readily
during spermatogenesis than oogenesis. Therefore, the
causative mutation in a male with sporadic hemophilia
is more likely to have arisen during spermatogenesis
in the maternal grandfather than in oogenesis in the
mother. It follows that mothers of males with sporadic
hemophilia are likely to be constitutional hemophilia
carriers. Observational population studies confirm
this prediction and show that approximately 90% of
mothers of males with sporadic hemophilia are carriers and therefore, have significant risk transmitting
hemophilia to future male offspring.2

Fig. 16.2 Example pedigree allowing calculation of the probability
of carriership and transmission of hemophilia. The female proband
(III.2-arrowed) has a maternal grandfather (I.1) with hemophilia.
Since I.1 is a male with hemophilia, the mother of the proband (II.2)
is an obligate carrier of hemophilia. The proband III.2 therefore has a
50% chance of hemophilia carriership. Since the probability that a
hemophilia carrier will carry a fetus that is a male with hemophilia at
each pregnancy is 25%, the absolute probability that the unborn
fetus IV.1 will be affected with hemophilia is 12.5%.

I
2

90% chance of being
hemophilia carrier

II
1

23% chance of being
male with hemophilia

III
1

Fig. 16.3 Example pedigree showing estimated probability of
hemophilia carriership and transmission in a family with sporadic
hemophilia The female proband (II.2-arrowed) already has a son
who is affected with hemophilia (III.3) but has no other family
history. The mutation responsible for hemophilia in III.1 is most likely
to have occurred during spermatogenesis in individual I.3 and so
the proband II.2 is likely to be a carrier of hemophilia. The estimated
probability that II.2 is a carrier of hemophilia is approximately 90%
and so the probability that each subsequent pregnancy will yield a
male with hemophilia is approximately 23%.

195

Section 5. Hemorrhagic disorders

Laboratory detection of hemophilia
carriership
Determining the probability of hemophilia carriership
by pedigree analysis is essential for the genetic counseling process. However, all women who are potential
hemophilia carriers should also be offered laboratory
carriership detection. Two complementary approaches
are available; coagulation factor activity assays and
mutation analysis.

Coagulation factor activity assays
Female carriers of hemophilia usually show reduced
activities of coagulation factor VIII or IX (hemophilia
A and B, respectively) to levels of 40%–80% that of
unaffected individuals.3 However, there is wide variation in factor activity between carriers and there is significant overlap with women who are not hemophilia
carriers. Measurement of coagulation factor activity
may therefore guide identification of hemophilia carriers but is insufficient for definitive diagnosis. The ratio
of FVIII to Von Willebrand factor (VWF) may be helpful, as these two molecules normally circulate in the
plasma with 1:1 stoichiometry. Thus carrier status may
be suspected if the ratio falls below 0.7, despite the
absolute FVIII level being normal.4

Genetic detection of carriership

196

Female hemophilia carriers are heterozygous for
mutations in F8 or F9 and demonstration of a
hemophilia – associated mutation in these genes is sufficient to diagnose carriership. It is good practice to
confirm hemophilia carriership with genetic testing
even in women identified as obligate carriers by pedigree analysis. Definitive exclusion of hemophilia carriership in potential carriers requires demonstration
that the hemophilia mutation in the family is absent.
In this circumstance, prior knowledge of the causative
mutation in a male with hemophilia or an obligate
female carrier from the family is essential.
Testing the potential for transmission of
hemophilia in asymptomatic women raises complex moral issues for the individual and families
undergoing testing. The full implications of genetic
testing should therefore be discussed during counseling and informed written consent is mandatory.
Counseling should include specific discussion about
the limitations of F8 and F9 genetic analysis.

Mutations associated with hemophilia
Although more than 1800 F8 mutations have now been
identified in individuals with hemophilia A, many
defects are recurrent and have been recognized in multiple affected families. A major structural rearrangement of the F8 gene resulting from an inversion involving intron 22 accounts for approximately 50% of cases
of severe hemophilia A. Other recurrent mutations
associated with severe hemophilia A include point
mutations, non-sense mutations, deletions, or other
major structural changes in F8 that prevent expression
of the gene. Mild hemophilia A and hemophilia B are
usually associated with point mutations in F8 and F9,
respectively, although heterogeneity between affected
families means that previously unreported mutations
are common.
The mutation databases for hemophilia A (http://
europium.csc.mrc.ac.uk/WebPages/Main/main.htm)
and
hemophilia
B
(http://www.kcl.ac.uk/ip/
petergreen/haemBdatabase.html) contain bibliographic references and phenotypic data from previously reported families with hemophilia. These
resources are valuable for confirming that a newly
identified mutation in a hemophilia family is causative
and in predicting the future clinical phenotype of
affected males.

Limitations and hazards of genetic diagnosis
of carriership
1. Failure to detect causative mutations.
Approximately 5% of hemophilia mutations are
not detected by analysis of the coding sequence of
F8 or F9. Similarly, some mutations such as large
deletions may be readily detected in males but not
in heterozygous female carriers. In these
circumstances, detection of carriers currently
requires techniques such as linkage analysis. This
may not be informative in all families and,
because of genetic recombination events, has
lower diagnostic accuracy than direct mutation
detection by sequencing.
2. False-negative carrier detection because of
somatic mosaicism. An individual is a somatic
mosaic for hemophilia when a spontaneous
hemophilia mutation occurs in a somatic cell
during early embryogenesis rather than during
gametogenesis in one or other parent. In an
affected embryo, the hemophilia mutation is

Chapter 16. Genetic counseling and pre-natal diagnosis in hemophilia

therefore present in some, but not all, cells
including the germ cells. Those germ cells which
contain the hemophilia mutation may then go on
to form gametes. For women who are somatic
mosaics, this population of gametes is then
capable of transmitting hemophilia to the
subsequent generation.
Somatic mosaicism was identified in a female
proband in more than 10% of families with severe
hemophilia and, in some cases, the hemophilia mutation was present in up to 25% of maternal cells.5 In
this circumstance, standard genetic testing of DNA
obtained from peripheral blood cells may not detect
a hemophilia mutation and somatic mosaic mothers
may therefore be mis-classified as “not hemophilia carriers.” Somatic mosaicism should be considered in all
women who have a son with hemophilia but no other
family history and who have been classified as “not a
hemophilia carrier” by standard genetic testing. For
families with sporadic hemophilia B, one study has
estimated that women with this background have a
risk of hemophilia B in a second fetus of ⬍6%.6 This
very low probability may be similar for all forms of
hemophilia and should be discussed during genetic
counseling for all potential carrier women.

Pre-natal diagnosis of hemophilia
Women who have been identified as hemophilia
carriers by pedigree analysis and laboratory investigation may be offered several different options for
pre-natal diagnosis. Involvement of healthcare professionals with expertise in fetal medicine is essential.
Pre-natal diagnosis in hemophilia may be performed currently for two different reasons:
r to offer a more accurate probability of whether a
fetus will be affected with hemophilia by fetal
sexing to assist management of delivery;
r to offer early definitive diagnosis of a hemophilia
in male fetus by first trimester pre-natal genetic
diagnosis to enable the option of termination of
an affected pregnancy.

Pre-natal fetal sexing

r Fetal sexing may be performed by non-invasive
(ultrasound or free fetal DNA (ffDNA) analysis)
or invasive (chorionic villus sampling (CVS) or
amniocentesis) techniques.

r If a female fetus is identified, hemophilia is
excluded and hemostatic precautions at delivery
are unnecessary.
r Fetal ultrasound allows gender to be identified
reliably in most pregnancies from about 18–20
weeks’ gestation.
r PCR detection of the fetal SRY locus in ffDNA
circulating in maternal plasma is highly specific
for a male fetus. This assay requires a maternal
venous blood specimen of ⬍20 mL and has ⬎99%
diagnostic accuracy at 10–12 weeks’ gestation in
expert centers.7 A recent study evaluating the
method in 196 women, including hemophilia
carriers, showed 100% accuracy as early as 7
weeks’ gestation.8

First trimester pre-natal genetic diagnosis

r First trimester pre-natal genetic diagnosis allows
definitive diagnosis of hemophilia in a fetus but
requires invasive testing by CVS.
r CVS is an important option for confirmed
hemophilia carriers who are considering
termination of a male fetus with hemophilia. The
uptake of this approach is very low in most
reported series of pregnancies in hemophilia
carriers.
r CVS for pre-natal genetic diagnosis is performed
at 11–14 weeks’ gestation and carries a
miscarriage rate of approximately 1%. Earlier
procedures have resulted in fetal limb reduction
defects, particularly if performed before 10 weeks’
gestation. Hemophilia carrier mothers with low
coagulation factor levels may need treatment to
cover the procedure. (See Chapter 15.)
r Placental cells obtained by CVS are first used to
determine fetal sex. Detection of the hemophilia
mutation present in the family is then performed
on male fetuses.
r Advances in very early fetal sexing by ffDNA
analysis may enable female fetuses to be identified
before 11–14 weeks so that CVS is then
unnecessary. Confirmation of gender by fetal
ultrasound at 18 weeks’ gestation is then
recommended.
r Amniocentesis is an alternative technique for
pre-natal genetic diagnosis in hemophilia and can
safely be performed from 15 weeks’ gestation. This
technique is therefore less suitable for women
contemplating termination of pregnancy.

197

Section 5. Hemorrhagic disorders

Amniocentesis is associated with miscarriage rates
of 0.5%–1%, but higher miscarriage rates and fetal
talipes have been associated with amniocentesis
performed before 15 weeks. Cord blood sampling
is unsuitable for pre-natal diagnosis of hemophilia
because of bleeding risk in an affected
fetus.

Future techniques for pre-natal diagnosis
Third trimester amniocentesis and mutation detection
Amniocentesis performed at around 36 weeks enables
genetic diagnosis of hemophilia in male fetuses and
carries a risk of preterm labor of approximately 1%
in experienced centers. This approach allows hemostatic precautions to be applied only to male fetuses
with hemophilia, so that, unaffected male fetuses can
be delivered without these constraints. The risk to the
fetus of delivery precipitated by the amniocentesis is
very small at this late gestation. The potential clinical
benefits of this approach in hemophilia are currently
under evaluation.

Pre-implantation diagnosis
Pre-implantation sexing with re-implantation of
female or unaffected male embryos requires standard
in vitro fertilization techniques, with harvesting of
cells from embryos at the 8-cell stage for analysis.
Single-cell PCR enables detection of specific mutations
in male embryos.9 These approaches are technically
feasible in hemophilia and have now been performed
in small numbers of successful pregnancies.

Mutation detection using ffDNA or fetal cells
in maternal blood
Detection of hemophilia mutations in ffDNA or in fetal
cells in maternal blood in pregnancy potentially offers
non-invasive pre-natal diagnosis of hemophilia. This
approach requires highly efficient purification of fetal

198

material from maternal blood and may only be feasible in the third trimester when ffDNA and fetal cells
are most abundant. This approach is currently at early
development stage.

Genetic counseling for other heritable
bleeding disorders
Genetic counseling, carrier detection and pre-natal
diagnosis should also be considered in families with
heritable bleeding disorders other than hemophilia,
which may also present bleeding risk to an affected
fetus. Most of these rare bleeding disorders show autosomal recessive inheritance (e.g. severe Factor X deficiency, severe Factor V deficiency, Type III VWD) and
genetic counseling requires discussion about the transmission of homozygous or compound heterozygous
mutations from both parents. Affected fetuses are usually sporadic and arise in families with no bleeding history in heterozygous “carrier” ancestors. For mothers
who are known heterozygous “carriers” or who themselves are homozygous or compound heterozygous for
a recessive bleeding disorder, accurate prediction of
fetal bleeding risk may require partner testing. This
is particularly important in consanguineous partnerships where the risk of transmission of homozygous
recessive mutations is high.
Genetic counseling for the rare bleeding disorders
should reflect that the relationship between plasma
coagulation factor activity and bleeding risk in affected
individuals is less predictable than in hemophilia and
that some disorders show variable penetrance. Since
the range of reported mutations in the rare bleeding disorders is less than for hemophilia, detection of
previously undescribed mutations in affected families
is common. Uncertainty about whether a candidate
mutation is the true disease-associated mutation may
hamper genetic carrier detection and pre-natal diagnosis in some families.

Chapter 16. Genetic counseling and pre-natal diagnosis in hemophilia

References
1.

Ludlam CA, Pasi KJ, Bolton-Maggs P et al. A
framework for genetic service provision for
haemophilia and other inherited bleeding disorders.
Haemophilia 2005; 11: 145–163.

Kasper CK, Lin JC. Prevalence of sporadic and familial
haemophilia. Haemophilia 2007; 13: 90–92.

Plug I, Mauser-Bunschoten EP, Broker-Vriends AH
et al. Bleeding in carriers of haemophilia. Blood 2006;
108: 52–56.

Shetty S, Ghosh K, Pathare A, Mohanty D. Carrier
detection in haemophilia A families: comparison of
conventional coagulation parameters with DNA
polymorphism analysis – first report from India.
Haemophilia 2001; 5: 243–246.
Leuger M, Oldenburg J Lavergne J-M et al. Somatic
mosaicism in Haemophilia A: a fairly common event.
American Journal of Human Genetics 2001; 69: 75–87.

Green PM, Saad S, Lewis CM, Gianelli F. Mutation
rates in humans I: overall and sex-specific rates
obtained from a population study of haemophilia B.
American Journal of Human Genetics 1999; 65:
1572–1579.

Avent N, Chitty LS. Non-invasive diagnosis of fetal
sex; utilisation of free fetal DNA in maternal plasma
and ultrasound. Prenatal Diagnosis 2006; 26: 598–
603.

Bustamente-Aragones A, Rodrguez de Alba M,
Gonzalez-Gonzalez C et al. Foetal sex determination
in maternal blood from the seventh week of gestation
and its role in diagnosing haemophilia in the foetuses
of female carriers. Haemophilia 2008; 14: 593–598.

Michaelidis K, Tuddenham EG, Turner C et al. Live
birth following the first mutation specific
pre-implantation genetic diagnosis for haemophilia A.
Thrombosis and Haemostasis 2006; 95: 373–379.

199

Section

Microangiopathies

Section 6
Chapter

Microangiopathies

Pre-eclampsia
Eleftheria Lefkou and Beverley Hunt

Introduction
Pre-eclampsia (PET) is a pregnancy-related multisystem syndrome, that is characterized by newonset of hypertension (blood pressure greater than
140/90 mmHg) after 20 weeks of gestation and proteinuria (greater than 1+, or urinary excretion of protein ≥ 300 mg/24 hours) resolving after delivery. PET
is also termed toxemia, pregnancy-induced hypertension, and pre-eclamptic toxemia. Symptoms can occur
any time after 20 weeks of gestation or even start in the
first few days after delivery, and always resolve within
a few days to weeks after delivery of the placenta. Early
onset PET is when it develops before the 34th week
of gestation and late onset PET when it presents after
the 34th week of pregnancy. Predisposing factors are
shown in Table 17.1. It is not known why some women
develop pre-eclampsia, while others with the same risk
factors do not.
Eclampsia occurs when PET is complicated by
seizures.
Chronic hypertension is defined as systolic pressure ≥140 mmHg and/or diastolic pressure ≥90
mmHg that antedates pregnancy, that is present before
the 20th week of pregnancy, or persists longer than 12
weeks’ postpartum.
PET with chronic hypertension is diagnosed when
a pregnant woman has a history of chronic hypertension and then develops features suggestive of PET after
the 20th week of pregnancy.
Gestational hypertension, or transient hypertension of pregnancy refers to the situation that is characterized by elevated blood pressure (⬎140/90 mmHg)
after the 20th week of gestation, but without proteinuria, that occurs uniquely during pregnancy and
resolves after birth.

The aim of this chapter is to provide a basic understanding of PET and a detailed understanding of
hematological complications and their management.

Epidemiology
Pre-eclampsia (PET), the commonest medical complication of pregnancy, affecting approximately 2%–14%
of all pregnancies, remains a major cause of maternal
and fetal morbidity and mortality worldwide. It is estimated that 50 000 women die annually worldwide due
to PET and eclampsia. In the United States the incidence of PET is approximately 5%–8%, with 75% of
cases being mild and 10% of cases due to early onset
PET. According to the latest report from the Confidential Enquiry into Maternal and Child Health it remains
the second major cause of maternal mortality and morbidity in the UK after venous thromboembolism.1 The
incidence of PET in the UK is reported as 2%–8%, with
a fatality rate of 18/ 100 000 pregnancies. Mild PET is
under-reported and so the true incidence is potentially
much higher.
PET is associated with intrauterine growth restriction (IUGR), in one-third of cases. Premature delivery to prevent the progression of PET is responsible for
15% of all preterm births. Infants of women with PET
have a fivefold increase in mortality compared with
infants of mothers without the disorder.
The recurrence likelihood for PET is reported as
60% if it had occurred ⬍34 weeks’ gestation and 10%–
20% if occurred near term. The key to appropriate
management is early clinical recognition.

Diagnosis of PET
The diagnosis of PET is based on the maternal history,
signs, and symptoms (Table 17.2). The current aim of

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

203

Section 6. Microangiopathies

Table 17.1 Risk factors for pre-eclampsia (PET)

Table 17.2 Signs and symptoms of severe pre-eclampsia

Nulliparity

Blood pressure greater than 160/110 mm Hg

High body mass index (BMI) (⬎35 at booking)

Chronic hypertension

Impaired kidney function (serum creatinine concentration
⬎110 ␮mol/L, urine protein greater than 5 grams in a 24-hour
urine collection) or low urine production (less than 500 mL in
24 hours)

Diastolic pressure ⬎89 mmHg at booking

Persistent severe headache

Proteinuria at booking

Papilledema and/or visual disturbances (blurred vision, diplopia,
blind spots, flashes of light, or squiggly lines).

Multiple gestation (twins, triplet pregnancies)

Previous pregnancy with PET or IUGR child
Family history of PET (in mother or sisters)
Black race
Maternal age under 20 and possibly maternal age over 35 to 40
Diabetes mellitus or insulin resistance

Hyperreflexia, brisk tendon reflexes (3+)
Pulmonary edema, shortness of breath
Nausea, vomiting
Abdominal pain, persistent new epigastric pain or tenderness

Renal disease

Impaired functional liver tests (elevated alanine
aminotransferase, aspartate aminotransferase)

Thrombophilias and hyperviscosity syndromes

Thrombocytopenia (⬍100×109 /L)

Underlying maternal collagen vascular disease

Microangiopathic hemolytic anemia

Presence of antiphospholipid syndrome, or antibodies
Increased circulating testosterone
Protein D deficiency in mother during pregnancy
Trisomy 13

Table 17.3 Complications of PET

High altitude

(A) Maternal
Central nervous system
r Eclampsia
r Cerebral edema
r Cerebral hemorrhage
r Retinal edema
r Retinal blindness
r Cortical blindness
Liver
r HELLP syndrome
r Acute liver failure
r Hepatic rupture
Renal system
r Acute renal failure
r Renal cortical necrosis
r Renal tubular necrosis
Respiratory
r Pulmonary edema
r Laryngeal edema
Hemostatic system
r Thrombocytopenia
r DIC
r Microangiopathic hemolytic anemias
Cardiovascular system
r Risk factor for later cardiovascular disease
Labor
r Placental infarction
r Placental abruption
r Preterm delivery

Mirror syndrome
Fetal (genetic) factors from donor eggs
Father of Hispanic origin
Parental specific genes

204

ante-natal care is to monitor for signs of PET at each
clinic visit, with assessment of blood pressure, urinalysis and the presence of edema. These visits occur more
frequently in the third trimester of pregnancy, especially in women with risk factors. Most women with
PET experience only mildly increased blood pressure
and small amounts of proteinuria. Edema, especially
in the face and hands, is a frequent sign of PET, but
is not pathognomonic, for many women without PET
also develop edema during pregnancy. Other forms of
hypertensive disorders also occur in pregnancy and
should be considered in the differential diagnosis of
PET.
The maternal manifestations of PET can affect
almost every organ, depending on severity. Possible
complications of PET in the mother and in the fetus
are listed in Table 17.3.

(B) Fetal–Neonatal
r IUGR
r Prematurity
r Death
r Neurological complications
r Later cardiovascular disease

Chapter 17. Pre-eclampsia

Current concepts on the pathogenesis
of PET
Placental dysfunction is the central feature in the
development of PET. In 1939 Ernest Page introduced
the concept that PET may be due to the reduced perfusion of the placenta.

The two stage model of PET
Currently, the development of PET is hypothesized,
to be in two stages, according to a theory introduced
by the Oxford Group in 1991, and supported and
expanded by Roberts.2 The first stage is reduced placental perfusion and the second the maternal response
to this – maternal endothelial cell activation. Failure of
endovascular trophoblast invasion is thought to lead to
relative under perfusion of the placenta. Under conditions of hypoperfusion, the placenta probably releases
factors into the circulation, which then trigger maternal endothelial dysfunction.
Early in normal gestation, cytotrophoblast cells
invade the decidua and myometrium. These cells also
invade endovascularly, replacing first the endothelium
and then the media of the spiral arteries. This creates a system of flaccid, low resistance, large diameter,
unresponsive arterioles, that increase placental perfusion. The outcome is an increment in blood flow to the
fetus and lack of adrenergic vasomotor control. The
endothelial lining is replaced by the cytotrophoblast
cells, which adapt to mimick an endothelial pattern of
adhesion molecule expression.
In PET, this vascular phenotype is not expressed
and the pattern of invasion is much more superficial. There is a restriction of trophoblast invasion
into the spiral arteries, particularly those within the
myometrium. These decidual vessels may later show
atherosis, and superimposed thrombosis augments
hypoperfusion. It seems plausible that, consequent to
these changes, placental hypoperfusion causes a state
of relative hypoxia.
Various factors have been postulated as the substance produced from the placenta that affects blood
flow, arterial pressure, and maternal endothelial cell
activation (ECA). These factors include oxidative
stress, cytokines such as tumor necrosis factor ␣ (TNFa) and IL-6, insulin-like growth factors, nitric oxide
(NO), heparin-binding endothelial growth factor-like
growth factor, endothelin-1, arachidonic acid metabo-

lites, angiotensin II type-1 receptor autoantibody
(AT1-AA), and angiogenic factors.
Recently, the focus has been on angiogenic factors. It has been proposed that PET could be related
to an imbalance between proangiogenic (as vascular
endothelial growth factor, VEGF) and antiangiogenic
factors as Fms-like tyrosine kinase (sFlt-1) and soluble
endoglin (sEng).
sFlt-1 is an endogenous inhibitor of both VEGF
and PGF and may regulate placental angiogenesis by
preventing the interaction between circulating VEGF
and PGF with their proangiogenic receptors. Level
of sFlt-1 in the plasma of women with PET are elevated compared with normal pregnancies. When sFlt1 is exogenously administered via adenovirus mediated gene transfer in pregnant rats and mice, there
are increases in arterial blood pressure and proteinuria, as well as decreased levels of VEGF and PIGF,
similar to these observed in PET.3,4 Another observation was that VEGF infusion attenuates the increased
blood pressure and renal dysfunction observed in
pregnant rats overexpressies sFlt-1.5 Also uteroplacental ischemia has been shown to increase plasma and
placental sFlt-1, and to decrease the levels of VEGF and
PIGF in late gestation of rats and baboons.6 Endoglin
(Eng) is a component of the transforming growth
factor (TGF)-beta receptor complex and a hypoxia
inducible protein, and is related to cellular proliferation and NO signaling. Soluble Eng (sEng) has been
shown to act as an anti-angiogenic factor, possibly via
the inhibition of TGF-beta of binding to cell surface
receptors.7 Recent works showed that sEng inhibits in
vitro endothelial cell tube formation and that adenovirus mediated increase of both s-flt-1 and sEng in
pregnant rats resulted in IUGR and in a syndrome
resembling PET.8 Recently, sEng levels have been proposed as a predictor of PET.9
Whichever factor provokes maternal ECA, when
the latter is established, it leads to upregulation
of a number of inflammatory molecules, including
adhesion molecules. These procedures change the
endothelium phenotype from antithrombotic to prothrombotic, with a decrease in the formation of the
vasodilator and antiplatelet agents prostacyclin and
nitric oxide, the production of endothelin, and finally
the downregulation of anticoagulant systems.
Redman’s research group in Oxford proposed that
the endothelial dysfunction seen in PET is part of
a wider inflammatory response and that placental
hypoperfusion is not necessarily the sole primary

205

Section 6. Microangiopathies

First stage: reduced placental perfusion, abnormal implantation/vascular remodeling
Why? Unknown. Proposed factors:
Hypoxia, ischemia, oxidative stress, altered
NK cell signaling, syncytial debris, altered
hemeoxygenase expression, etc.
Reduced perfusion of
the placenta

Oxidative stress, cytokines (TNF-a, IL-6), insulin-like growth factors, NO, heparin-binding endothelial growth factorlike growth factor, endothelin-1, arachidonic acid metabolites, angiotensin II type-1 receptor autoantibody (AT1-AA)
and angiogenic factors (sFlt-1, sEng)

↑ Blood flow and arterial pressure
Maternal endothelial cell activation (ECA)

Endothelin, reactive oxygen species (ROS), thromboxane, 10-HETE, ↑ on vascular sensitivity to
angiotensin II, ↓ vasolidators (as nitric acid (NO) and prostacyclin)

Generalized dysfunction of the maternal vascular endothelium

Maternal constitutional factors

Second stage: maternal syndrome
Fig. 17.1 Summary of current concepts on the pathogenesis of PET. The two stages model of PET.

206

event. They argue that pregnancy normally elicits an
inflammatory response.10 This is evidenced by changes
in granulocytes and monocytes such as increased
intracellular production of reactive oxygen species
and upregulation of surface molecules such as CD11b
and CD64, as well as release of L-selectin, which is
related to granulocyte activation. During PET there
is increased activation of platelets, neutrophils, and
monocytes and an increase in the release of microparticles when compared with normal pregnancy. Perhaps these inflammatory changes are a response to
the presence of fetal (or paternal) antigens. If so, then
abnormalities of the normal immunomodulation seen
at the feto-placental interface could act to trigger PET.
HLA-G is important in the prevention of recognition
of the placenta as “non-self” and there is a reduction
of expression of HLA-G in PET along with abnormal

responsiveness of maternal lymphocytes towards fetal
cells.
Microparticles are fragments of cell membranes
released into the circulation as a result of cellular activation or apoptosis and can have a procoagulant effect.
Microparticles in pregnancy are derived from a number of cells, but the predominant population is platelet
derived. Vesicles prepared from syncytiotrophoblast
microvillous membranes (STBM) have been shown to
suppress the proliferation of endothelial cells in vitro.
They also affected an in vitro model of endothelial celldependent arterial relaxation. The numbers of STBM
detected in the circulation of pre-eclamptic women
have been shown to be significantly elevated compared
with those with normal pregnancies.
A summary of the currently favored pathogenesis
of PET is shown in Fig. 17.1.

Chapter 17. Pre-eclampsia

Relation between PET and IUGR/ fetal
growth restriction (FGR)
The consequences of placental dysfunction can be
twofold – intra-uterine growth restriction and the
maternal symptoms and signs of PET. What is not
understood is why some women only have FGR, while
others have both FGR and PET. It has been suggested that the maternal syndrome of PET may only
occur in women with “constitutional factors” (genetics, environmental, dietary, behavior, etc.) that render
the mother sensitive to the effects of reduced placental
perfusion.
Constitutional factors that have been proposed to
act as the inductors of the maternal syndrome of
PET, include several dietary factors, metabolic conditions such as diabetes, insulin resistance and uric acid,
low melatonin levels, obesity, metabolic syndrome,
folic acid and hyperhomocysteinemia, hyperlipidemia
with elevated triglycerides, free fatty acids and LDL
cholesterol and reduced HDL, maternal vitamin D
deficiency, and thrombophilia. The factors that cause
ECA may contribute to the development or severity
of PET.

Relation of PET with later cardiovascular risk
in women and their babies
Despite PET and FGR occurring only in pregnancy,
they have been shown to have long-term consequences. Mothers, who have had PET or have delivered a baby with FGR, experience a 2–8-fold increased
risk of atherosclerotic cardiovascular disease (CAD)
in later life.11 It is unclear whether PET causes CAD
or whether these two entities share the same causal
origin. It has also been shown that the earlier PET
presents in pregnancy, the more severe the maternal
CAD is. Women with PET before 37 weeks of gestation had eight times more cardiovascular deaths than
woman with normal pregnancy 14 years later.
There is a large body of epidemiological studies
showing that the long-term consequences of FGR in
the baby last well into adulthood. These individuals have a predisposition to develop a metabolic syndrome later in life, manifesting as obesity, hypertension, hypercholesterolemia, cardiovascular disease,
and type 2 diabetes, in agreement with the theory
of early origin of CAD, also known as “the Barker
hypothesis.”12

A recent study showed positive associations
between maternal pre-pregnancy levels of triglycerides, cholesterol, low-density lipoprotein, and
baseline systolic blood pressure and subsequent
development of PET.13 The authors concluded that
the presence of cardiovascular risk factors prior to
pregnancy, are predisposing to PET. The prevalence
of chronic hypertension is significantly higher among
women with a history of PET (46.7%) as well as those
with previous IUGR (8.9%).14 Women with PET and
FGR with chronic hypertension on follow-up had
increased carotid intimal-media thickness, suggesting
a predisposition to atherosclerosis. Women with
previous PET have significantly higher fasting glucose
levels, waist circumference, body mass index, and
higher prevalence of metabolic syndrome compared
to normal women

New modifications of current theories
on the pathogenesis of PET
As long as the initial causative factor for PET remains
unrecognized, different theories continue to be generated, some of them challenging the currently accepted
origins of PET.
It has been suggested that early (before 34 weeks)
and late onset (after 34 weeks) PET are two different clinical entities with different pathogenesis, origins, etiology, severity, and clinical expression. Certainly, FGR is more strongly associated with severe
rather than with milder pregnancy-induced hypertension.15 According to this theory, early PET is associated
with reduced perfusion but PET at term may not, suggesting different genetic origins for early and late PET.
Huppetz, in a recent paper, has challenged the
placental origins of PET and proposed that PET is
a syndrome of early placental formation.16 He suggested that an insult results in aberrant development
and differentiation of the villous syncitiotrophoblast
causing impaired maintenance of the placental barrier. This subsequently leads to the release of necrotic
and aponecrotic fragments culminating in a systemic
inflammatory response of the mother. According to
this theory FGR is due, in contrast, to a failure of
extravillous trophoblast invasion. This new concept
clearly separates the origins of PET and FGR, and proposes alterations in different trophoblast differentiation pathways as origins of both syndromes.
Genome-wide expression analysis in rodents
showed that spontaneous differentiation of

207

Section 6. Microangiopathies

trophoblast stem cells is associated with the acquisition of an endothelial-cell like thromboregulatory
gene expression program.17 This program is developmentally regulated and conserved between mice and
humans. They further showed that trophoblast cells
sense, via the expression of protease activated receptors, the presence of activated coagulation factors.
Engagement of these receptors results in cell-type
specific changes. These observations define candidate
fetal genes that are potential risk modifiers of PET
and suggest that hemostasis can affect trophoblast
physiology and thus affect placental function in the
absence of frank thrombosis. It is postulated that PET
is not only due to a maternal cause, but also that fetal
genes could contribute to the development of the
disease.

Thrombophilia and PET
Acquired thrombophilia
Mothers with antiphospholipid antibodies have a predisposition to PET and FGR. Indeed, the development of these conditions before 34 weeks in a woman
with antiphospholipid antibodies has now become
defining criteria for obstetric antiphospholipid syndrome. This is discussed in more depth in Chapter 11.
Other acquired conditions that predispose to thrombosis such as myeloproliferative disease would also be
expected to predispose to PET (see Chapter 19).

Genetic thrombophilias

208

An association between PET and inherited thrombophilias was first reported by Dekker et al. in 1995,
who proposed that maternal thrombophilia could act
as a genetic constitutional factor for the development
of PET.18 Since then, a large number of retrospective
and case-controlled studies have examined the association between different types of thrombophilic mutations and PET. The results of published reports have
been inconsistent. Meta-analysis of all case-control
studies suggests that only FVL mutation is associated with a minor increased risk of PET (odds ratio,
l.18; 95% confidence interval, 1.14 to 2.87). Overall, studies suggest that women with genetic thrombophilia have more severe PET than those without, but thrombophilia itself does not precipitate the
condition.

Prediction of PET
As the exact causative factor that provokes PET is not
yet known, at present there is no clear strategy for its
prevention and so the clinical and research focus has
been on early detection and prediction.
Hyperuricemia is an established marker of severe
PET, correlating with the histological severity of renal
lesions, and clinically with adverse fetal outcomes, but
has a low negative predictive value.
Uterine artery Doppler screening between 20 and
24 weeks identifies mothers at high risk for developing
adverse pregnancy outcomes. The correlation between
elevated uterine artery resistance and a high risk of
PET and/or FGR was first demonstrated at the end of
second trimester, probably reflecting the ongoing process of trophoblast invasion into the spinal arteries.
Bilateral notching at 20–24 weeks identifies the pregnancies that will have FGR and PET, although there is
a high false positive rate.19
An algorithm of placental and endothelial markers between 20 and 24 weeks’ gestation was developed
and showed good prediction of the later development
of PET.20 This study proposed six markers as potential predictive indicators: HDL cholesterol, PAI-1/PAI2 ratio, leptin, and PIGF. At 20 weeks’ of gestation, an
algorithm of these markers distinguished PET from
the low risk group. At 24 weeks’ of gestation the positive predictive value was even better. Increased levels of
soluble fms-like tyrosine kinase 1 (sFlt-1) and reduced
levels of soluble placental growth factor (PIGF) have
been shown to predict the subsequent development
of PET, as early as 5 weeks before the onset of PET.
Human cancer patients treated with anti-VEGF antibody developed hypertension and proteinuria.
In association with increased levels of sFlt-1,
symptoms were dramatically worse, and typical of
HELLP syndrome, leading the authors to postulate
that increased levels of sFlt-1 were responsible for PET,
but the combination of increased sFlt-1 and sEng led
to HELLP syndrome. In a longitudinal analysis, the
rise in soluble endoglin concentrations occurred earlier and was more marked in pregnancies with subsequent pre-eclampsia.
Soluble endoglin (sEng) is a co-receptor for
transforming growth factor ␤1 and ␤3, expressed
on trophoblasts. Its levels are increased in preeclampsia15 and in pregnant rats this has been
associated with increased vascular permeability and
hypertension. Other serum markers that have been

Chapter 17. Pre-eclampsia

proposed to predict PET as early as the first trimester,
are placental protein 13 (PP13), placenta associated
plasma protein A (PAPP-A), and long pentraxin 3
(PTX3). All of those markers still need further evaluation in larger multicenter trials.

Management of pre-eclampsia
At present, the sole effective therapy for pre-eclampsia
is delivery and removal of the placenta. Symptoms usually improve within days. Therefore, early diagnosis
and timely delivery are imperative for maternal and
peri-natal survival.

Prevention of PET
Several drugs have been tried for the prevention of
PET. Despite the first promising publications, it has
been shown later that there is a lack of evidence for calcium, vitamin C, and E in PET’s prevention. The main
drugs that are used for the prevention of PET are antihypertensives and antithrombotics.

Prevention of PET with antihypertensives
Antihypertensive drugs are used for secondary prevention of PET, in women with mild to moderate
hypertension developing or pre-existing to pregnancy.
Data from several studies showed that, although there
was a reduction in hypertension, it was unlikely that
this had a major impact on the progression to PET.
Furthermore, it has been argued that the impact to
the fetus of lowering maternal blood pressure could
provoke FGR. Although there is no big randomized
trial, beta-blockers are more likely to have such an
impact (eight trials, 810 women; relative risk 1.56,
1.10 to 2.22). The antihypertensive drug methyldopa
has often been used in gestational hypertension. Side
effects include depression and drowsiness. Other drugs
that can be used are labetalol and calcium channel
blockers. Atenolol is relatively contraindicated in pregnancy due to possible FGR; absolutely contraindicated are angiotensin converting enzyme inhibitors
and angiotensin receptor antagonists due to possible
teratogenicity. Diuretics should be avoided in general,
and should be kept only for special indications such as
renal or cardiac diseases.

Prevention of PET with antithrombotics
Antiplatelet agents
The Collaborative low-dose Aspirin Study in Pregnancy (CLASP study), was a randomized trial of low

dose aspirin for the prevention and treatment of PET
among 9364 pregnant women.21 The women were
randomly assigned 60 mg aspirin daily or matching
placebo. To simulate real obstetric practice, the entry
criteria were broad and embraced women thought to
be at risk of PET and FGR from 12 to 32 weeks’ gestation. Primiparous women, women with pre-existing
hypertension or a history of FGR, PET, or stillbirth
and women with established PET could all be entered
in the study: 74% were entered for prophylaxis of
PET, 12% for prophylaxis of FGR, 12% for treatment
of PET, and 3% for treatment of FGR. Overall, the
use of aspirin was associated with a reduction of only
12% in the incidence of proteinuric PET, which was
not significant. Nor was there any significant effect on
the incidence of IUGR or of stillbirth and neonatal
death. Aspirin did, however, significantly reduce the
likelihood of preterm delivery (7% aspirin vs. 2% control); absolute reduction of 5 per 100 women treated.
There was a significant trend towards progressively
greater reductions in proteinuric pre-eclampsia, the
more preterm the delivery. Aspirin was not associated
with a significant increase in placental hemorrhage or
in bleeding during preparation for epidural anesthesia, but there was a slight increase in use of blood
transfusion after delivery. Low dose aspirin appeared
safe for the fetus and newborn infant, with no evidence of an increased likelihood of bleeding. The rate
of stillbirth, neonatal death, or fetal growth retardation occurring before 32 weeks was 5.3% in the aspirin
group as compared with 10.6 % in the placebo group.
These findings do not support routine prophylactic or
therapeutic administration of aspirin in pregnancy to
all women at increased risk of pre-eclampsia or IUGR.
Low dose aspirin may be justified in women judged to
be especially liable to early-onset PET severe enough
to need very preterm delivery. In such women it seems
appropriate to start low dose aspirin prophylactically
early in the second trimester.
The Cochrane Library Update summarizing data
from 37 560 women for 59 trials of aspirin to prevent PET showed that the use of aspirin is associated with a 17% reduction in the risk of pre-eclampsia
(46 trials, 32 891 women, relative risk (RR) 0.83, 95%
confidence interval (CI) 0.77 to 0.89), an 8% reduction in the relative risk of preterm birth (29 trials,
31 151 women, RR 0.92, 95% CI 0.88 to 0.97); NNT
72 (52 119)), and a 14% reduction in fetal or neonatal
deaths (40 trials, 33 098 women, RR 0.86, 95% CI 0.76
to 0.98); NNT 243 (131, 1 666) and a 10% reduction in

209

Section 6. Microangiopathies

small-for-gestational age babies (36 trials, 23 638
women, RR 0.90, 95% CI0.83 to 0.98). The authors concluded that antiplatelet agents, largely low dose aspirin,
have moderate benefits when used for prevention of PE
and its consequences.22
The Perinatal Antiplatelet Review of International Studies (PARIS) Collaborative Group published
a meta-analysis, included 31 randomized trials of
PET primary prevention enrolling a total of 32. 217
women and their 32.819 infants.23 According to their
results antiplatelet agents, particularly aspirin, moderately reduce the relative risk for PET, preterm births
before 34 weeks’ gestation, and serious adverse pregnancy outcomes. For women randomized to receive
antiplatelet agents, the relative risk of developing PET,
compared with women in control groups, was 0.90
(95% confidence interval (CI) 084–0.97). The risk of
delivering before 34 weeks’ gestation was 0.90 (95% CI,
0.83–0.98) and of having a pregnancy with a serious
adverse outcome was 0.90 (95% CI, 0.85–0.96). Use of
antiplatelet agents was not associated with any significant effect on the risk for death of the fetus or newborn, risk of having an infant born small for gestational
age, or risk for bleeding events for either the women or
their babies. No subgroups of women who were substantially more or less likely to benefit from antiplatelet
agents than any other were identified.23 Despite these
two large meta-analyses, further studies are required
to assess which women are most likely to benefit, when
treatment is best started, and at what dose.

Heparin and antithrombin concentrates

210

Heparin as monotherapy or in combination with
aspirin has also been suggested for the prevention of
PET in women with high risk pregnancies, but data are
not yet sufficient for a final conclusion. For example,
a recent study investigated the effect of low molecular
weight heparin (LMWH) on pregnancy outcome, on
the maternal blood pressure values, and on uteroplacental flow in angiotensin-converting enzyme (ACE)
non-thrombophilic women, with insertion/deletion
(I/D) polymorphism, with history of PET.24 The study
included 80 women, 41 treated with dalteparin 5000
IU/day, and 39 untreated (control group). This study
suggests that LMWH may reduce the recurrence of
PET, of negative outcomes, and the resistance of uteroplacental flow, and also prevents maternal blood pressure increase in ACE DD homozygote women with a
previous history of PET.

Antithrombin (AT) levels are reduced in PET. Previous randomized controlled trials of AT therapy in
PET between 24–35 weeks’ gestation have shown significantly improved maternal symptoms and birth
weight.25 A further trial examined AT therapy in
severe PET in women presenting before 32 weeks’ gestation. 42 patients were enrolled and each received
3000 IU per day for 7 days compared to albumin
582 mg/day for 7 days. An equal number of women
discontinued the intervention in the AT and placebo
(albumin) groups. AT treatment improved or at least
preserved fetal biophysical status. It prolonged the
pregnancy to reach 34 weeks and fetal growth rate
was preserved. However, AT treatment of PET is still
largely confined to research settings.

Planning for the optimal timing of delivery
One can justify PET, of any severity, presenting after
34 weeks as an indication for delivery. If earlier than
34 weeks, the balance of expectant management is set
against risk to the mother, but potentially benefits the
child in terms of risks of prematurity. Generally, hemodynamic instability, fetal distress, and rapid disease
progression are indications for delivery. There is no
evidence base to support these decisions, as only small
trials of expectant management prior to 34 weeks vs.
delivery have been carried out.
If an induced pre-term delivery is contemplated,
it may be necessary to give prostaglandins to ripen
the cervix. Steroid therapy to improve fetal lung maturity should also be considered, in discussion with the
pediatric team. In general, a vaginal delivery is considered safer than Cesarean section for those with complications of PET. For both forms of delivery, a platelet
count of greater than 50 × 109 /l is recommended, and
platelet transfusions may be necessary to achieve this.
Regional anesthesia is also generally preferred, but
depends on the platelet count, and guidelines recommend a count of greater than 80 × 109 /L, in the setting of normal platelet function. Coagulation parameters should also be checked prior to delivery because
of the risk of DIC in PET.
It should be emphasized that the disease does not
abate immediately post-delivery and that seizures can
occur up to a week later. Hence, seizure prophylaxis,
anti-hypertensive therapy, and frequent monitoring
should be continued for an appropriate period, e.g. 12–
48 hours for seizure prophylaxis and close monitoring,

Chapter 17. Pre-eclampsia

up to 12–16 weeks or indefinitely for anti-hypertensive
therapy.

Other pharmaceutical management of PET
Anti-hypertensive drugs for the management of PET
The most used anti-hypertensive drugs in the management of PET are methyldopa, labetalol, and nifedipine.25,26 Labetalol is quite safe and effective, decreasing heart rate and having fewer side effects than
other drugs (lack of reflex tachycardia, hypotension, or
increased intracranial pressure).25 Best avoided drugs
are high dose diazoxide, due to increased risk for
hypotension and Cesarean section, and the serotonin
receptor antagonist kentaserin.25
In general, angiotensin converting enzyme (ACE)
inhibitors, angiotensin receptor-blocking drugs
(ARB), and diuretics should be avoided. Nifedipine
should be given orally and not sublingually.
Concern has been about hydralazine as first-line
treatment (due to the potential unpredictable hypotension) and the combination of nifedipine and magnesium sulfate.

Magnesium sulfate
Magnesium sulfate is the drug of choice for the prevention and treatment of pre-eclampsia. The epidemiological and basic science evidence suggesting that magnesium sulphate when given to early pregnancy in
women considered at risk of preterm birth may be neuroprotective for the fetus, has now being confirmed
by a recent Cochrane systematic review.26 It acts by
causing cerebral vasodilation, thereby reversing the
ischemia produced by cerebral vasospasm during an
eclamptic episode. Data suggest that women receiving
magnesium sulfate therapy have a 58% lower risk of
eclampsia than placebo and that also reduces the risk
for maternal death.26 A possible side effect is flushing,
which occurs in one-quarter of women.

Suggesting guidelines for the management
of established PET
r Close in or outpatient monitoring of vital signs,
deep tendon reflexes, neurological examination.
r Bed rest and relaxation.
r Fetal monitoring: external fetal monitor,
oxcytocin challenge test, biophysical profile.
r Give steroids to accelerate fetal lung maturation
when ⬍ 34 weeks of gestation; betamethasone

r
r
r
r

12 mg IM/day for 2 doses, or dexamethasone 6 mg
IM/ 12 hours × 4 doses.
Careful fluid restriction to reduce the risk of fluid
overload. Total fluid intake should be limited to
80 mL/h (max 150 mL/h), or 1 mL/kg/h, urine
output can be tolerated as low as 10 mL/h.
Give supplemental oxygen.
Maintain diastolic blood pressure ⬍ 110 mmHg/
and systolic ⬍ 160 mmHg with anti-hypertensive
drugs.
Give prophylactic intravenous magnesium sulfate
for the prevention of eclampsia during labor and
the postpartum.
Laboratory monitoring; complete blood count,
platelets count; coagulation studies in severe PET
(PT, PTT, fibrinogen, FDP) urea, serum
creatinine, uric acid, serum electrolytes, liver
functional tests, lactate dehydrogenase.

Suggesting guidelines for the management
of eclampsia
r Close monitoring.
r Give oxygen.
r Fluid restriction is advisable to reduce the risk of
fluid overload. Total fluid should be limited to
80 mL/h, or 1 mL/kg/h.
r Give magnesium sulfate. Alternative drugs
include diazepam, phenytoin.
r Give steroids if ⬍34 weeks’ gestation.
r Urgent delivery.

Hematological complications of PET
All the changes taking place during PET due to
endothelial cell activation can produce hematological
complications.
Frequent (at least every 8 hours) full blood count
and coagulation screen should be performed in case of
severe PET, or where there is suspicion of subsequent
development of hematological complications.

Thrombocytopenia
The most common hematological complication of
PET is thrombocytopenia, occurring in 18% of preeclamptic women. This is probably due to platelet
and endothelial activation generating thrombin and

211

Section 6. Microangiopathies

causing platelet consumption. In general, the severity
of thrombocytopenia is related to the severity of PET.
If the platelet count is greater than 40 000 × 109 /L,
the risk of bleeding is small. In the majority of cases
thrombocytopenia resolves after delivery, but rarely
may continue to fall after birth. Severe thrombocytopenia persisting after delivery could be a possible
indicator of developing microangiopathic hemolytic
anemia.

Management of thrombocytopenia in PET
Platelet counts of ⬎ 50 × 109 /L in patients with otherwise normal coagulation are regarded as safe for normal vaginal delivery and Cesarean section. Concerns
over the risk of hematoma formation and neurological damage have led to the use of regional anesthesia
not being recommended unless the platelet count is
⬍ 75× 109 /L with a normal coagulation screen. This
recommendation is based on consensus rather than on
evidence.
If platelet count ⬍50 × 109 /L and there is no bleeding, then no treatment is necessary unless there is
active bleeding, when it is appropriate to transfuse
platelets.

Disseminated intravascular
coagulation (DIC)

212

DIC is a clinicopathological syndrome characterized
by a systemic activation of coagulation leading to
microvascular deposition of fibrin, and thus to consumption of coagulation factors, platelets and physiological anticoagulants. This produces a reduction in
platelet count, a fall in fibrinogen, and a prolongation
of the activated partial thromboplastin time (APTT)
and international normalized ratio (INR).
Prolongation of PT and APTT with severe thrombocytopenia and low fibrinogen levels (⬍1.0 g/L) are
signs of a developing DIC-like state and hence frequent estimation of platelet count, fibrinogen (using
Clauss method), prothrombin time (PT), and APTT
is strongly recommended. Laboratory evidence of a
consumptive coagulopathy should be sought before
microvascular bleeding becomes evident, so that
appropriate and aggressive action can be taken to
address the underlying cause.
DIC occurs in about 10%–12% of all cases of PET
and in 7% of severe PET. The etiology of DIC in

pre-eclampsia is not well understood, but is probably
a consequence of endothelial cell activation. In only
10%–15% of DIC cases in PET, it can become more
systematic and even lethal. In PET there is a low grade
fibrin deposition in the renal and placental microcirculation.
DIC in obstetric patients could be a complication
of other obstetric conditions or of none related directly
with pregnancy. The most common causes of DIC in
obstetrics, besides PET, are abruption placentae and
amniotic-fluid embolism (occurring in more than 50%
of obstetric cases), and retained dead fetus, sepsis, and
septic abortion.

Management of DIC
Management of DIC involves (1) treating the cause and
(2) replacement of missing hemostatic components
with blood products. Rarely, chronic DIC requires low
dose anticoagulation to “switch off” the stimulus to
DIC.
Hematological treatment consists of platelets, FFP,
and cryoprecipitate (see Table 17.3, Chapter 13c),
but avoiding circulatory overload. Novel therapeutic
strategies are based on current insights into the pathogenesis of DIC, and include anticoagulant strategies
(e.g. directed at switching off coagulation stimulus)
and strategies to restore physiological anticoagulant
pathways (such as activated protein C concentrate).
These have not been evaluated adequately in the management of DIC in pregnancy and postpartum.

HELLP syndrome
Definition
HELLP syndrome (hemolysis, elevated liver enzymes,
low platelets) occurs in the second and third trimester
of pregnancy and presents occasionally postpartum.
There are no clear definition criteria for HELLP.

Epidemiology of HELLP
This disorder complicates between 0.5% and 1% of
pregnancies and is associated with a maternal morbidity ranging between 1% and 4%. HELLP syndrome is
reported in PET with an incidence ranging between
2% and 50% (5% and 15%), depending on the population studied and the diagnostic criteria used: 70%
of cases occur ante-natally and 30% occur within the
first 48 hours’ to 7 days’ postpartum. 20% of women
who develop HELLP post-labor had no evidence
of PET before delivery. The incidence of HELLP is

Chapter 17. Pre-eclampsia

Table 17.4 Differential diagnosis of HELLP syndrome
Acute fatty liver of pregnancy
Gal bladder disease
Gastroenteritis
Appendicitis
Diabetes insipidus
Hemolytic uremic syndrome
Thrombotic thrombocytopenic purpura
Idiopathic thrombocytopenic purpura
Acute renal failure
Pyelonephritis
Glomerulonephritis
Peptic ulcer
Flair of systemic lupus erythematosus
Viral hepatitis

significantly increased among white middle-class and
older multiparous women. DIC is founded in approximately 20%–30% of women with HELLP. Recurrence
rates in subsequent pregnancies is 3% for HELLP,
10%–14% for IUGR and 18%–20% for PET.

Clinical presentation of HELLP
The clinical presentation is with fatigue and malaise
for a few days, followed by nausea, vomiting, shoulder, neck, epigastric or right upper quadrant pain,
headache, and visual disturbances. Right upperquadrant or epigastric pain is thought to be due to
obstruction of blood flow in the hepatic sinusoids,
which are blocked by intravascular fibrin deposits.
Usually, the patients present with significant weight
gain, due to the associated generalized edema, and
with proteinuria greater than 1+ (in 90% of cases).
Severe hypertension is not a constant or a frequent
finding in HELLP syndrome. That is why it can usually be misdiagnosed as having another disease (listed
in Table 17.4).

Pathophysiology of HELLP syndrome
The pathophysiology is not clear, but it is helpful to
consider that it represents PET confined to the liver,
which may result in necrosis of areas of the liver.
According to one theory, pre-eclamptic patients are
already prone to spontaneous hemorrhages. The liver
is thought to be particularly prone because fibrin split
products can deposit in the reticuloendothelial system of the liver. Multiple previous subclinical sponta-

neous hemorrhages within the small hepatic sinusoids
and arterioles may go unnoticed symptomatically and
leave the liver in a fragile state. Fibrin thrombi may be
left uncleared in the liver. Occasionally, a trigger (such
as DIC) may cause extreme hypoperfusion of the liver,
leading to infarction.
As the liver is the primary site of plasma protein
production and pregnancy is a hypermetabolic condition, a specific plasma protein profile was noted in
women with HELLP syndrome compared with normal control cases. The primary candidate identified
was serum amyloid A (SAA), which was significantly
different between the HELLP cases and controls. However, further work is needed to determine if this is truly
a predictive marker for the development of HELLP or
merely a surrogate of liver impairment.

Complications of HELLP
Possible complications of HELLP syndrome include
subcapsular hematoma of the liver, liver rupture,
excessive bleeding, DIC, pulmonary edema, acute
renal failure, abruptio placentae, peri-natal asphyxia,
fetal death, and maternal death.

Diagnosis of HELLP syndrome
The diagnosis is made by the findings of fragmentation on the blood film, low platelets and abnormal liver
function tests, and with abdominal ultrasound. The
patient may or may not have signs of PET.

Management of HELLP syndrome
Stabilization of hypertension, if present, and other
manifestations of HELLP, such as seizures or DIC are
required as well as fetal monitoring. The only certain therapeutic measure is prompt delivery, and in
the majority of cases women have complete recovery
within 24–48 hours after labor, although some women
may continue to have symptoms for up to 14 days.
In the majority of patients, normalization of platelet
count and resolution of HELLP occurs 5 days postpartum. If these signs of disease persist beyond 5
days postpartum (and indeed if they don’t begin to
improve within 48 hours of delivery), the diagnosis
of HELLP should be reconsidered. Ideally, all women
with HELLP should be referred to a tertiary hospital. Anti-hypertensive drugs, steroids, and plasma
exchange/plasmapheresis have also been used with
variable results.
A Cochrane review summarized the evidence on
the effects of corticosteroids on maternal and neonatal

213

Section 6. Microangiopathies

mortality and morbidity in women with HELLP syndrome.28 From the five studies reviewed (n = 170),
three were conducted antepartum and two postpartum. Four of the studies randomized participants to
standard therapy, or to the administration of dexamethasone. One study compared dexamethasone with
betamethasone. The conclusions were that there is
insufficient evidence to determine whether steroid
use in HELLP decreases the major maternal and
peri-natal morbidity and the maternal and peri-natal
mortality.
Platelet transfusions and HELLP syndrome
A randomized trial of women with class 1 HELLP
syndrome received either dexamethasone (n = 26) or
dexamethasone and platelet transfusions (n = 20).
Liver function tests were significantly higher in the
steroid plus platelets group. Platelet count normalized significantly faster in the dexamethasone only
group, and the postpartum stay was more prolonged
in the dexamethasone and platelet group. The group
that received platelets reported complications such as
wound dehiscence, wound infection and pulmonary
edema.11 A previous report of intrapartum use of
platelets when platelet count was ⬍40 × 109 /L did
not find a significantly lower incidence of hemorrhagic
complications. As a result, platelet transfusion is not
often used in the management of HELLP.

Massive bleeding secondary to
placental abruption

214

Placental abruption is defined as the premature separation of a normally located placenta. Patients with
defective placentation and abnormal placental vasculature, such as in PET, are predisposed to ischemia
and rupture of these placental vessels, which is thought
to lead to placental abruption. Other risk factors
include smoking and cocaine use. Presenting features include mild vaginal bleeding, signs of hypovolemia, fetal compromise, uterine contractions or
hypertonicity, DIC, and renal failure. Ultrasonography may be useful to confirm the position of the placenta, or the presence of a large hemorrhage, but is
insensitive.
The management of placental abruption, whether
expectant or with delivery depends on the extent of the
abruption, the gestational age of the fetus, and the presence of fetal or maternal compromise. A full review is
beyond the scope of this chapter and is covered in other

sources. In general terms, however, delivery may be
vaginal (usually due to the stimulation of rapid labor in
response to the abruption), or by Cesarean section. The
latter scenario may occur in the case of failed progression of labor or in maternal or fetal instability. Expectant management with or without the use of tocolytics
may be possible if the presentation of bleeding is less
acute and earlier in the pregnancy.
DIC often occurs in association with abruption,
particularly with a complete abruption, and may follow within hours. The specific management of DIC has
already been mentioned. The hemostatic management
of massive bleeding is presented in Chapter 13c.
The maternal complications of placental abruption
include massive hemorrhage, DIC, renal failure, and
amniotic fluid embolism. Fetal complications relate
primarily to premature delivery, i.e. stillbirth (adjusted
relative risk of 8.9), growth restriction (adjusted relative risk of 2.0), and complications of prematurity.

Differential diagnosis of PET and HELLP
by microangiopathic hemolytic
anemias (MAHA)
The differential diagnosis of thrombotic thrombocytopenic purpura (TTP) and hemolytic uremic syndrome (HUS) from PET and HELLP may be difficult
(see Chapter 18). TTP is diagnosed during pregnancy
or postpartum, with 75% of episodes occurring around
the time of delivery.
Postpartum HUS is a rare syndrome of unknown
cause, not related to E. coli (D-). The prognosis is poor
for both the mother and the fetus. It is recognized that
HUS recurs in subsequent pregnancies, although the
reason for that is not known. Many pregnant women
who survive after HUS develop chronic hypertension
and chronic renal failure later in life. Plasma exchange
(PE) has low response rates.

Acute fatty liver of pregnancy (AFLP)
HELLP syndrome should be distinguished from AFLP,
a rare condition, also associated also thrombocytopenia, but without microangiopathic hemolytic anemia.
Clinical presentation is similar with HELLP, occurring
almost always in the third trimester. DIC accompanies
AFLP in 90% of cases. Maternal mortality is approximately 15% and fetal mortality ⬍5%.

Chapter 17. Pre-eclampsia

Summary

r Pre-eclampsia (PET), the new onset of
hypertension after 20 weeks’ of gestation and
proteinuria, resolving after delivery, affects
approximately 2%–14% of all pregnancies and
remains a major cause of maternal and fetal
morbidity and mortality worldwide.
r Placental dysfunction is considered to be the
central feature in the development of PET.
r Current hypothesis is that PET is a two-stage
disease: the first stage is reduced placental
perfusion and the second stage is the maternal
response to this with endothelial cell activation.
r Proposed placental factors produced from the
placenta that affect blood flow, arterial
pressure and maternal endothelial cell activation
(ECA) include oxidative stress, cytokines
(TNF-a,IL-6), and angiogenic factors (VEGF,
s-FLT-1, sEng).
r Maternal constitutional factors that have been
proposed to act as inductors of the maternal
syndrome of PET, include several dietary factors,
metabolic conditions (diabetes, insulin resistance,
and uric acid), obesity, metabolic syndrome, folic

r
r

acid and hyperhomocysteinemia, hyperlipidemia,
maternal vitamin D deficiency, and
thrombophilia.
PET is associated with fetal growth restriction
(FGR), in one-third of cases.
Despite PET and FGR occurring only in
pregnancy, they have been shown to have
long-term consequences for both mother and
fetus. Mothers who have had PET or who have
delivered a baby with FGR, experience a 2–8-fold
increased risk of atherosclerotic cardiovascular
disease (CAD) in later life.
The key to good management is early detection
and secondary prevention with anti-hypertensive
and antithrombotic drugs (aspirin, heparin).
Hematological complications of PET include
thrombocytopenia, disseminated intravascular
coagulation (DIC), HELLP syndrome, and
massive bleeding after placental abruption.
Differential diagnosis includes microangiopathic
hemolytic anemias: (thrombotic
thrombocytopenic purpura, TTP, hemolytic
uremic syndrome, HUS), and acute fatty liver of
pregnancy.

215

Section 6. Microangiopathies

References
1. Lewis, G (ed). The Confidential Enquiry into Maternal
and Child Health. (CEMACH). Saving Mothers; Lives:
reviewing maternal deaths to make motherhood safer
2003–2005. The Seventh Report on Confidential
Enquiries into Maternal Deaths in the United Kingdom,
2007.
2. Roberts JM, Gammill HS. Preeclampsia: recent
insights. Hypertension 2005; 46: 1243–1249.
3. Maynard SE, Min JY, Mercham J et al. Excess placental
soluble fms-like tyrosine kinase 1 (sFlt-1) may
contribute to endothelial dysfunction, hypertension,
and proteinuria in preeclampsia. Journal of Clinical
Investigations 2003; 111: 649–658.
4. Lu F, Longo M, Tamayo E et al. The effect of
over-expression of sFlt-1 on blood pressure and the
occurrence of other manifestations of preeclampsia in
unrestrained conscious pregnant mice. American
Journal of Obstetrics and Gynecology 2007; 196:
396.
5. Li B, Ogasawara AK, Yang R et al. KDR (VEGF
receptor 2) is the major mediator for the hypotensive
effect of VEGF. Hypertension 2002; 39: 1095–1100.
6. Gilbert JS, Babcock SA, Granger JP. Hypertension
produced by reduced uterine perfusion in pregnant
rats is associated with increased soluble fms-like
tyrosine kinase-1 expression. Hypertension 2007; 50:
1142–1147.
7. Levine RJ, Lam C, Qian C et al. Soluble endoglin and
other circulating antiangiogenic factors in
preeclampsia. New England and Journal of Medicine
2006; 355: 992–1005.
8. Venkatesha S, Toporsian M, Lam C et al. Soluble
endoglin contributes to the pathogenesis of
preeclampsia. Nature Medicine 2006; 12:
642–649.
9. Masuyama H, Nakatsukasa H, Takamoto N,
Hiramatsu Y. Correlation between soluble endoglin,
vascular endothelial growth factor receptor-1 and
adipocytokines in preeclampsia. Journal of Clinical
and Endocrinological Metabolism 2007; 92:
2672–2679.
10. Redman CW, Sacks GP, Sargent IL. Preeclampsia: an
excessive maternal inflammatory response to
pregnancy. American Journal of Obstetrics and
Gynecology 1999; 180: 499–506.
11. Irgens HU, Reisaeter L, Irgens LM, Lie RT. Long term
mortality of mothers and fathers after pre-eclampsia:
population based cohort study. British Medical Journal
2001; 323: 1213–1217.

216

12. Barker DJB (ed.) Foetal and Infant Origins of Adult
Disease. London: BMJ Publishing Group, 1992.

13. Magnussen EB, Vatlen LJ, Lund-Nilsen TI, Salvesen
KA, Smith GD, Romundstad PR. Pregnancy
cardiovascular risk factors as predictors of
pre-eclampsia: population based cohort study. British
Medical Journal 2007; 225: 978–981.
14. Berends AL, de Groot CJM, Sijbrands EJ et al. Shared
constitutional risks for maternal vascular-related
pregnancy complications and future cardiovascular
disease. Hypertension 2008; 51: 1034–1041.
15. Rasmussen S, Irgens LM. History of fetal growth
restriction is more strongly associated with severe
rather than milder pregnancy-induced hypertension.
Hypertension 2008; 51: 1231–1238.
16. Huppetz B. Placental origins of preeclampsia:
challenging the current hypothesis. Hypertension 2008;
51: 970–975.
17. Sood R, Kalloway S, Mast AE, Hilard CJ, Weiler H.
Fetomaternal cross talk in the placental vascular bed:
control of coagulation by trophoblast cells. Blood 2006;
107(8): 3173–3181.
18. Papageorgiou AT, Yu CK, Bindra R, Pandis G,
Nicolaides KH. Multicenter screening for
pre-eclampsia and fetal growth restriction by
transvaginal uterine artery Doppler at 23 weeks’ of
gestation. Ultrasound Obstetrics and Gynecology 2001;
18: 441–449.
19. Chappell LC, Seed PT, Briley A et al. A longitudinal
study of biochemical variables in women at risk of
preeclampsia. American Journal of Obstetrics and
Gynecology 2002; 187:127–136.
20. CLASP: a randomised trial of low dose aspirin for the
prevention and treatment of pre-eclampsia among
9364 pregnant women. CLASP (Collaborative low
dose Aspirin Study in Pregnancy) Collaborative
Group. The Lancet 1994; 343: 619–629.
21. Askie LM, Duley L, Henderson-Smart DJ, Stewart LA.
PARis Collaborative Group. Antiplatelet agents for
prevention of pre-eclampsia: a meta-analysis of
individual patient data. Lancet 2007; 369: 1791–
1798.
22. Askie L, Duley L, Henderson-Smart D, Stewart L.
Antiplatelet agents for prevention of pre-eclampsia: a
meta-analysis of individual patient data. The Lancet
2007; 369: 1791–1798.
23. Mello G, Parretti E, Fatini C et al.
Low-molecular-weight heparin lowers the recurrence
rate of preeclampsia and restores the physiological
vascular changes in angiotensin-converting enzyme
DD women. Hypertension 2005; 45: 86–91.
24. Maki M, Kobayashi T, Terao T et al. Antithrombin
therapy for severe preeclampsia: results of a
double-blind, randomized, placebo-controlled trial.

Chapter 17. Pre-eclampsia

BI51.017 Study Group. Thrombosis and Haemostasis
2000; 84: 583–590.
25. Doyle LW, Crowther CA, Middleton P et al.
Magnesium sulphate for women at risk of preterm
birth for neuroprotection of the fetus. Cochrane
Database Systems Review 2009; 1: CD004661.
26. McCoy S, Baldwin K. Pharmacotherapeutic options
for the treatment of preeclampsia. American Journal
Health Systems Pharm. 2009; 66: 337–344.

27. Matchaba P, Moodley J. Corticosteroids for HELLP
syndrome in pregnancy. Cochrane Database Syst. Rev.
2004; (1): CD 002076.
28. Fonseca JE, Mendez F, Catano C, Arias F.
Dexamethasone treatment does not improve the
outcome of women with HELLP syndrome: a
double-blind, placebo-controlled, randomized clinical
trial. American Journal of Obstetrics and Gynecology
2005; 193: 1591–1598.

217

Section 6
Chapter

Microangiopathies

Thrombotic thrombocytopenic purpura
and other microangiopathies
Marie Scully and Pat O’Brien

Introduction
Thrombotic microangiopathies (TMAs) describe the
clinical and pathohistological effects of thrombosis in
small vessels. There is usually thrombocytopenia and
anemia and review of the blood film confirms the
microangiopathic process, with evidence of red cell
fragmentation and often polychromasia. One of the
earliest diagnoses was by Moschowitz in 1924, who
described a young woman with anemia and thrombocytopenia, neurological and renal symptoms, and signs
with fever. This described the typical pentad of features of acute thrombotic thrombocytopenic purpura
(TTP). However, in pregnancy, the differential diagnosis may be very difficult and often clinical suspicion in
conjunction with laboratory parameters requires differentiation from other TMAs, which are specific to
this period. The diagnostic challenge is the differentiation from acute fatty liver of pregnancy (AFLP), preeclampsia (PET) or eclampsia, HELLP (hemolysis, elevated liver enzymes, low platelets), antiphospholipid
syndrome (APS), systemic lupus erythematosus (SLE),
hemolytic uremic syndrome (HUS), and disseminated
intravascular coagulation (DIC) (See Table 18.1).

Moderate to severe thrombocytopenia
presenting during pregnancy
Thrombocytopenia is defined by a platelet count
⬍150 × 109 /L. It results from increased destruction
and/or decreased production and can affect 10%
of pregnancies. The most common is gestational
thrombocytopenia, accounting for 75% of all cases.
Rarely, the count is below 70 × 109 /L, typically in
the third trimester and it returns to normal within 12
weeks postpartum. It is thought to result from a hemodilutional effect in pregnancy and placental platelet

218

destruction. There is very little risk of hemorrhage to
the mother or the fetus.
ITP (immune thrombocytopenic purpura) occurs
in 5% of pregnancies with thrombocytopenia and is
a result of immunological peripheral platelet destruction. Maternal treatment and precautions during delivery may be required, but rarely does it have an effect on
the fetus (Chapter 4).
PET and HELLP account for 21% of all cases of
thrombocytopenia in pregnancy; the platelet count
(and other pathological features) usually return to normal within 3–5 days after delivery.

Placental profiles in high risk
pregnancies
Abnormal uterine artery blood flow in the second
trimester is indicative of an increased risk of placental pathology later in the pregnancy, including intrauterine growth restriction (IUGR) and PET. Uterine
artery Doppler examination is often carried out at
around 24 weeks’ gestation in women considered to be
at increased risk of these disorders. Increased resistance in the uterine arteries (indicated by increased
pulsatility index or “notched” waveforms) are associated with a sixfold increased risk of thrombotic placental injury, leading to IUGR and/or PET, compared with
normal uterine artery Dopplers. However, the sensitivity of this test is poor, so its use is usually restricted
to high risk women. In early pregnancy, increased
levels of biochemical markers such as alpha fetoprotein (AFP), beta-human chorionic gonadotrophin (␤HCG), and decreased levels of placental protein 13
(PP–13), in the absence of Down syndrome and spina
bifida, are associated with an increased risk of PET,
IUGR, placental abruption, and intra-uterine fetal
death. These biochemical markers can improve the

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

Chapter 18. Thrombotic thrombocytopenic purpura and other microangiopathies

Table 18.1 Typical features in pregnancy associated microangiopathies

Abdominal
symptoms

Renal
impairment

Neurological
symptoms

+++

−

+++

MAHA

Thrombocytopenia

Coagulopathy

HBP

PET

HELLP

TTP

+++

HUS

AFLP

SLE

APS

−

MAHA: microangiopathic hemolytic anemia, HBP: High blood pressure.
PET: pre-eclampsia, HELLP: hemolysis, elevated liver enzymes and low platelets, TTP: thrombotic thrombocytopenia HUS: hemolytic
uremic syndrome AFLP: acute fatty liver of pregnancy SLE: systemic lupus erythematosis APS:
±: possibly occurs.
+++: definitive feature.

predictive value of uterine artery Doppler imaging for
the prediction of the smaller subset of women with a
high risk of later developing serious problems related
to placental disease.

Thrombotic thrombocytopenic
purpura (TTP)
TTP is an acute life-threatening disorder associated
with thrombocytopenia, microangiopathic hemolytic
anemia, and symptoms related to microvascular
thrombosis. Clinically, in addition to a low platelet
count (below 150 × 109 /L, but more usually ⬍50 ×
109 /L), patients are anemic secondary to fragmentation hemolysis with an associated acute consumption
of folate. Corresponding blood film changes include
polychromasia, anemia, reduced platelets, and fragmented red blood cells. Bilirubin is often raised, but
the direct antiglobulin test is negative and the clotting screen is normal. Lactate dehydrogenase (LDH)
is increased, often out of proportion to the degree of
hemolysis, due to associated tissue ischemia.
Von Willebrand factor (VWF), a plasma glycoprotein synthesized by megakaryocytes and endothelial
cells, normally circulates as multimers of 500–20 000
kDa. Ultra-large VWF multimers (ULVWFM), which
have a molecular weight greater than 20 000 kDa, and
are not normally detected in plasma, were initially
detected in patients with chronic relapsing TTP. Subsequently, a deficiency of VWF-cleaving protease in
patients with TTP, was defined in 2001 as “a disintegrin and metalloprotease with thrombospondin type
1 motif, member 13” or ADAMTS 13.1 This enzyme is
required to break down ULVWFMs. Failure to do so,

due to an inherited deficiency or acquired reduction
of ADAMTS 13, or due to antibodies to ADAMTS 13,
for example, leads to platelet adhesion and aggregation
of UL VWFMs and resulting microvascular thrombosis. Hence, platelet transfusions are relatively contraindicated in TTP, as infusions potentiate the effects
of platelet aggregation on UL VWFMs.
Pregnancy is a precipitating cause of acute TTP,
accounting for approximately 10%–25% of all cases of
TTP in women. From the Oklahoma registry, 19 of the
61 women of child-bearing age presented with TTP
during pregnancy or postpartum.2 TTP is more common in women (3:2), and 45% of all cases of TTP occur
in women of child-bearing age. There is also a risk
of relapse of TTP during subsequent pregnancies in
women diagnosed with TTP.
Other pregnancy-related thrombotic microangiopathies, such as pre-eclampsia / HELLP and
hemolytic–uremic syndrome may further complicate
the diagnosis of TTP. Gestational thrombocytopenia,
which occurs in around 7% of pregnancies and is a
diagnosis of exclusion, may explain a reduction in
platelet counts, when all other laboratory parameters
are normal. Management approaches differ for these
conditions, although differentiation may be clinically
challenging.

Hemostatic changes of normal
pregnancy-Factor VIII, Von Willebrand
Factor (VWF), and ADAMTS 13
Normal pregnancy is associated with marked changes
in hemostasis, which are hormonally mediated and

219

Section 6. Microangiopathies

protect against severe hemorrhage at the time of
delivery, but ultimately result in a hypercoagulable
state. Factor VIII and VWF increase in parallel in the
first half of pregnancy; thereafter, the increase in VWF
is greater throughout the remainder of pregnancy,
returning to normal levels over the 6 weeks’ postpartum. Reciprocal changes of VWF and ADAMTS 13
have been documented. Therefore, with the increased
VWF in pregnancy, ADAMTS 13 would be expected
to decrease. A review of ADAMTS 13 in normal
women with no history of TTP documented a reduction in ADAMTS 13 activity in the second and third
trimesters of pregnancy. A further study in healthy
women confirmed a reduction in ADAMTS 13 activity
after the first trimester (weeks 12–16) up until the end
of the post-natal period when the levels normalized to
pre-pregnancy levels. ADAMTS 13 activity was lower
in non-pregnant nulliparous women (mean 65%) compared with parous women (mean 83%). In pregnancy
and post-delivery, mean ADAMTS 13 activity was
slightly, but non-significantly, lower in primigravidae
than in multigravidae (68% vs. 74%). ADAMTS 13 was
unaffected by platelet count, but was higher in smokers
than in non-smokers during pregnancy (mean 79% vs.
70%, respectively). There was a significant correlation
between higher VWF:Ag levels and lower ADAMTS
13 activity.3 The reason for the decrease in ADAMTS
13 during pregnancy may be twofold. First, enzyme
levels decrease with excess substrate, VWF. Second,
a hormonal influence, possibly estrogen, may lower
ADAMTS13 levels. A role for the effect of estrogen on
parity, and ADAMTS 13 levels, are in line with estrodiol levels in the pregnant and non-pregnant state.

Women presenting with acute TTP
during pregnancy

220

Women presenting with TTP during pregnancy
appear to fall into two groups: those with congenital
TTP and those with acquired, antibody mediated
TTP. Congenital TTP may first present during pregnancy and these women are more likely to relapse
in subsequent pregnancies. Diagnosis is confirmed
with ADAMTS 13 activity ⬍5%, no evidence of an
inhibitor, and confirmation by mutational analysis
of the ADAMTS 13 gene, revealing a homozygous
or compound heterozygous abnormality. To date,
the published literature includes 14 patients, eight of
whom received plasma during pregnancy.

In women who present with acquired TTP related
to pregnancy, the literature presents varying outcomes.
Successful pregnancy outcome can be achieved in
women with an initial episode of TTP.4 In the Oklahoma registry,5 there were 11 women who had a total
of 17 pregnancies subsequent to a diagnosis of acute
TTP in pregnancy. Two of these pregnancies were
associated with TTP recurrence and neither infant survived. In women with no TTP in a subsequent pregnancy (15/17), infant survival was 80%. However, it
appears from the remaining published literature, with
the proviso that these are small case series and there is
likely to be some reporting bias, that the risk of recurrence in subsequent pregnancies is approximately 50%,
and infant survival rates are around 67%.

Risk associated with pregnancy
in women with previous acquired
idiopathic (non-pregnancy associated)
TTP
A particular concern in women who have had acute
TTP unrelated to pregnancy is the risk of relapse from
TTP during a subsequent pregnancy. From the Oklahoma Registry,5 of 7 women with idiopathic TTP, 3
had recurrent relapsing TTP. In the 12 subsequent
pregnancies following a diagnosis of TTP, 3 developed TTP in pregnancy and infant survival was 67%.
Interestingly, in women who did not relapse from
TTP during pregnancy (9/12 pregnancies, 75%), infant
survival was only 33% (3/9). From the literature to
date, including 20 women who had a total of 26
pregnancies following the diagnosis of acute TTP,
17/26 had a relapse of TTP during pregnancy and
infant survival was 15/26. In those patients in whom
ADAMTS 13 testing was available, normal levels prepregnancy/onset of pregnancy were associated with a
lower likelihood of relapse. Another important feature of women reported in the literature is the number
of complications documented associated with thrombotic microangiopathies, such as pre-eclampsia and
HELLP syndrome, as well as reduced fetal survival (see
Table 18.2). It could be hypothesized that women with
TTP are at increased risk of prothrombotic complications and increased risk of placental infarction, despite
normal routine TTP-based laboratory parameters.
Thrombotic microangiopathies during pregnancy
may be clinically indistinguishable and very difficult
to treat. With the normal reduction in ADAMTS 13

Chapter 18. Thrombotic thrombocytopenic purpura and other microangiopathies

Table 18.2 Complications in pregnancy in women with a history of TTP

Reference

Number of
pregnancies

In utero fetal
death

Maternal
death

Pre-eclampsia/
HELLP

1
8

–

11/2

11x first trimester spontaneous abortions

3 (set of twins)

–

Other

1
–

–

29∗

–

1x fetal distress, 1x placental abruption

–

1x Hypertension, 1x first trimester
spontaneous abortion

∗ : Includes patients with HUS, but excludes those presenting with bloody diarrhea, therefore 29 pregnancies in 18 women.
HELLP: hemolysis with elevated liver enzymes and low platelets.

Table 18.3 Thrombotic Thrombocytopenic Purpura Presenting During Pregnancy

Trimester

Case series
References

Number of women
diagnosed with TTP
during pregnancy

First

Second

Third/postpartum

Total

from the onset of the second trimester, it had originally
been proposed that this was the time of increased
presentation of acute TTP. However, it now appears
that the greatest risk is in the third trimester or postpartum (see Table 18.3).

Treatment of TTP in pregnancy
The combination of thrombocytopenia and MAHA
encompasses a number of diagnoses in pregnancy and
it is often difficult to differentiate TTP from these. The
primary decision is whether delivery will be associated with remission of the TMA (as in PET or HELLP)
or whether plasma exchange should be instigated, as
recovery following delivery is unlikely and there is
a risk of multi-organ dysfunction/ death. A further
complicating issue is the development of HELLP/PET

following delivery, which may occur in 20%–30% of
cases of TTP in pregnancy.
If TTP develops in the first trimester, plasma
exchange (PEX) may allow continuation of pregnancy
with delivery of a live infant. However, as HELLP/preeclampsia or TTP can present in the post-natal period
or there may be progression of symptoms despite delivery, PEX is the most appropriate option. With the
availability of ADAMTS 13 activity measurement and
detection of inhibitors to ADAMTS 13 (or more specifically IgG antibodies), it may be possible to distinguish
TTP from other pregnancy associated TMAs, specifically if ADAMTS 13 activity is ⬍5% and/or if IgG antibodies are present. In HELLP syndrome, ADAMTS
13 activity is reduced (median 31%, range 12%–43%)
but with no inhibitor/antibodies to ADAMTS 13 and
higher VWF levels.

221

Section 6. Microangiopathies

222

Steroids may be useful in HELLP syndrome and
in TTP, but for different reasons. They have been used
empirically in TTP because of the underlying autoimmune basis of the disorder, and in HELLP may accelerate recovery from delivery.
However, women presenting with thrombocytopenia, MAHA, neurological features (such as
stroke/TIAs, seizures, encephalopathy), and renal
impairment, should be treated with PEX until the
diagnosis of TTP is excluded. In women with congenital TTP, the risk of relapse in a subsequent pregnancy
is such that elective plasma therapy during pregnancy
is warranted. Plasma infusions may be satisfactory;
however, to deliver sufficient volumes, PEX may be
required. The optimal frequency of plasma replacement is unknown; the half-life of ADAMTS 13 is
2–3 days and plasma therapy every 2 weeks appears
satisfactory.4
In women with acquired TTP, it is not as easy to
predict who are likely to relapse and the literature is
sparse in this area. The previous history of TTP and
the ADAMTS 13 activity at the onset of pregnancy
may be helpful in differentiating patients most likely
to relapse. A normal ADAMTS 13 at the onset of pregnancy appears to predict women at reduced risk of subsequent relapse.4 However, if there is low ADAMTS 13
activity (⬍5%) at the onset of pregnancy, consideration
should be given to elective therapy to prevent relapse.
In contrast, women with normal ADAMTS 13 activity
at the onset of pregnancy, who maintain normal routine laboratory parameters, ADAMTS 13 activity, and
antibody/inhibitor levels throughout pregnancy, do
not usually require intervention for TTP. A reduction
in ADAMTS 13 activity (⬍10%) may be the trigger for
elective therapy to prevent microvascular thrombosis
during pregnancy.
Supportive therapy during pregnancy has not
been addressed in the literature; specifically, low dose
aspirin (LDA) and/or prophylactic low molecular
weight heparin (LMWH). All patients in our cohort
are maintained on LDA throughout pregnancy and
women with a documented thrombophilia or a past
history of venous thromboembolism (VTE) associated
with TTP are started on prophylactic LMWH. The
aim is to optimize implantation and preserve placental
function as abnormalities of the utero-placental circulation, resulting in insufficiency are established in the
first trimester. LDA/LMWH may be beneficial in other
thrombophilic disorders during pregnancy, reducing
the risk of placental abnormalities secondary to infarc-

Table 18.4 Physiological changes during normal pregnancy

Test

Change in pregnancy

Bilirubin

Unchanged

Aminotransferases

Unchanged

Alkaline phosphatase

Increase two to fourfold

Cholesterol

Increase twofold

Prothrombin time

Unchanged

Fibrinogen

50% increase

Hemoglobin

Decrease in later pregnancy

White cells

Increase

tion. However, this therapy has not been formally
evaluated in pregnancy associated TTP. There are no
data on the microvascular effects of “subacute” TTP
before presentation with thrombocytopenia. Therefore, women with a previous pregnancy loss due to
TTP or low ADAMTS 13 activity at the onset of pregnancy can be assumed to be at increased risk of further
episodes of placental disorders in subsequent pregnancies. Interestingly, especially as reported in the Oklahoma registry data, there were a large number of first
trimester losses in such women. This may be due to the
underlying TTP risk, but there is no conclusive histological confirmation.
Therefore, women with congenital TTP require
therapy with plasma, either as infusions or as
PEX. In women with acquired, previous acute TTP
episodes, the baseline ADAMTS 13 activity, and
inhibitor/antibody status at the onset of pregnancy
may be useful in the identification of those most likely
to relapse. Monitoring of enzyme activity in those
with normal early pregnancy levels may be useful,
but in women with low (⬍5%) ADAMTS 13 activity
and/or raised IgG antibody levels, which appear to
be at increased risk of relapse, elective PEX may be
useful. Adjunctive therapy with LDA in all women
+/− prophylactic LMWH, should be added to help
prevent complications related to placental thrombosis.

Liver disease in pregnancy
There are some changes in liver function in normal
pregnancy (see Table 18.4), but clinically abnormal
liver function can be detected in 3%–5% of all pregnancies. The cause may be coincidental to pregnancy
or pre-existing chronic liver disease may be documented. However, in the majority of cases, pregnancy
itself is the precipitant. Hyperemesis gravidarum

Chapter 18. Thrombotic thrombocytopenic purpura and other microangiopathies

typically occurs in the first trimester and intrahepatic
cholestasis of pregnancy (ICP) in the second or third
trimesters. PET, HELLP and acute fatty liver of pregnancy (AFLP) are also associated with abnormal liver
function.

Intrahepatic cholestasis
of pregnancy (ICP)
ICP has been associated with impaired sulphation
and abnormalities of progesterone metabolism. Clinically, initially there is pruritus, which in 10%–25%
progresses to jaundice associated with 10–20-fold
increases in aminotransferases, but a less marked rise
in bilirubin. The diagnosis is helped by measuring bile
acid levels. Treatment is supportive and ursodeoxycholic acid (UDCA) is used. Steroids, although useful for fetal lung maturation pre-delivery have not
been shown to be beneficial compared with UDCA
therapy. The main risk of raised bile acid levels is
to the fetus; there is an increased risk of placental
insufficiency but more importantly an association with
sudden intrauterine fetal death, the precise cause of
which is not clear. Resolution of the condition occurs
with delivery. However, recurrence occurs in 45%–
70% of subsequent pregnancies or with use of the combined oral contraceptive pill, the progesterone only pill
(mini-pill) appears not to increase the risk of recurrence.

Acute fatty liver of pregnancy (AFLP)
This is a rare disorder (incidence estimated at 1/13 000
deliveries), but is an acute life-threatening illness associated with significant maternal and peri-natal mortality.6 Typically, it presents in the third trimester,
between the 30th and 38th weeks of pregnancy,
although it has been rarely described in the first
and second trimesters. It usually affects primigravid
women, although reports of recurrence in subsequent
pregnancies have been documented.
Clinically, presentation is non-specific with
headache, fatigue, nausea, vomiting (70%), and right
upper quadrant or epigastic pain (50%). Progression
of the illness is often rapid and, early in the presentation, there may be gastrointestinal hemorrhage,
coagulation abnormalities, acute renal failure, infection, pancreatitis, and hypoglycemia. Later in the
disease process, liver failure and encephalopathy may
occur. Early delivery is imperative and improvement

occurs over 1–4 weeks’ postpartum, although an
improvement in liver function is usually seen within a
few days of delivery.
Diagnosis is suggested by the clinical features and
may be confirmed by liver biopsy. Histologically, there
is characteristic microvesicular steatosis and with Oil
Red O staining, cytoplasmic vesiculation as a result
of microvesicular fat. However, because of the acute
presentation and laboratory features including coagulopathy, it is usually not possible to undertake liver
biopsy, and the diagnosis is made by a combination of
clinical and biochemical features.
In routine laboratory tests, there may be a raised
white cell count and thrombocytopenia with normoblasts on the blood film. There is DIC (with prolonged PT, APPT, and reduced fibrinogen). Urea,
creatinine, and uric acid levels are raised, there are
elevated ammonia levels and hypoglycemia. Serum
aminotransferases are markedly raised and alkaline
phosphatase are three to four times the normal level
(although this is raised in normal pregnancy because
of placental production).
The primary differential diagnoses are acute fulminant hepatitis and severe HELLP, although the latter
are less likely to be associated with hypoglycemia and
prolonged PT. The histological features of liver biopsy
are described above.
Pathogenesis: with advances in molecular biology,
it has become evident that AFLP may result from mitochondrial dysfunction. There is a strong association
between AFLP and a deficiency of the enzyme long
chain 3-hydroxyacyl-CoA dehydrogenase (LCHAD)
in the fetus, a disorder of mitochondrial fatty acid
beta-oxidation. ␤-oxidation of fatty acids is a major
source of energy for skeletal muscle and the heart,
while the liver oxidizes fatty acids under conditions
of prolonged fasting, during illness, and at periods of increased muscular activity. Mitochondrial ␤oxidation of fatty acids is a complex process. LCHAD
is part of an enzyme complex, the mitochondrial trifunctional protein (MTP), associated with the inner
mitochondrial membrane. MTP contains four ␣ and
four ␤ subunits. A hydratase enzyme is located in the
amino-terminal domain and LCHAD is located in the
carboxy-terminal region of the ␣ subunit. The ␤ subunit contains thiolase enzymatic activity. Defects in the
MTP complex are recessively inherited and are due
to an isolated LCHAD deficiency, specifically associated with G1548C mutation, with relatively normal
hydratase and thiolase activities. In complete MTP

223

Section 6. Microangiopathies

deficiency, there is a marked reduction in all three
enzymes. A few hours after birth, children with these
disorders, which are primarily LCHAD, present with
non-ketotic hypoglycemia and hepatic encephalopathy, progressing to coma or death if untreated.
Studies suggest an association between fetal MTP
defects and AFLP. In one study, in every pregnancy
in which the fetus had an LCHAD deficiency, the
mother developed AFLP or HELLP syndrome. Subsequent work in pregnancies without a LCHAD deficient fetus found that the pregnancy progressed normally, with no liver dysfunction. In another study of
prospectively screened mothers who developed AFLP
(27 pregnancies) or HELLP (81 pregnancies), 5 fetuses
in the AFLP group, but none in the HELLP group, had
an MTP mutation.
The precise mechanism by which a LCHADdeficient fetus causes AFLP in a heterozygote mother
remains unclear. However, there are several hypotheses. The mother who is heterozygote for an MTP
defect has reduced capacity to oxidize long chain fatty
acids. The stress of pregnancy associated with altered
metabolism, increased lipolysis, and decreased ␤ oxidation, and the hepatotoxic LCHAD produced by the
fetus or placenta may accumulate in the maternal circulation. Therefore, approximately one in five women
who develop AFLP may carry an LCHAD-deficient
fetus. Screening of newborn infants at birth for this
disorder of fatty acid oxidation can be lifesaving and
allows for genetic counseling in subsequent pregnancies.

Hemolysis, elevated liver enzymes
and low platelets (HELLP)

224

This is a microangiopathy associated with endothelial cell injury, fibrin deposition, platelet activation
and consumption, and areas of hepatic hemorrhage
and necrosis. The underlying precipitating cause is
unknown but it occurs only in pregnancy and the
incidence is between 0.17% and 0.85% of all live
births. Maternal mortality is 3%–4%, with fetal mortality reaching approximately 25%, mainly due to prematurity. Diagnostically, there is considerable overlap
with other TMAs especially PET, and they may represent different points on a single pathological spectrum
(see Chapter 17). There are no obvious precipitating
factors associated with development of HELLP and
it typically presents between the second and third
trimesters, although approximately a quarter of all

cases are postpartum.7 Typical presenting symptoms
include upper abdominal pain and tenderness, nausea,
vomiting, malaise, headache, and rarely jaundice.
There are no clinical or laboratory factors that are
diagnostic, but bilirubin is not usually raised. Aminotransferases can be marginally increased or up to
20-fold. HELLP syndrome may be classified according
to the degree of thrombocytopenia: HELLP 1 (≤ 50 ×
109 /L), HELLP 2 (between 50 × 109 and 100 × 109 /L)
and HELLP 3 (between 100 × 109 and 150 × 109 /L).
Serious maternal complications include DIC, placental abruption, acute renal failure, pulmonary
edema, and hepatic failure, occasionally requiring liver
transplantation. Hepatic rupture is a further rare,
acute, life-threatening complication.

Pre-eclampsia (PET)
This is classically defined as the triad of hypertension,
proteinuria, and edema, but is best thought of as a multisystem disorder resulting from endothelial damage.
It is a leading cause of maternal and neonatal morbidity and mortality, affecting 5%–10% of all pregnancies. It is more common in primigravid women. It
rarely occurs before 24 weeks of gestation and the incidence rises as pregnancy advances, being most common in the third trimester. Liver involvement is common although rarely severe and is the most common
cause of hepatic tenderness and liver dysfunction in
pregnancy. It is an indicator for delivery because of
the increased risk of severe eclampsia, hepatic rupture,
DIC, and necrosis. The high peri-natal morbidity and
mortality are partly due to the association with placental insufficiency and IUGR, but partly due to premature delivery for maternal indications. Severe PET is
complicated in 2%–12% of cases by HELLP syndrome,
consistent with the idea that they lie on a spectrum of a
single disorder. Renal impairment, eclampsia (convulsions), and abnormalities of the coagulation system are
further complications.

Hemolytic uremic syndrome (HUS)
D+ (diarrhea positive) HUS is typically preceded by
an illness with a verotoxin-producing bacteria, usually
E.coli 0157:H7. Atypical, D– (diarrhea negative) HUS,
is rare, with an incidence of 1/25 000 pregnancies, and
in nearly all documented cases associated with pregnancy, occurs postpartum. Atypical HUS (aHUS) may
be familial and has a poorer prognosis, with a mortality of 25% acutely and 50% requiring chronic renal

Chapter 18. Thrombotic thrombocytopenic purpura and other microangiopathies

therapy. Like all TMAs, it is a disease of microvascular
endothelial activation, cell injury, and thrombosis, but
associated with complement deregulation, leading to
an increase in activity in the alternative pathway. Mutations within the complement regulatory proteins and
activating components are found.
Typically, the presentation in HUS is of MAHA,
thrombocytopenia, and renal impairment. The primary pathology is in the renal arterioles and interlobular arteries, with widespread endothelial cell swelling,
leading to exposure of the underlying basement membrane. The vessel lumens are occluded by red cells and
platelet fibrin thrombi. The pre-glomerular pathology distinguishes it from D+HUS and TTP. There is
consequently excess complement activation particularly along glomeruli, arteriolar endothelium, and
basement membranes. More than 50% of cases result
from mutations in complement genes controlling the
alternative complement pathway. Mutations may affect
complement regulatory genes, such as Factor H, I or
MCP, or complement activating genes, Factor B (CBF),
or C3 (C3). Single nucleotide polymorphisms and antibodies, such as to Factor H, have also been found
to play a role. Factor H mutations, mostly heterozygote, account for 15%–30% of all cases of aHUS.8 MCP
mutations account for 10%–13% of aHUS patients, the
majority being heterozygote, with approximately 25%
homozygous/compound heterozygote.

Treatment
This is primarily supportive, including red cell transfusion, blood pressure control, and renal dialysis. The
role of plasma therapy remains undetermined, but has
been successful in some cases.

Exacerbation of systemic lupus
erythematosis (SLE)
SLE is an autoimmune disease, the active phase of
which may be associated with thrombocytopenia,
hemolytic anemia, pancytopenia, and an increase in
double-stranded DNA. The disorder is multisystem
and, typically, there are associated skin and joint
symptoms. Serum complement levels may be normal or decreased. An acute exarcerbation occurs in
25%–30% of women during pregnancy, but it may
occur for the first time during pregnancy9 or in
the postpartum period. An acute episode of lupus
nephritis, associated with hypertension and proteinuria, may be difficult to differentiate from HELLP or
pre-eclampsia.
Antiphospholipid antibodies (aPL) may be present
in 30%–49% of women with lupus and further increase
the risk of thrombotic events, the risk of tissue
ischemia and TMA. Thrombocytopenia is present in
a minority.

Disseminated intravascular
coagulation (DIC)
In pregnancy, DIC must not be forgotten as a cause
of MAHA with an abnormal clotting screen. Usually,
there is an underlying precipitating cause that must be
treated and it can be a complication of any of the above
TMAs in severe cases. Treatment of DIC requires
platelet transfusions to maintain a count ⬎50 × 109 /L,
fresh frozen plasma, and cryoprecipitate, depending on the level of abnormality of the coagulation
parameters.

225

Section 6. Microangiopathies

References
1. Sadler JE. Von Willebrand factor, ADAMTS13, and
thrombotic thrombocytopenic purpura. Blood 2008;
112: 11–18.
2. Vesely SK, George JN, Lammle B et al. ADAMTS13
activity in thrombotic thrombocytopenic
purpura–hemolytic uremic syndrome: relation to
presenting features and clinical outcomes in a
prospective cohort of 142 patients. Blood 2003; 102:
60–68.
3. Sanchez-Luceros A, Farias CE, Amaral MM et al. von
Willebrand factor-cleaving protease (ADAMTS13)
activity in normal non-pregnant women, pregnant and
post-delivery women. Thrombosis and Haemostasis
2004; 92: 1320–1326.
4. Scully M, Starke R, Lee R et al. Successful management
of pregnancy in women with a history of thrombotic
thrombocytopaenic purpura. Blood Coagulation and
Fibrinolysis 2006; 17: 459–463.
5. Vesely SK, Li X, McMinn JR, Terrell DR, George JN.
Pregnancy outcomes after recovery from thrombotic
thrombocytopenic purpura–hemolytic uremic
syndrome. Transfusion 2004; 44: 1149–1158.
6. Riely CA. Acute fatty liver of pregnancy. Seminars
Liver Disease 1987; 7: 47–54.
7. Rath W, Faridi A, Dudenhausen JW. HELLP
syndrome. Journal of Perinatal Medicine 2000; 28:
249–260.
8. Caprioli J, Noris M, Brioschi S et al. Genetics of HUS:
the impact of MCP, CFH, and IF mutations on clinical
presentation, response to treatment, and outcome.
Blood 2006; 108: 1267–1279.
9. Cortes-Hernandez J, Ordi-Ros J, Paredes F et al.
Clinical predictors of fetal and maternal outcome in
systemic lupus erythematosus: a prospective study of
103 pregnancies. Rheumatology (Oxford). 2002; 41:
643–650.

226

10. Ezra Y, Rose M, Eldor A. Therapy and prevention
of thrombotic thrombocytopenic purpura during
pregnancy: a clinical study of 16 pregnancies.
American Journal of Hematology 1996; 51:
1–6.
11. Ducloy-Bouthors AS, Caron C, Subtil D et al.
Thrombotic thrombocytopenic purpura: medical and
biological monitoring of six pregnancies. European
Journal of Obstetrics and Gynecology Reproduction
Biology 2003; 111: 146–152.
12. Shamseddine A, Chehal A, Usta I et al. Thrombotic
thrombocytopenic purpura and pregnancy: report of
four cases and literature review. Journal of Clinical
Apheresis, 2004; 19: 5–10.
13. Castellá M, Pujol M, Juliá A et al. Thrombotic
thrombocytopenic purpura and pregnancy: a
review of ten cases. Vox Sanguinis 2004; 87: 287–
290.
14. Ridolfi RL, Bell WR. Thrombotic thrombocytopenic
purpura. Report of 25 cases and review of the
literature. Medicine (Baltimore) 1981; 60: 413–
428.
15. Bell WR, Braine HG, Ness PM, Kickler TS. Improved
survival in thrombotic thrombocytopenic
purpura-hemolytic uremic syndrome. Clinical
experience in 108 patients. New England Journal of
Medicine 1991; 325: 398–403.
16. Thompson CE, Damon LE, Ries CA, Linker CA.
Thrombotic microangiopathies in the 1980s: clinical
features, response to treatment, and the impact of the
human immunodeficiency virus epidemic. Blood 1992;
80: 1890–1895.
17. Hayward CP, Sutton DM, Carter WH, Jr. et al.
Treatment outcomes in patients with adult thrombotic
thrombocytopenic purpura-hemolytic uremic
syndrome. Archives in Internal Medicine 1994; 154:
982–987.

Section

Malignant conditions

Section 7
Chapter

Malignant conditions

Myeloproliferative disorders
Claire Harrison and Susan E. Robinson

Introduction

Previous reports of MPD in pregnancy

The myeloproliferative disorders (MPDs) encompass
chronic myelogenous leukemia (CML), polycythemia
vera (PV), myelofibrosis (PMF), primary thrombocythemia (PT also known as essential thrombocythemia or ET), rarer entities such as chronic neutrophilic leukemia, chronic eosinophilic leukemia,
chronic myeloproliferative disease unclassifiable, and
the mast cell diseases. This chapter will concentrate
upon the management of the more common classical Philadelphia negative MPDs; PT, PV, and PMF in
pregnancy.

A recent meta-analysis reported the outcome of 461
pregnancies in women diagnosed with PT.1 The mean
age was 29 years and the mean platelet count at the
beginning of pregnancy was 1000 × 109 /L declining to
599 × 109 /L in the second trimester. The live birth
rate was 50%–70%, first trimester loss occurred in
25%–40%, and late pregnancy losses in 10%. Rates of
placental abruption (3.6%) and intrauterine growth
restriction (IUGR) (4.5%) were higher than in the general population. Postpartum thrombotic episodes were
reported in 5.2% of pregnancies and pre/postpartum
hemorrhage in 5.2%. A summary of 208 historical
cases of PT collated from case series that included
greater than six pregnancies produced comparable
data (presented in Table 19.1). The literature for pregnancies affected by PV is sparse; pregnancy outcome
in a case series of 18 pregnancies in PV combined with
20 historical reports was concordant with the pregnancy outcomes in PT (and is summarized in Table
19.2).2 In PV first trimester loss was the most frequent
complication (21%), followed by late pregnancy loss
(18%), IUGR (15%) and premature delivery (13%),
which included three neonatal deaths resulting in a
50% survival rate. Maternal morbidity was also significant including three thromboses, one large postpartum hemorrhage, four cases of pre-eclampsia and
one maternal death associated with evidence of a deep
vein thrombosis, pulmonary emboli, sagittal sinus
thrombosis and disseminated intravascular coagulation. Lastly, PMF is the least prevalent MPD in women
of child-bearing age. A report of four pregnancies in
PMF combined with four historical cases suggested a
50% risk of fetal loss; however, no maternal complications of thrombosis or disease progression were noted
but the numbers are probably too small to draw any
firm conclusions (summarized in Table 19.3).3

Epidemiology
The incidence of the classical Philadelphia negative MPDs combined is approximately 6/100 000–
9/100 000, with a peak in frequency between 50 and 70
years of age; they are less frequent in women of reproductive age.
Thrombosis and hemorrhage are a major cause of
morbidity in MPD patients; progression to myelofibrosis or an acute leukemia occur less frequently. Historical case reports of pregnancy in MPDs have suggested significant maternal morbidity and poor fetal
outcome. An increase in awareness of MPDs, advanced
maternal age, and automation of blood counts to
include a platelet count has led to an increase in the
diagnoses of MPDs in women of a reproductive age.
Hence issues concerning the management of these disorders in pregnancy are a real clinical challenge to
hematologists and obstetricians that is compounded
by a lack of clinical data and evidence-based guidance.
This chapter provides a summary of the epidemiology,
pathogenesis, and diagnosis of the MPDs in pregnancy
and a management strategy developed from current
experience attained in a tertiary referral center.

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

229

108

Total

1 VTE

1 VTE
2 TIA

Detail not
available

208

43 (2 TOP, 1
Ectopic)

Detail not
available

1 CVA
1 VTE

30 (1 Ectopic) Detail not
available

15 (1 TOP)

Detail not
available

1 Epistaxis

Detail not
available

1 Eclampsia 2
Pre-eclampsia 1
Vaginal bleed

3 Vaginal bleeds

Detail not available

1 TIA 2 Acquired vWD
3 Vaginal bleeds 2
Epistaxis

1 PE

3VTE

2 VTE 2TIA 1
Hemorrhage

Detail not available

1 Phlebitis 1 Leg ulcer
1 PPH

Detail not available

FTD: full term delivery; IUGR: intrauterine growth restriction; TOP: elective termination of pregnancy
Adapted from refs. 4 and 5.

Number Number of Previous
Previous
Maternal
Reference of pts
Pregnancies thrombosis hemorrhage outcome

Table 19.1 Summary of reported pregnancies affected by PT

124
(60%)

85 (41%)

26 (1 15
Twin)

60 (29%)

Detail not
available

20 (10%) 5 (2%)

Live
birth Pregnancy Loss
Loss
total loss total <12/40 >12/40 IUGR

7 (3%)

22 (11%)

102
(49%)

Live birth
premature Live
Placental delivery
birth
abruption <37/40
FTD

1
1

Total

18 (1 twin)

3
2

No
No

1 CVA 1
thrombo
phlebitis

None

Yes (1 patient) No

Yes, CVA

Superficial
thrombo
phlebitis

No
No

Yes

No
No

1 yes

Varied:
No
Venesection,
Aspirin,
Interferon
LMWH,
vitamin C+E

Hydroxyurea No
9/40 then nil

Heparin 3/52 No
postpartum

Nil

Aspirin,
heparin∗∗
venesection

None

Nil
Nil

Aspirin +
No
dipyrimadole

1
2

1 death 4
PET 2 PE 1
PPH

2
0

0
0

2 (5 + 7
months) PET

2 (24/40 and
28/40)∗∗∗

1 (35/40) PET 1

0
0

1 TOP 0

0
0

1 (34/40,
IUGR), 1
(36/40), 1
26/40) (NND)

1 (7 months,
PET), 1
(8 months)

1 (32/40)

0
0

1
2
PET

Live birth
premature Live
Pregnancy
Stillbirth
Placental delivery
birth
loss total FTM (gestation) IUGR abruption <37/40
FTD

22
16
3 NNDs

Alive PET in 1 11

Alive

Alive, PE
24/7
postpartum

Alive PPH

Alive PET

Alive PE
postpartum

Alive PET

Alive
Alive PET

Death∗

Live
birth
total

FTM: first trimester miscarriage, IUGR: intrauterine growth restriction, FTD: full term delivery, NND: Neonatal death, TOP: termination of pregnancy, PE: pulmonary embolism, PET: pre-eclampsia,
PPH: postpartum hemorrhage, LMWH: low molecular weight heparin
∗ the patient died with evidence of deep vein thrombosis, pulmonary embolism, sagittal Sinus thrombosis and disseminated intravascular coagulation
∗∗ postpartum heparin after 2nd pregnancy, LMWH throughout third pregnancy, aspirin throughout both pregnancies
∗∗∗ multiple placental infarcts in first and abnormal uterine artery Doppler waveforms and Severe IUGR in third pregnancy
Adapted from ref. 2.

Treatment
Number No. of
Previous
Previous
during
High Maternal
Author of Pts Pregnancies thrombosis hemorrhage pregnancy risk outcome

Table 19.2 Summary of literature regarding pregnancy in PV

3 (nb. 2
preceeding
PMF
diagnoses)
1

Digital
ischemia
Digital
ischemia
Digital
ischemia

Adapted from ref. 3.
IUGR: Intrauterine growth restriction, FTND: full term normal delivery.

Total

Placental
infarctions
Placental
infarctions

Previous
Previous
Author Patient Pregnancy thrombosis hemorrhage

Table 19.3 Summary of literature regarding PMF in pregnancy

Aspirin

Aspirin, LMWH

Aspirin

Interferon ␣

Aspirin

None

Supportive

Treatment
during
pregnancy

None

Supportive

Treatment
Pre
pregnancy

Postpartum
hemorrhage
No
complications
No
complications

Disseminated
TB

No
complications
No
complications
No
complications

No
complications

Maternal
outcome

24/40 cardiac 0
malformation
0
0

30 (placental
infarctions)
27 (placental
infarctions)
0

1 FTND

1 Elective
delivery at 34
wks due to
IUGR birth
weight 2000 g

1 Elective
induction at 36
wks
0

Live birth
First
premature
trimester
Stillbirth
Placental delivery
miscarriage (gestation) IUGR abruption <37 wks

Chapter 19. Myeloproliferative disorders

Historical reports of pregnancy in MPD are likely
to be subject to selection bias, favoring cases associated with a poor outcome; prospective surveillance of
pregnancies in MPDs and the development of structured evidence-based guidance would be of benefit in
this field.

Pathogenesis
MPDs result from the transformation of a hematopoietic progenitor cell and are characterized by overproduction of mature blood cells. The proliferation
of one single cell type predominates, resulting in
increased numbers of granulocytes (CML), erythrocytes (PV), platelets (PT), or fibroblasts (PMF). A
single, acquired point mutation in the Janus kinase
2 (JAK2) gene occurring in the majority of patients
with PV and almost half of those with ET and PMF
was discovered in 2005. The mutation is a guanine to
thymidine substitution that substitutes phenylalanine
for valine at position 617 (V617F) of the JAK2 protein. This residue is located within the JH2 pseudokinase domain, which negatively regulates the JH1 catalytically active kinase domain. The wild-type JAK 2
protein binds to multiple cytokine receptors including
the erythropoietin, thrombopoietin, and granulocytecolony stimulating factor receptors that are essential for hemopoietic stem cell biology and differentiation. The JAK2 protein with the V617F mutation
enables constitutive, cytokine independent activation
of the JAK-STAT, PI3K, and MAPK signal transduction pathways at various stages of development and in
various lineages of hemopoietic cells (Fig. 19.1). Four
further mutations affecting the JAK2 exon 12 that were
recently identified define a distinctive myeloproliferative syndrome that affects patients who previously
received a diagnosis of PV or idiopathic erythrocytosis. Finally, two further mutations in the thrombopoietin receptor MPL W515L/K have been described in
patients with PMF(5%) and PT(1%). The reported
mutations have been shown to produce an MPD-like
phenotype in various murine models.

Pathogenesis of placental infarction
and thrombosis
Thrombosis is consistently identified as the leading
cause of maternal mortality in apparently healthy normal pregnancies. Thrombotic occlusion of the placental circulation may be a late manifestation of placental

dysfunction or an independent mechanism of pregnancy morbidity. In women with PT, placental thrombosis was reported in pregnancies which resulted in
late fetal loss, pre-term delivery, and IUGR. IUGR,
which is associated with utero-placental dysfunction,
is known to occur in other acquired and inherited
causes of thrombophilia.
Multiple factors are likely to contribute to the
pathogenesis of thrombosis in MPDs, including the
degree of thrombocytosis, leukocytosis, raised hematocrit, activation of platelets and leukocytes, the formation of platelet leukocyte aggregates, circulating
prothrombotic and endothelial factors, and their interactions.6 It is of interest to our understanding of how
the MPD phenotype contributes to placental dysfunction that independently, studies of MPDs and preeclampsia both report increased platelet activation,
platelet monocyte aggregate formation, and microparticle formation. It is currently unclear whether the
presence of the JAK2 V617F and MPL W515L/K mutations increase the risk of thrombosis in MPDs and,
if so, by what mechanism. However, recently a study
of 103 pregnancies in 62 women with PT, identified
the JAK2 V617F mutation as an independent predictor of pregnancy complications (P = 0.01). A total of
17 (71%) of 24 women carrying the JAK2 mutation had
complications at first pregnancy. The study concluded
that women with PT with the JAK2 V617F mutation
had a twofold higher risk of developing complications
than those without the mutation. A recent matched
case control study of unexplained first pregnancy loss
involving 32 683 patients is in support of these findings. In 3496 pairs of women the JAK2 V617F mutation occurred more frequently in patients with pregnancy loss (1.06%) than in control subjects (0.20%).

Diagnosis
The diagnosis of MPDs in pregnancy requires an
increased awareness of these disorders occurring in
pregnant women. Suspicion may be secondary to an
abnormal full blood count, a thrombotic, or hemorrhagic event and should prompt referral to a hematologist. In view of pregnancy morbidities and the
likelihood of improved outcome with intervention,
these women benefit from a diagnosis being made
pre-conceptually, during pregnancy or the postpartum period. The following section details local policy
including adaptation for diagnostic investigations in
pregnancy.

233

Section 7. Malignant conditions

Fig. 19.1 Jak-Stat pathways.

Extracellular domain
Wild-type JAK2 without

Wild-type JAK2 with

JAK2 with V617F mutation

erythropoeitin

without erythropoeitin

EPO

JAK

V617F

P
P

STAT

STAT dimer

Intracellular domain
P

RAS-MAPK
P13K

STAT dimer
P

RAS-MAPK

P13K

Nucleus

Enhanced DNA transcription

Primary thrombocythemia

234

There is no diagnostic hallmark for this condition.
The diagnosis is made by excluding other MPDs
and a reactive or secondary cause of a thrombocytosis. Causes of a reactive thrombocytosis include
iron deficiency anemia, chronic inflammation (e.g.
rheumatoid arthritis, or inflammatory bowel disease),
splenectomy, acute hemorrhage, and malignant disease. Where conditions co-exist that may cause a
reactive thrombocytosis, this may make the diagnosis more difficult. In pregnancy the platelet count may
fall especially during the second and third trimesters,
thereby masking the diagnosis.
Historically, the diagnostic criteria for PT were
those of the polycythemia vera study group; 40 years
on, continual development of the diagnostic criteria

for MPDs set the stage for the World Health Organization diagnostic criteria 2001, modified in 2007.
The revised WHO criteria require characteristic bone
marrow morphology (this is a controversial aspect not
universally accepted), a platelet threshold of 450 ×
109 /L and molecular analysis for the JAK 2 V617F
mutation and other clonal markers. Investigations
should include a blood count, blood film, hematinics, renal, and liver profile, CRP, ANA and RhF,
genetic screen for the JAK2 V617F, MPL W515L/K,
and bcr/abl mutation and abdominal ultrasound scan.

Polycythemia vera
An erythrocytosis requiring investigation is defined
as a packed cell volume (PCV) greater than 0.48 in

Chapter 19. Myeloproliferative disorders

non-pregnant women; in pregnancy this threshold has
not formally been defined. To determine whether there
is an absolute increase in PCV or an erythrocytosis
or an apparent increase due to reduced plasma volume has traditionally required a red cell mass study,
which would be contraindicated in pregnancy. Red
cell mass scans have been largely superseded by testing for the presence of the JAK2 V617F mutation,
which indicates the presence of the majority of PV
cases. The JAK2 V617F mutation negative erythrocytosis cases may still be PV without a genetic marker or
with a JAK2 exon 12 mutation; alternatives include a
pseudo/apparent, primary congenital, secondary congenital or acquired, or an idiopathic erythrocytosis, all
of which require definition.
The current British Committee for Standards in
Haematology guideline for investigation and management of erythrocytosis7 suggest a staged approach
to investigation as the differential diagnosis is broad
and secondary causes must be excluded. This is followed by investigations to confirm or refute a diagnosis of a JAK2 V617F positive PV. The majority of
patients (excluding borderline erythrocytosis) and all
ex- and current smokers will require a chest X-ray.
This should be avoided in pregnancy unless there is
a strong suspicion of a causative lung pathology, in
which case appropriate screening should be used. Urinalysis is a simple effective screen for renal disease,
which should be performed in all patients at the initial visit. Patients may present with co-morbidity, thus
regardless of a diagnosis of PV a review of secondary
causes is pertinent. Additional investigation of possible secondary causes will vary according to symptoms
or signs present.

Myelofibrosis
Myelofibrosis is very rare indeed in women of childbearing age. To achieve this diagnosis it is necessary
to exclude other MPDs (PV, PT and CML) as well as
disorders in which marrow fibrosis can develop as a
secondary feature such as metastatic carcinoma, lymphoma, irradiation, TB, and leishmaniasis. The following features are generally necessary to confirm a
diagnosis of MF: splenomegaly, increased bone marrow fibrosis (coarse reticulin fibers arranged in parallel
in trephine biopsy), a leukoerythroblastic blood film
(immature red cells and myeloid precursors with tear
drop-shaped red cells) and the exclusion of secondary
causes of myelofibrosis (see above). In all suspected

cases of MF a bone marrow aspirate and trephine are
required.

Treatment options
Women of reproductive age with a diagnosis of MPD
should receive information and assurance regarding
management and outcome of future pregnancies. If
fertility issues arise, optimal disease management may
need to be re-addressed prior to a timely referral
for standard fertility investigation. A risk assessment
according to disease status, concomitant illnesses, and
prior obstetric history forms the basis for a discussion
of the risks and benefits of therapeutic options in pregnancy. According to perceived risk, the therapeutic
options include aspirin, heparin, venesection, cytoreductive agents, and thromboembolic deterrent stockings. From pre-conceptual planning to the postpartum
period, access to joint care from an obstetrician with
experience of high risk pregnancies and a hematologist in a multidisciplinary setting is paramount.
The pre-conception to postpartum management
plan should include:
r informed multidisciplinary care and education;
r risk assessment and discussion of therapeutic
options and implementation of an appropriate
management plan;
r additional monitoring during pregnancy;
r further optimization of disease control, if fertility
is an issue, prior to timely referral for standard
investigation;
r A comprehensive delivery and postpartum plan.
This approach enables optimal disease control with
the aim of increasing the possibility of conception,
implantation, and maintenance of placental function.
Thus reducing complications secondary to placental
dysfunction, such as IUGR and pregnancy losses. An
emphasis upon the prevention of thrombosis and hemorrhage and management of events pre- and postpartum is also required.
Two key treatment aims, in high risk non-pregnant
patients with MPDs, are to attain a platelet count less
than 400 × 109 /L and a PCV less than 0.45 and possibly
less than 0.42 in women who remain symptomatic. In
pregnancy an increase in the plasma volume reduces
both the platelet count and PCV, which is likely to
further alter blood cell rheology. Interestingly, it has
been suggested that the decrease in platelet count is
greater than that expected in a normal pregnancy. One

235

Section 7. Malignant conditions

theory is whether the placenta produces an interferonlike substance. An understanding of this physiological
dilutional effect and the brisk return to pre-pregnancy
levels in the postpartum period in MPD pregnancies, are important when considering optimal monitoring intervals and suitable treatment targets. The target PCV and platelet count in pregnant women with
MPDs are ongoing debates, but appropriate targets are
probably a platelet count of less than 400 × 109 /L and
a hematocrit certainly less than 0.45 and probably in
the mid gestation appropriate range.

Aspirin
Low dose aspirin is considered safe in pregnancy in
accordance with the Collaborative Low dose Aspirin
Study in Pregnancy (CLASP), although its use for
thromboprophylaxis in MPDs has never been assessed
by a controlled trial. The European Collaborative Lowdose Aspirin in Polycythaemia Vera (ECLAP) study
supports the use of low dose aspirin in non-pregnant
patients with PV. Aspirin has been the most widely
used therapy (in at least half of published pregnancies) for pregnancies affected by PT. Although the evidence is both retrospective and based on small numbers, the use of low dose aspirin in pregnancy in myeloproliferative disorders seems advantageous, and a low
risk strategy for the pregnancy. A recent update of
the largest case series of pregnant women with PT to
date provided analysis of pregnancy outcomes treated
with aspirin vs. those managed by observation alone.
There was no evidence that therapy with aspirin positively influenced pregnancy outcome in women with
PT; however, interpretation should be cautious as with
all retrospective reports.

Low molecular weight heparin

236

A study of women with the Factor V Leiden, prothrombin gene mutation or protein S deficiency, and
one fetal loss demonstrated aspirin to be inferior to
low molecular weight heparin (LMWH) in terms of
live birth rate and birth weight. The successful use of
LMWH in other pregnancies at high risk of thrombosis and in reducing fetal morbidity has drawn attention
to the possibility of its use, in addition to aspirin, during pregnancy in women with MPDs with a previous
thrombosis or pregnancy-related events.
Our regime for LMWH use, if necessary, in
pregnant patients with MPDs is:

r ante-natal dose of LMWH, e.g. 40 mg enoxaparin
daily or 5000 IU dalteparin daily, increasing to 12
hourly from 16 weeks onwards;
r at low body weight (e.g. ⬍ 50 kg), lower doses of
LMWH may be required, e.g. 20 mg enoxaparin
daily or 2500 IU dalteparin daily;
r In obese patients (e.g. BMI ⬎ 30 in early
pregnancy), higher doses of LMWH may be
required, e.g. 40 mg enoxaparin 12 hourly or 5000
IU dalteparin 12 hourly;
r postpartum dose of LMWH, e.g. 40 mg
enoxaparin daily or 5000 IU dalteparin daily for 6
weeks;
r if uterine artery Dopplers are repeatedly
abnormal, increase to a therapeutic dose of
LMWH.
In women with previous arterial thrombotic events
or in those with recurrent thrombosis on warfarin
prior to pregnancy, therapeutic doses of subcutaneous
LMWH may be required. Some patients may require
monitoring of anti-Xa levels.

Graduated elastic compression
stockings (GECS)
GECS may be used ante-natally and during the postpartum period. There are no trials to support such
practice, but the British Society for Haematology
(BSH) guidelines suggest that all women with previous venous thrombosis or a thrombophilia should be
encouraged to wear GECS throughout their pregnancy
and for 6–12 weeks after delivery.

Cytoreductive therapy
Cytoreduction is used where necessary to reduce the
platelet count or a raised PCV that is resistant to venesection, but these agents should preferably be avoided
in pregnancy, particularly in the first trimester. None
of the cytoreductive drugs mentioned in this chapter have a product licence for use in pregnancy. The
expected natural fall of the platelet count and hematocrit during pregnancy may reduce the need for
cytoreduction or venesection. However, in high risk
situations where cytoreduction is deemed necessary
(see below), interferon ␣ (IFN-␣) is the drug of choice.
There are no reports of teratogenic effects in animals
or adverse effects in the admittedly small numbers of
pregnancies exposed to this drug. However, some evidence suggests that IFN-␣ may decrease fertility and

Chapter 19. Myeloproliferative disorders

Table 19.4 95% ranges for hematological variables during
pregnancy

Table 19.5 High risk MPD pregnancy criteria
r

Gestation

First
Second
Third
trimester trimester trimester

Hb (g/dL)

11–14.3

10–13.7

9.8–13.7

PCV (l/L)

0.31–0.41

0.30–0.38

0.28–0.39

Platelet count (109 /L)

174–391

171–409

155–429

Adapted from ref. 8.

so it is best avoided in women with difficulty conceiving. In relation to hydroxyurea, the outcomes of small
numbers of pregnancies have been published and
these are mainly without fetal complications, although
one stillbirth and one malformed infant have been
reported after exposure to hydroxyurea. Teratogenicity in animals has also been reported. Thus the use of
hydroxyurea is probably contraindicated at the time
of conception and during pregnancy. The use of anagrelide in pregnancy is similarly not recommended
because of insufficient documentation of its use in this
situation and because of the possibility of thrombocytopenia in the fetus.

Venesection
Although the natural fall in the hematocrit or PCV
in pregnancy may obviate the need for venesection, it
is an option in resistant cases. If venesection fails to
control the PCV, then cytoreduction should be considered. The target PCV for a pregnant woman has yet to
be determined. A reasonable target PCV would be the
middle of the gestation appropriate range (Table 19.4).
There is currently no evidence for maintaining it lower
than this in pregnancy, although this has been an area
of controversy.

Recommendations for management
of MPDs in pregnancy
An overview of the small groups of MPD patients and
individual cases described in the literature does not
enable confident management guidelines to be drawn
up. The following are recommendations based on current knowledge of this and other thrombophilic states
and upon personal experience in a tertiary referral
unit. Good communication between consultant obstetrician and hematologist is essential.

r
r
r

Previous venous or arterial thrombosis in mother (whether
pregnant or not)
Previous hemorrhage attributed to MPD (whether pregnant
or not)∗
Previous pregnancy complication that may have been
caused by a MPD:
– Three pregnancy losses ⬍ 10 weeks
– One or more pregnancy losses ⬎24 weeks
– Intra-uterine growth restriction or other evidence of
placental dysfunction
– Intra-uterine death or stillbirth (with no obvious other
cause)
– A significant ante- or postpartum hemorrhage
(requiring red cell transfusion)
– Severe pre-eclampsia (necessitating preterm delivery
⬍ 37 weeks)
Platelet count rising to ⬎1500 × 109 /L prior to pregnancy
or during pregnancy∗
Diabetes mellitus or hypertension requiring treatment
JAK2 V6127F mutation, the status of this as a risk marker is
currently unclear

Note ∗ these criteria would be an indication for cytoreductive
treatment but not LMWH.

Pre-conceptual meeting
The patient should ideally have a pre-conceptual meeting with both an obstetrician and a hematologist to
discuss a plan of management for a future pregnancy,
including the necessity for cytoreductive therapy. Ideally, this should be written out and copied to the
patient.

Control of platelet count and haematocrit
If a patient is already taking hydroxyurea or anagrelide,
this should be gradually withdrawn before conception,
followed ideally by a wash-out period of 3 months for
hydroxyurea following the last dose. The platelet count
and PCV must be closely monitored thereafter. Careful venesection should be commenced if the hematocrit rises above the gestational appropriate range.
Cytoreduction with IFN-␣ may be necessary in cases
with a raised PCV resistant to venesection or persistent
thrombocytosis. Most patients with a clear indication
for cytoreductive therapy pre-pregnancy will require
cytoreduction during pregnancy.
Cytoreduction with IFN-␣ may also be necessary
if any of the factors are present, or if they develop
in the index pregnancy, which in our experience
defines a high risk MPD pregnancy (Table 19.5). Treatment should be guided by monitoring the full blood
count and by maintaining the platelet count less than

237

Section 7. Malignant conditions

400 × 109 /L and the PCV in the appropriate gestational range.

Management of thrombotic risk
Assessment of need for antithrombotic medication
The assessment of the need for antithrombotic medication should preferably be done in the pre-conceptual
meeting, but on-going individual risk assessment
should occur and may warrant commencing or
increasing thromboprophylaxis.

Aspirin
In the absence of clear contraindications, i.e. asthma,
history of peptic ulceration, or current hemorrhage,
all patients should be on aspirin (initially 75 mg once
daily) throughout the pregnancy and for at least 6
weeks after delivery. In the event of a platelet count in
excess of 1000 × 109 /L, acquired von Willebrand disease should be excluded prior to commencing aspirin.

Low molecular weight heparin
Consider the use of subcutaneous LMWH during
pregnancy in addition to cytoreduction in patients
with any of the high risk MPD pregnancy factors
listed in Table 19.5 with the exception of hemorrhage
and extreme thrombocytosis. LMWH is an option to
be introduced in patients with persistently abnormal
uterine artery Dopplers. Once adequate hemostasis
has been achieved postpartum, all women should be
offered 6 weeks of LMWH thromboprophylaxis in the
absence of a prior history of a significant hemorrhage.
Caution should be applied to cases where women have
a previous history of a significant hemorrhage with
a platelet count ⬍1000 × 109 /L and no other obvious cause except for platelet dysfunction secondary to
a MPD.

Graduated elastic compression stockings
Consider the use of GECS as a supplementary therapy
throughout pregnancy and for 6–12 weeks after delivery in accordance with current BSH guidelines.

Maternal and fetal monitoring
238

Fetal monitoring
The local protocol for fetal monitoring includes scans
at 12 and 20 weeks. If the uterine artery Doppler ultrasound at 20 weeks is abnormal, it should be repeated
at 24 weeks; if abnormal then, consideration should
be given to increasing or escalating therapy. Regular
growth scans should also be performed.
Uterine artery doppler scanning at 20 and 24 weeks
will reveal whether the woman has bilateral notching. The persistence of bilateral notching indicates
increased resistance to flow and possible placental dysfunction. The presence of persistent bilateral notching
at 24 weeks should prompt commencing or increasing
LMWH to 40 mg s/c twice daily, and further treatment
escalation including interferon ␣ and the possibility of
an early delivery may need to be considered.

Delivery
Prior to labor or Cesarean section, it is important to
discuss the implications of the use of thromboprophylaxis for epidural or spinal anesthesia with the woman
and obstetric anesthetist following locally agreed protocols. If a woman develops a hemorrhagic problem
while on LMWH, the treatment should be stopped
and hematological advice sought; a platelet transfusion may be useful in patients with MPDs. It should be
remembered that excess blood loss and blood transfusion are risk factors for VTE, so thromboprophylaxis
should be begun or reinstituted as soon as the immediate risk of hemorrhage is reduced. The third stage of
labor should be managed actively.

Postpartum thromboprophylaxis
The time of greatest risk for VTE associated with
pregnancy is the immediate puerperium period. The
prothrombotic pregnancy phenotype does not revert
back to normal until 6 weeks after delivery. All MPD
patients should receive 6 weeks of postpartum LMWH
thromboprophylaxis unless contraindicated. Aspirin
should also be continued for at least 6 weeks. As discussed above, where women have a history of hemorrhage, the addition of postpartum heparin should be
cautious and considered on an individual case basis.

Maternal monitoring

Postpartum assessment

Full blood count monitoring, blood pressure, and
urine testing should be performed 4 weekly until 24
weeks and thereafter at 2-weekly intervals.

The platelet count and PCV may rise dramatically
postpartum, but can usually be controlled with cytoreductive therapy or venesection. Cytoreductive therapy

Chapter 19. Myeloproliferative disorders

suitable immediately post-delivery if required include
hydroxyurea, IFN-␣ and anagrelide, the choice and
dose depending on previous experience in that patient.
Hydroxyurea, IFN-␣, and possibly anagrelide are
excreted in breast milk, so breast feeding is then contraindicated. In the current literature there is no evidence that pregnancy predisposes MPD patients to
acceleration of their disease to PMF or acute leukemia,
nor would this be anticipated. The most significant
risk is of thrombosis in the mother and adverse fetal
outcome. A frequent question asked by these patients
is what is the chance of their children being affected
by MPD. Until recently, it was believed that familial MPD was relatively rare and, whilst this is true
for large kindreds with many affected individuals, it
has become apparent that up to 8% of MPD patients
may have an affected relative, usually a cousin, aunt,
uncle, etc. Parent:child combinations are extremely
uncommon and routine testing of children is not
recommended.

Dilemmas
The following section includes a series of challenging
cases, which enable discussion of management options
accordingly in these women with complex pregnancies.

Case 1
A 37-year-old lady with a diagnosis of PT is referred
for a tertiary opinion regarding conception and
pregnancy management. The referral to a
hematologist and initial diagnosis of PT followed a
full blood count screen by her GP. The obstetric
history includes one full-term spontaneous vaginal
delivery 12 years previously. Following remarriage
3 years ago, she has undergone two spontaneous
miscarriages ⬍10 weeks. The current platelet count
is 1700 ×109 /L, current medication includes aspirin
75 mg daily.

In an attempt to establish when the patient developed PT prior to the official diagnosis, any previous
full blood counts could be reviewed. However, this is
unlikely to change the management in this case. The
diagnosis of PT and obstetric history are suggestive of
poor pregnancy outcome secondary to PT. The platelet
count enables us to stratify the patient as being at high
risk of a vaso-occlusive event and outside of pregnancy
would suggest benefit from commencing cytoreduct-

ive therapy. With the stated aim to conceive in the near
future, the appropriate cytoreductive agent would be
interferon ␣ as this could be continued throughout
pregnancy. A screen for an acquired von Willebrand’s
disease should be completed prior to continuation of
aspirin. The woman should be monitored in her local
clinic and the dose of interferon adjusted to maintain a platelet count ⬍400 × 109 /L. If a pregnancy
is confirmed, monitoring according to the treatment
algorhythm should commence and follow-up could be
shared between the local hospital and tertiary referral unit. The patient does not currently meet the criteria to commence LMWH ante-natally, although may
benefit from heparin and aspirin for 6 weeks postpartum. However, a discussion within the multidisciplinary pre-conceptual meeting in view of patients age
and prior obstetric history may conclude the addition
of LMWH from conception as suitable on an individual basis regardless that the history is of two not three
miscarriages ⬍10 weeks. Clearly, these complex cases
need to be managed upon an individual basis and management plans may need to encompass aspects which
are outside of general guidance. Close collaboration
between the local hematology and obstetric unit and
the tertiary center may enable delivery outside of the
tertiary center dependent upon the progress of the
individual pregnancy.

Case 2
A pregnant 26-year-old woman diagnosed with PV
following investigation of menorrhagia and
epistaxis 4 years previously is referred for an opinion
at 9 weeks’ gestation. There is no prior obstetric
history, and the current management is low dose
aspirin and venesection. Her current blood count
reveals Hb 15 g/dL, HCT 0.47, plt 567 ×109 /L.

The history of hemorrhage and or epistaxis may
be attributed to MPD in this case and, if so, the
woman would meet the criteria to consider interferon
in this pregnancy, but this is a relatively soft indication and the authors would not use this treatment in
this setting. However, this history should be examined carefully as hemorrhage is rare in these conditions and, even though the platelet count is not
markedly abnormal, it would be wise to screen for
von Willebrand’s disease. With no history of thrombotic events or pregnancy complications, there is no
indication for ante-natal heparin prophylaxis unless

239

Section 7. Malignant conditions

uterine artery dopplers subsequently suggest impaired
placental function; heparin should be given for 6 weeks
postpartum. Aspirin should be continued throughout
pregnancy. The current PCV is outside the appropriate
range for the first trimester and venesection should be
considered if tolerated.

Case 3
A 32-year-old woman diagnosed with PMF 3 years
ago following a recent hepatic vein thrombosis
attends clinic to discuss treatment options
regarding future pregnancies. She has one healthy
daughter of 5 years delivered by Cesarean section
following a trial of labor which failed to progress.
Current medication includes aspirin, warfarin, and
hydroxyurea. The question of future cord blood
stem cell storage is also raised. Her current blood
count in Hb 11 g/dL, PCV 0.35, plt 147, Wbc 7.6 ×
109 /L.

In the pre-conception planning, a management
plan regarding anticoagulation, cytoreductive therapy,
and review of concomitant liver disease is required.
The issues of stem cell storage needs to be addressed
regarding reasoning and practicalities. Clearly, the
woman may have a personal interest in cord stem
cell storage and this needs to be addressed preconceptually. This complex case would benefit from
follow-up and delivery in a tertiary referral center.
The aspirin should be continued throughout pregnancy; testing should be performed in the fortnight
following a possible conception in order to stop the
warfarin as early as possible and commence LMWH.
In view of the history of a hepatic vein thrombosis, a therapeutic dose of LMWH should be used
and switched back to warfarin postpartum. Three
months prior to conception the hydroxyurea needs
to be stopped and interferon commenced. Optimization of any concomitant liver pathology and portal
hypertension secondary to the previous hepatic vein

240

thrombosis is important. The obstetric history needs
to be reviewed in light of whether portal hypertension
and varices are present. The presence of varices may
require banding and additional medication, which
should be instigated and followed up by the gastroenterology team. The planned mode of delivery according to concomitant pathology needs to be addressed in
a multidisciplinary meeting.

Case 4
A pregnant 31-year-old female with diabetes is
referred at 10/40 with a platelet count of 759 ×
109 /L from the combined endocrine obstetric clinic
and is subsequently diagnosed with PT.

Although there is no previous obstetric history, a
diagnosis of diabetes suggests a high risk pregnancy.
Outside of pregnancy, the patient would be in the high
risk of vaso-occlusive event group secondary to the
diagnosis of diabetes. Low dose aspirin should be commenced and continued postpartum. A prophylactic
dose of LMWH once daily increased to twice daily at
16 weeks followed by 6 weeks’ postpartum prophylactic once daily LMWH should be considered. Interferon
would also be considered in this case.

Case 5
A 29-year-old pregnant woman with a diagnosis of
PT with no prior obstetric history or thrombotic
events attends clinic at 15/40.
Hb 12 g/dL, HCT 0.34, Plt 580 × 109 /L.

Low dose aspirin throughout pregnancy continued
indefinitely postpartum, combined with 6 weeks postpartum heparin prophylaxis would be appropriate in
this case. Additional treatment will depend upon how
the pregnancy progresses.

Chapter 19. Myeloproliferative disorders

References
1. Barbui T, Finazzi G. Myeloproliferative disease in
pregnancy and other management issues. Hematology
2006; 1: 246–252.
2. Robinson S, Bewley S, Hunt BJ, Radia DH, Harrison
CN. The management and outcome of 18 pregnancies
in women with polycythemia vera. Haematologica
2005; 90: 1477–1483.
3. Tupule S, Bewley S, Robinson SE et al. The
management and outcome of 4 pregnancies in women
with idiopathic myelofibrosis. British Journal of
Haematology 2008; 142: 480–482.
4. Griesshammer M, Struve S, Harrison CM. Essential
thrombocythemia/polycythemia vera and pregnancy:
the need for an observational study in Europe.
Seminars in Thrombosis and Hemostasis 2006; 32:
422–429.
5. Elliott MA, Tefferi A. Thrombocythaemia and
pregnancy. Best Practice Research Clinical
Haematology 2003; 16: 227–242.
6. Harrison CN. Platelets and thrombosis in
myeloproliferative diseases. Hematology 2005; 1:
409–415.
7. McMullin MF, Bareford D, Campbell P et al.
Guidelines for the diagnosis, investigation and
management of polycythaemia/erythrocytosis.
British Journal of Haematology 2005; 130: 174–
195.
8. Bain BJ. Blood Cells, 3rd edn. Oxford, UK: Blackwell
Science, 1995.
9. Bellucci S, Janvier M, Tobelem G et al. Essential
thrombocythemias. Clinical evolutionary and
biological data. Cancer 1986; 58: 2440–2447.
10. Beard J, Hillmen P, Anderson CC et al. Primary
thrombocythaemia in pregnancy. British Journal of
Haematology 1991; 77: 371–374.
11. Leone G, De S, V, D’Addosio A. [Essential
thrombocythemia pregnancy]: Haematologica 1991;
76: 365–367.
12. Pagliaro P, Arrigoni L, Muggiasca ML et al. Primary
thrombocythemia and prgnancy: treatment and
outcome in fifteen cases. American Journal of
Hematology 1996; 53: 6–10.
13. Randi ML, Rossi C, Fabris F, Girolami A. Essential
thrombocythemia in young adults: treatment and
outcome of 16 pregnancies. Journal of Internal
Medicine 1999; 246: 517–518.
14. Cincotta R, Higgins JR, Tippett C et al. Managament
of essential thrombocythaemia during pregnancy.
Australia and New Zealand Journal of Obstetrics and
Gynaecology 2000; 40: 33–37.

15. Bangerter M, Guthner C, Beneke H
et al. Pregnancy in essential thrombocythaemia:
treatment and outcome of 17 pregnancies.
European Journal of Haematology 2000; 65: 165–
169.
16. Wright CA, Tefferi A. A single institutional experience
with 43 pregnancies in essential thrombocythemia.
European Journal of Haematology 2001; 66: 152–
159.
17. Candoni A, Fanin R, Michelutti T et al. Pregnancy and
abortion in women with essential thrombocythemia.
American Journal of Hematology 2002; 69: 233–234.
18. Niittyvuopio R, Juvonen E, Kaaja R et al. Pregnancy in
essential thrombocythaemia: experience with 40
pregnancies. European Journal of Haematology 2004;
73: 434–436.
19. Crowley JP, Barcohana Y, Sturner WQ. Disseminated
intravascular coagulation following first trimester
abortion in polycythemia vera. Increased platelet
count apparently contributed to fatal outcome.
Rhode Island Medical Journal 1987; 70: 109–
112.
20. Centrone AL, Freda RN, McGowan L. Polycythemia
rubra vera in pregnancy. Obstetrics and Gynecology
1967; 30: 657–659.
21. Ruch WA, Klein RL. Polycythaemia vera and
pregnancy: report of a case. Obstetrics and Gynecology
1964; 23: 107–111.
22. Subtil D, Deruelle P, Trillot N, Jude B. Preclinical phase
of polycythemia vera in pregnancy. Obstetrics and
Gynecology 2001; 98: 945–947.
23. Hochman A, Stein JA. Polycythemia and pregnancy.
Report of a case. Obstetrics and Gynecology 1961; 18:
230–235.
24. Harris RE, Conrad FG. Polycythemia vera in the
childbearing age. Archives of Internal Medicine 1967;
120: 697–700.
25. Ruggeri M, Tosetto A, Castaman G, Rodeghiero F.
Pulmonary embolism after pregnancy in a patient with
polycythemia vera. American Journal of Hematology
2001; 67: 216–217.
26. Pata O, Tok CE, Yazici G et al. Polycythemia vera and
pregnancy: a case report with the use of hydroxyurea
in the first trimester. American Journal of Perinatology
2004; 21: 135–137.
27. Robinson S, Bewley S, Hunt BJ et al. The management
and outcome of 18 pregnancies in women with
polycythemia vera. Haematologica 2005; 90;
1477–1483.
28. Taylor UB, Bardeguez AD, Iglesias N, Gascon P.
Idiopathic myelofibrosis in pregnancy: a case report

241

Section 7. Malignant conditions

and review of the literature. American Journal of
Obstetrics and Gynecology 2004; 167: 38–39.
29. Gotic M, Cvetkovic M, Bozanovic T, Cemerikic V.
[Successful treatment of primary myelofibrosis
with thrombocytosis during pregnancy with
alfa-interferon] Srp Arh Celok Lek 2001; 129: 304–308.

242

30. Tulpule S, Bewley S, Robinson SE, Radia D,
Nelson-Piercy C, Harrison CN. The management and
outcome of 4 pregnancies in women with idiopathic
myelopfibrosis. British Journal of Haematology 2008;
142: 480–482.

Section 7
Chapter

Malignant conditions

Effects of chemoradiotherapy
for hematological malignancy
on fertility and pregnancy
Seonaid Pye and Nina Salooja

Introduction
Advances in treatment for hematological malignancies
over the last two decades have led to marked improvements in survival, and for many patients this translates to cure. Consideration of the long-term sequelae of treatments administered are therefore becoming
increasingly relevant to the overall management strategies of these disorders and, since many of the potentially treatable hematological cancers occur in children and young adults, this includes concerns about
future fertility. The chemotherapy agents and radiotherapy used to treat leukemias and lymphomas can
affect reproductive potential in a variety of ways. In
this chapter we will consider:
r the general effects of chemoradiotherapy on
female fertility;
r the incidence of infertility following radiation and
chemotherapy;
r the likely outcome of pregnancy in a patient
treated for hematological malignancy;
r the strategies available for the preservation of
fertility in patients who require potentially
sterilizing treatment.

General effects of chemoradiotherapy
on fertility
Normal reproduction requires interplay between the
gonads and the hypothalamic–pituitary–endocrine
axis. In addition, the uterus must be receptive to
implantation and capable of effecting appropriate
growth in pregnancy. Damage to hormone producing
cells in the hypothalamus, pituitary, or gonads can lead
to infertility as well as more direct damage to the germ
cells, reproductive tracts, or sexual organs.

In females the production of germ cells (oocytes)
ceases before birth. Thereafter, the number of oocytes
decrease throughout life, either by a mechanism of preprogrammed cell death (physiological apoptosis) or
else post-menarche, in menstruation. When the number of oocytes falls below a critical number, ovulation
and ovarian function cease. A female’s fertility potential is therefore related to the number of oocytes, which
are present in the ovary. Chemotherapy and radiotherapy both lead to an additional irreversible reduction in
oocyte numbers by mechanisms which involve activation of apoptotic pathways.
The chance of retaining ovarian function following
such external insults to germ cell numbers depends on
the starting number of oocytes and this is related to the
age of the patient. Thus young women who start with
high numbers of oocytes are more likely to recover
menses and fertility following chemoradiotherapy,
although they remain at risk of premature ovarian
failure. Older women with relatively low numbers of
oocytes remaining in the ovary pre-treatment, frequently experience immediate and irreversible cessation of ovarian function. Clinically, this may be accompanied by severe vasomotor symptoms and biochemically it is associated with low serum levels of estradiol
with significantly elevated levels of FSH and LH.

The incidence of infertility
following radiation
The clinical effects of radiotherapy on fertility depend
on the dose and radiation field in addition to patient
age as discussed above. Animal studies have shown
that increasing doses of ovarian radiation lead to loss
of primordial follicles in a dose-dependent manner.
The dose at which 50% of human oocytes are lost
(LD50) has been estimated to be ⬍2Gy.1 Considering

The Obstetric Hematology Manual, ed. Sue Pavord and Beverley Hunt. Published by Cambridge University Press.

C Cambridge University Press 2010.

243

Section 7. Malignant conditions

Table 20.1 Radiation sites relevant to reproductive potential

Site of irradiation

Tissues relevant to
fertility potential

Irradiation to the cranium, for
example:
(a) total body irradiation
(b) craniospinal irradiation
(c) direct cranial irradiation

Hypothalamic–pituitary
axis

Irradiation to the abdomen or pelvis,
for example:
(a) total body irradiation
(b) total lymphoid irradiation
(c) craniospinal irradiation
(d) direct irradiation to the pelvis or
abdomen

Ovaries
Uterus
Reproductive tract

the radiation fields, treatment impinging on either the
cranium or reproductive tract can lead to impairment
of fertility (Table 20.1).

Direct cranial irradiation
Irradiation of the hypothalamic–pituitary axis leads to
a classical pattern of hormone loss, with growth hormone being the most sensitive and first to be affected
followed by the gonadotrophins. In patients with
intracranial disease associated with acute leukemia,
direct cranial radiation may be administered. Doses
in the range 18–24 Gy may result in isolated growth
hormone (GH) deficiency. In addition, subtle disturbances in the menstrual cycle may occur, although their
relevance to future reproductive potential is unclear.
Delayed puberty has been described in young girls who
receive doses above 24 Gy.
Patients with pituitary and primary brain tumors
can be given higher doses of radiation exceeding 50 Gy.
Such doses can damage pituitary function directly,
but features of hypopituitarism in patients who have
received doses less than 50 Gy are more likely to be secondary to hypothalamic dysfunction.

Direct abdomino-pelvic radiation

244

As discussed above, radiation to the ovaries leads to
damage which is dose related. Abdomino-pelvic irradiation also damages the uterus with adverse effects
documented on the endometrium, myometrium, and
vasculature. In a study which investigated length and
blood flow of the uterus in ten women aged 15–
31 following 20–30 Gy abdominal radiotherapy, there
were significant reductions in uterine length and blood
flow compared with women whose treatment had not
included abdominal radiation.2

Radiation induced damage might also be expected
to impair implantation and/or growth and development of a fetus and there is data to support this. In a
study where 38 patients were given 20–26.5 Gy abdominal radiotherapy in childhood, there were four documented conceptions but no live births, all miscarrying
in the second trimester.3

Total body irradiation
This affects all of the radiation sites relevant to fertility
potential. It is usually given together with chemotherapy, as conditioning prior to stem cell transplantation(SCT) and serves two separate functions:
r suppression of the host immune system to allow
donor engraftment;
r eradication of hemopoiesis in the host bone
marrow.
Doses of 8–15 Gy are administered either as a single dose or in fractions. At these doses, the effects on
the hypothalamic–pituitary axis are usually minimal,
but both ovarian function and uterine function are
compromised. The incidence of ovarian failure is high.
In a single center study, which included 144 women
who had received total body irradiation (TBI) as conditioning for SCT, all became amenorrheic immediately post-transplant and only 9 of the 144 recovered
menses at a median of 4 years following treatment.4 All
who regained ovarian function were aged less than 25
at the time of SCT.
Uterine function following TBI has been less extensively studied, but in a study which included 12 women
who had received TBI in childhood, there was a reduction in uterine volume to 40% of adult size despite the
use of sex steroid replacement therapy.5 These patients
had received either unfractionated TBI at a midline
dose of 8.5–10 Gy (n = 4) or a total midline dose of
10.9–11.7 Gy in three fractions (n = 8). These data
suggest that the adverse effects of radiation may be
more marked if given pre-pubertally, before optimum
growth of the uterus has been achieved.

The incidence of infertility following
chemotherapy
The likelihood of infertility following chemotherapy
depends on
r the drug(s) administered;
r the doses to which the patient is exposed;

Chapter 20. Effects of chemoradiotherapy

r the underlying disease;
r patient age.

Meirow6 provides an elegant analysis of sterilizing
effects of different classes of chemotherapeutic agents
(Fig. 20.1). Data on 168 patients treated with combination chemotherapy were evaluated and the odds
ratio for ovarian failure calculated for exposed versus

Chemotherapy agents can be divided into classes
based upon their mechanism of action (Table 20.2).

Table 20.2 Classes of chemotherapeutic agents and their action

Drug class/subclasses

Examples

Mechanism of action

Alkylating agents
Nitrogen mustards
Nitrosureas
Alkyl sulfonates
Triazines
Ethylenimines
Platinum drugs∗

Cyclophosphamide, chlorambucil, melphalan
BCNU (carmustine), Lomustine
Busulfan
Dacarbazine
Thiotepa, altretamine
Cisplatin, carboplatin

DNA damage

Methotrexate, 5-fluorouracil, 6-mercaptopurine,
gemcitabine, cytarabine (Ara-C), fludarabine

Interfere with nucleic acid or nucleotide synthesis

Antibiotics
Anthracyclines

Daunorubicin, doxorubicin, epirubicin, idarubicin

Various mechanisms, e.g. interference with
enzymes involved in DNA synthesis

Other

Bleomycin, mitomycin-C. mitoxantrone∗∗

Topoisomerase Inhibitors
Topoisomerase I
Topoisomerase II

Topotecan
Etoposide (VP-16), Mitoxantrone∗∗

Mitotic inhibitors
Taxanes
Vinca alkaloids
Epothilones

Paclitaxel, docetaxel
Vinblastine, vincristine
Ixabepilone (ixempra)

Tyrosine kinase inhibitors

Imatinib, dasatinib

Miscellaneous

Bortezomib (Velcade), L-asparaginase

Antimetabolites

∗∗

platinum drugs are grouped here with alkylating agents because they have a similar mechanism of action.
similar to doxorubicin but also acts as topoisomerase II inhibitor,

Fig. 20.1 In 168 cancer patients treated
by combination chemotherapy, the
overall ovarian failure rate was 34%,
representing an odds ratio of 1.0.
Medications were in five drug categories
(alkylating agents, platinum derivatives,
antibiotics, anti-metabolites, and plant
alkaloids) and analysis was performed on
these groups. The fraction contributed by
each of the chemotherapeutic classes was
analyzed by the odds ratio of exposed
versus non-exposed patients. The results
were adjusted for age. (Figure reproduced
with permission from Preservation of
Fertility, Tulandi and Gosden 2004, page
31.)

Odds ratio adjusted for age

∗

3.5
3
2.5
2
1.5
1
0.5
0
Alkylating
agents

Cis-platinum

Antibiotics

Antimetabolites

Plant alkaloids

245

Section 7. Malignant conditions

Table 20.3 Relationship between dose of cyclophosphamide and age

Age

Number of
patients

Number developing
amenorrhea

Average cumulative dose
at onset of amenorrhea

⬎40

5.2 g (range 1.4–8.4 g)

30–40

4 (2 subsequently resumed menses)

9.3 g (7–11 g)

20–30∗

20.4 g (14–24.5 g)

From Ref. 7.
∗ these patients received other treatment modalities in addition to CY.

unexposed patients. Results were then adjusted for age
by logistic regression analysis. These data show that
alkyating agents and platinum derivatives are associated with the highest risks of ovarian failure with odds
ratios of 3.98 and 1.7, respectively.
For many individual drugs, however, the true
age-related, dose-related incidence of infertility is
unknown because there are insufficient longitudinal
data using them as single agents. A notable exception
to this is cyclophosphamide for which there are extensive published data. This is because the drug is useful in the treatment of a variety of diseases that affect
women of child-bearing age: these include breast cancer, autoimmune disorders such as SLE, and hematological malignancies. Cyclophosphamide also plays an
important part in conditioning treatment given prior
to allogeneic stem cell transplantation where it can be
used as:
r a high dose single agent (for example, in patients
transplanted for severe aplastic anemia, SAA);
r in combination with busulfan (for example,
pediatric and adult transplantation for leukemia);
r together with total body irradiation (for example,
in adults transplanted for leukemia).

246

The relationship between ovarian failure and
age in women administered cyclophosphamide was
clearly demonstrated by a study in which premenopausal women with breast cancer were treated
with cyclophosphamide (CY) at a dose of 100 mg/day.7
The data are illustrated in Table 20.3 and show that a
total cumulative dose in excess of 11 g will lead to cessation of menstruation in most women over the age of
30 but not younger women.
Although the cumulative dose administered is
important as illustrated above, data from transplant
centers where cyclophosphamide is administered in
a single high dose as pre-transplant conditioning
suggest that this may be a particularly gonadotoxic

approach. Follow-up of 43 women with SAA who
received CY in doses of 200 mg/kg as pre-transplant
conditioning demonstrated acute cessation of menstruation in all 27 of these patients who were less than
26 years of age at the time of transplant subsequently
recovered ovarian function in comparison with only 5
of the 16 women aged ⬎26.4
More limited data are available on other
chemotherapeutic drugs used as single agents.
An association of busulfan with ovarian failure has
been noted as far back as the 1950s with cumulative
doses of 150–400 mg associated with acute amenorrhea. More recent data from transplant patients in
which high doses of busulfan (BU) are incorporated
into pre-transplant conditioning regimens further
highlights the gonado-toxicity of this agent. In a large
European multicenter evaluating pregnancy following
SCT, the combination of BUCY as pre-transplant
conditioning appeared more gonado-toxic than
CY/TBI as there were no pregnancies in patients
with malignant disease who had received BUCY in
standard doses for pre-transplant conditioning.8
There are fewer protocols involving use of chlorambucil as a single agent in young females. In a small
study of 10 pre-pubertal girls exposed to cumulative
doses of chlorambucil ranging from 9–28 mg/kg for
autoimmune disease, all had normal pubertal development including normal age at onset of menarche.9
Larger cumulative doses of 535–750 mg/m2 administered to women with breast cancer, however, are associated with ovarian failure.10
Data on drug combinations will now be discussed
in the context of the underlying disease.

Acute leukemias
Conventional treatments for acute myeloid leukemia
(AML) and acute lymphocytic leukemia (ALL) are
generally less gonado-toxic than those used to treat
lymphomas. Typical induction regimens for AML

Chapter 20. Effects of chemoradiotherapy

involve drugs such as cytarabine, daunorubicin, and
etoposide followed by consolidation treatment, which
may incorporate amsacrine or mitoxantrone. The incidence of persistent gonadal damage following treatment with anthracycline-based regimens during childhood or adulthood has been reported as ⬍10%. AML
survivors had a 6% incidence of acute ovarian failure
in the Childhood Cancer Survivors Study published in
2006.11
Acute lymphoblastic leukemia is the commonest childhood cancer, although it also affects adults.
In addition to induction and consolidation phases,
treatment of ALL also incorporates a maintenance
phase and CNS-directed therapy. The latter is generally
administered as intrathecal chemotherapy with cranial
or craniospinal irradiation reserved for those at high
risk (5%–20%) of CNS relapse. The drugs commonly
used in the treatment of ALL are glucocorticoids, vincristine, anthracycline, and asparaginase. High dose
methotrexate may be administered to those with high
risk disease and a tyrosine kinase inhibitor, imatinib, is
used for patients who have Philadelphia positive ALL.
The incidence of persistent gonadal damage in females
following treatment of childhood ALL with standard
UKALL protocols is less than 20%. Those who received
craniospinal irradiation or cyclophosphamide as part
of their treatment are at greatest risk. Data from the
Childhood Cancer Survivors Study demonstrated an
acute ovarian failure rate in ALL survivors of 14%.11

Chronic leukemias
Chronic lymphocytic leukemia is predominantly a disorder of the elderly and so will not be discussed further.
Chronic myeloid leukemia (CML) is typically a disorder of middle-aged adults, but a significant number of
cases occur in women of child-bearing age (15–49). In
the past, the mainstay of treatment for chronic myeloid
leukemia (CML) has been treatment with hydroxycarbamide (formerly known as hydroxyurea) followed by
stem cell transplantation. In the last 10 years, however,
there has been considerable success in managing CML
with tyrosine kinase inhibitors such as imatinib: the
first example of a molecularly targeted therapy. Imatinib was first administered to patients with CML in
1998 and has now been used to treat more than 60 000
patients worldwide. It is given orally and is generally
well tolerated. To date, there are limited data available on the effects of imatinib on fertility. Reproductive studies in animals have shown imatinib is terato-

Table 20.4 Treatment regimens for Hodgkin’s disease and
likelihood of gonadal failure

Risk of
gonadal
failure

Combination chemotherapy
regimens for Hodgkin’s
lymphoma

High risk (⬎80%)

MVPP
MOPP
ChlVPP/EVA
COPP

Intermediate risk

BEAM (BCNU, etoposide, cytarabine
and melphalan)
VEBEP
Alternating ABVD/MOPP or COPP
CHOP

Low risk (⬍20%)

VAPEC-B
BEACOPP
VEEP
ABVD

genic in rats, so patients are advised to avoid pregnancy
while taking it. Nonetheless, 180 pregnancies have
been reported in patients who were taking imatinib.12
Dosage data were not available in every case, but many
were receiving standard doses of 300–400 mg. Recent
data, however, suggest that higher doses of imatinib
may be associated with premature ovarian failure.13

Lymphomas
Hodgkin’s and non-Hodgkin’s lymphoma (HL and
NHL) together account for approximately 10% of
pediatric cancers (4% HL, 6% NHL). With modern treatments for HL in excess of 90% of children
and adolescents can expect to be cured. Alkylating
agents have played a major role in many of the combination chemotherapy protocols proposed for the
treatment of HL and many of these have therefore
been associated with infertility (Table 20.4). In the
1970s, treatment regimens containing nitrogen mustard such as MOPP (nitrogen mustard, vincristine,
procarbazine, and prednisolone) and MVPP (nitrogen
mustard, vinblastine, procarbazine, and prednisolone)
were used and were associated with oligo- or amenorrhea in approximately 20%–40% of women. In the mid
1970s, however, it was discovered that a new regimen
combining doxorubicin, bleomycin, vinblastine, and
dacarbazine (ABVD) was as efficacious a treatment
as MOPP or MVPP, but lacked the gonado-toxicity.
ABVD has become the modern gold standard of treatment and the risk of sterilization in women under the
age of 25 is almost zero.14 Women treated with inverted
Y-irradiation, in addition to alkylating agents, have

247

Section 7. Malignant conditions

been shown to have a significantly higher risk of premature menopause, however.
Non-Hodgkin’s lymphomas (NHL) can be broadly
subdivided into low grade and high grade lymphomas.
Low dose oral chemotherapy such as chlorambucil is
appropriate for many patients with low grade disease
and it has been discussed above. The gold standard
treatment for high grade NHL is a combination of
the following drugs: cyclophosphamide (750 mg/m2 ),
doxorubicin, vincristine, and prednisolone (CHOP).
Pre-menopausal women treated using this regimen
are likely to develop amenorrhea during chemotherapy, but the majority (95%) will resume menstruation
shortly after completion of treatment. The risk of permanent ovarian failure is highest in those aged 40 or
more at the time of treatment. Even when higher doses
of cyclophosphamide are used as in “mega-CHOP,”
in which 2–3 g/m2 cyclophosphamide are administered, it appears that the risk of persistent ovarian failure may be low with 92% (12/13) of women regaining ovarian function in one study.15 Women in this
latter study who were aged ⬍40 years were offered
GnRH analogs in parallel with their chemotherapy in
an attempt to preserve fertility. A second study which
combined data on various chemotherapy regimens for
NHL in premenopausal women found a higher incidence of ovarian failure of 44%,16 but the patient
characteristics and details of chemotherapy regimens
used were not reported and this is likely to underlie the discrepancy. Childhood NHL is treated with
similar protocols to childhood ALL and the risks of
infertility and delayed puberty are low. In a recent
prospective study of survivors of childhood NHL or
ALL, all 40 females treated with chemotherapy alone
or chemotherapy plus cranial irradiation underwent
spontaneous menarche. Whether these patients will
subsequently undergo premature ovarian failure is not
known, however.17

The likely outcome of pregnancy in
a patient treated for hematological
malignancy

248

There are several reasons why pregnancy outcome
might be adversely affected by prior treatment with
chemoradiotherapy. Irradiation to the uterus may
affect implantation, potentially predisposing to
miscarriage or intra-uterine growth retardation.
Chemotherapy agents such as cyclophosphamide

can cause gene mutations, chromosomal breaks, and
rearrangements raising the possibility of an increased
risk of congenital malformations.
Although the focus of this chapter is on the
relationship between chemoradiotherapy and fertility
potential, effects on other maternal organs may cause
complications for pregnancy and delivery. Cardiac and
pulmonary toxicities, for example, are well described
following some regimens. Patients at risk of such complications should have a cardiorespiratory review early
in pregnancy, including an echocardiogram and pulmonary function tests and may require assessment by
an anesthetist prior to delivery. Similarly, patients with
renal impairment will require expert review and monitoring throughout pregnancy.
There are some data to support concerns of an
adverse pregnancy outcome resulting from pelvic irradiation. A retrospective multicenter study identified
139 pregnancies in 111 female patients who had
received SCT.8 Of these 111 women, 39 had received
autologous stem cells and 74 had received allogeneic
stem cells. Of the latter group, 21 had been conditioned with TBI-containing regimens. In this study, the
majority of pregnancies were uncomplicated; however,
20% of female allograft recipients had pre-term singleton deliveries (normal incidence approximately 6%)
and 23% had low birth weight singleton babies (normal incidence approximately 6%). These complications were confined to women who had received total
body irradiation. The incidence of Cesarean section
was significantly higher amongst allografted women at
42% compared with approximately 16% in the normal
population, but the incidence of congenital anomalies
amongst offspring was not increased.
A further report from the Childhood Cancer Survivor Study looked at pregnancy outcome for 4029
pregnancies in 1915 women previously treated for cancers in childhood. The pregnancy outcome of the sibling closest in age to the patient was used for control
data. Their results showed that women treated with
pelvic irradiation tended to have smaller babies than
the controls and delivered earlier, at an average of 37.23
weeks vs. 38.47 weeks for controls. Use of daunorubicin or doxorubicin was also linked adversely to birth
weight but there was no clear dose–response relationship.18
Despite the theoretical concern of congenital
abnormalities arising in offspring of survivors of cancer treatment, available data do not demonstrate a substantial increase in risk in patients where conventional

Chapter 20. Effects of chemoradiotherapy

chemotherapy has been used prior to conception. It
may be, however, that existing studies do not have sufficient power to detect a small difference.

eral target proteins of relevance to embryonic development, such as cKIT and PDGFR.

Strategies for fertility preservation
Effects of chemotherapy
during pregnancy
The risks of teratogenicity are significant if chemotherapy is administered during pregnancy, particularly
in the first trimester. Management of hematological
malignancy in pregnancy will therefore depend on the
stage of pregnancy when a diagnosis is made and the
balance of risk between delaying treatment and teratogenicity. In patients who require curative chemotherapy in the first trimester, then therapeutic termination
should be discussed. This would include patients with
acute leukemia and also some patients with aggressive
or extensive/bulky lymphomas. In some cases it may
be possible to delay treatment to later in pregnancy
or even until after delivery. In patients with CML, for
example, leukapheresis can be used to temporarily
control the white cell and platelet counts. There
have been reports of certain chemotherapy agents
being administered in the second and third trimester
without complication, for example, adriamycin,
CHOP, and rituximab; however, data is limited.
Some highly teratogenic drugs, such as methotrexate
and dacarbazine, should be avoided at all stages of
pregnancy.
Recent data also highlight concerns about the
potential teratogenic effects of imatinib. As discussed
above, this tyrosine kinase inhibitor is used in the management of both CML and Philadelphia positive ALL.
In rats imatinib leads to exencephaly, encephaloceles,
and deformities of the skull bones. Female rats administered doses ⬎45 mg/kg (which equates to approximately half the maximum human dose of 800 mg/day,
based on body surface area) experienced significant
post-implantation loss with increased fetal resorption,
stillbirths, non-viable pups, and early pup mortality. In
a study which included data on pregnancy outcomes
for 125 women exposed to imatinib at conception and
during part or all of the first trimester there were 12 offspring identified with abnormalities, 3 of which were
terminated electively. Of the offspring with identifiable anomalies, 3 had strikingly similar complex malformations, which were unlikely to have occurred by
chance.12 Imatinib does not appear to damage chromosomes, but it is capable of interacting with sev-

Some women who require treatment for cancer have
more than one therapeutic option. In such cases,
women who hope to commence a family after treatment may be able to avoid potentially sterilizing treatment. In women who require pelvic irradiation, it may
be possible to laparoscopically transpose the ovaries
outside the field of radiation, leaving the ovarian blood
supply intact. This is not always successful, however;
not only can the ovaries migrate back into the field of
radiation, but complications can occur as a result of the
procedure such as chronic pain or formation of ovarian cysts. Alternatively, modified field radiation can
sometimes be planned to omit/reduce radiation to the
ovaries. Scatter radiation can nonetheless contribute to
ovarian failure and follow-up remains important.
In women whose treatment is highly likely to result
in infertility, cryopreservation of embryos prior to
treatment offers the best hope for parenthood postchemoradiotherapy. This technique requires ovarian
stimulation over a 2-week period, following which
mature oocytes in their second metaphase are collected. These oocytes are fertilized in vitro before freezing. This option is not open to all patients with cancer,
however. Ovarian stimulation takes 2 weeks, but it has
to be timed in relation to the menstrual cycle. Treatment delays can therefore extend to 6 weeks and this is
prohibitive for many patients with cancer. Additional
complexity arises if the patient lacks a male partner to
provide sperm. The option of donor sperm can be considered, but this requires careful counseling in relation
to future implications (see below).
In healthy women who attempt pregnancy with
transfer of thawed embryos, the pregnancy rate is
20%–30%. There are several reasons why the outcome of artificial reproductive techniques (ART) may
be lower in women with cancer, however. Firstly,
women with cancer do not always respond well to
stimulation regimens and the quality and number of
oocytes may be lower than expected. Secondly, many
patients will have compromised endometrial function
in addition to ovarian failure as a result of their treatment. This could potentially impede implantation or
fetal growth and development. There are some data
to support this from a European multicenter study
which included 9 women who conceived using ART

249

Section 7. Malignant conditions

following TBI.8 Among the pregnancies to these
women, the incidence of preterm delivery and low
birth weight offspring was high, and median birth
weights were lower than expected for gestational age.
Although there are a number of case reports of successful pregnancies using cryopreserved embryos following systemic cancer therapy, it is difficult to quote accurate success rates for these patients and some will elect
to use a surrogate if available to carry their embryos.

Experimental approaches
to restoring fertility
Freezing unfertilized oocytes
Although this is a promising option for women requiring sterilizing treatment who lack a partner to provide sperm, live birth rates following this procedure
are currently low at about 2%.19 This is too low to
justify routine use of this technique in clinical practice. Low pregnancy rates partly reflect the susceptibility of the mature unfertilized oocyte to thermal
and osmotic injury, which exceeds that of the preimplantation embryo. Furthermore, it is probable that
poorer quality embryos are generated following oocyte
cryopreservation. Vitrification, which involves ultrarapid cooling, has improved post-thaw oocyte survival
and pregnancy rates in small studies, but further information is required on the efficiency and safety of this
technique. Nonetheless, in recent years there have been
reports of successful pregnancies in women treated
for cancer using this technique. Yang and colleagues20
report a patient with Hodgkin’s disease who had frozen
oocytes thawed and fertilized by ICSI after cryopreservation for 6 years. During the time that the oocytes
were frozen, the patient had multiple relapses and was
treated with combination chemotherapy (ABVD followed by COPP followed by fludarabine) and total
body irradiation (200 cGy) in the context of a nonmyeloablative SCT. Nine embryos were obtained from
her frozen oocytes and all of these were implanted over
the course of three separate cycles. A surrogate was
used because of concerns of radiation damage to the
patient’s uterus, and a successful pregnancy occurred
after the final cycle of implantation.

Freezing ovarian tissue

250

This is currently the only option open to pre-pubertal
patients or to those women whose disease will not tol-

erate a significant delay in treatment. Cortical fragments containing primordial follicles with immature
oocytes can be obtained by laparoscopy and frozen.
Ideally, ovarian tissue should be obtained before the
patient has been exposed to chemotherapy, but this
is not always possible and is not an absolute requirement. Attempts to restore ovarian function and fertility have involved reimplanting the ovarian tissue,
either orthotopically adjacent to the ovary or heterotopically, for example, into the anterior abdominal wall. A major concern with this technique is the
possibility of reintroducing cancer cells, and a thorough histological assessment of the tissue should be
made before re-transplantation. In 2005 Meirow et al.
described a successful pregnancy in a patient with nonHodgkin’s lymphoma using cryopreserved ovarian tissue in conjunction with IVF and additional pregnancies have been reported since.21 However, it is likely
that ovarian tissue transplanted in this way will have
a limited lifespan and transplantation of ovarian tissue
should probably be reserved for assisting the restoration of fertility rather than for restoring hormone production.
The use of assisted reproductive techniques in cancer patients raises a range of ethical concerns, including several issues relating to consent. Consent takes
place when two or more people agree upon a course
of action and it implies that agreement occurred:
r without coercion;
r based on the provision of information; and
r that the participants have the ability to
understand the facts and implications of the
action (“competence”).
In the UK, consent for long-term cryopreservation of
gametes is governed by the Human Fertilization &
Embryology Authority (HFEA) and they have constructed guidance and consent forms, which are available at www.hfea.gov.uk. The consent of both partners
is required when embryos are cryopreserved and also
when the embryos are replaced. If either partner withdraws consent, the embryos cannot be used. In the
UK, young people aged 16–18 can consent to treatment under the Family Law Reform Act 1969 (“competent minors”). The position in younger patients was
established in the case of Gillick v West Norfolk Area
Health Authority (1985). As a result of this case, children who are of sufficient understanding and capable of expressing their own wishes (Gillick competent) can also make informed decisions. Under HFEA

Chapter 20. Effects of chemoradiotherapy

regulations, parents or guardians cannot give consent
on behalf of a child for the storage or use of gametes.
Immature germ cells obtained from gonadal tissue of
pre-pubertal children do not come under this remit,
however. The tissue can therefore be recovered with
parental consent if it is considered to be in the best
interest of the child.

Conclusions
Treatment of hematology cancers is constantly evolving to produce improved survival data and incorporate better tolerated agents. As a result of this, an
increasing number of young patients diagnosed with

hematological malignancies can now hope to lead relatively normal adult lives and for many this includes
the expectation of parenthood. Management of possible infertility should start before cancer treatment is
administered and, ideally, should include a full discussion of: (1) treatment options and the likelihood
of infertility associated with each option; (2) strategies
for preserving fertility if the chance of sterilization as
a result of treatment is high. Full data are not always
available, however, particularly where new drugs are
used or when experimental methods for preserving
fertility are considered. Long-term follow-up studies
of patients treated for cancer remains a central priority to provide the core information required for such
pre-treatment counseling.

251

Section 7. Malignant conditions

References
1. Wallace WHB, Thomson AB, Kelsey TW. The
radiosensitivity of the human oocyte. Human
Reproduction. 2003; 18: 117–121.

12. Pye S, Cortes J, Ault P et al. The effects of imatinib on
pregnancy outcome. Blood 2008; 111: 5505–5508.

2. Critchley HO, Wallace WH, Shalet SM et al.
Abdominal irradiation in childhood: the potential for
pregnancy. British Journal of Obstetrics and
Gynaecology. 1992; 97: 804–810.

13. Christopoulos C, Dimakopoulou V, Rotas E.
Primary ovarian insufficiency associated with imatinib
therapy. New England Journal of Medicine 2008; 358:
1079–1080.

3. Wallace WH, Shalet SM, Crowne EC et al. Ovarian
failure following abdominal irradiation in childhood:
natural history and prognosis. Clinical Oncology. 1989;
1: 75–79.

14. Meirow D, Dor J. Epidemiology and infertility in
cancer patients. In Preservation of Fertility, Tulandi T
and Gosden RG, ed. Taylor & Francis. 2004.

4. Sanders JE, Buckner CD, Amos D et al. Ovarian
function following marrow transplantation for aplastic
anemia or leukemia. Journal of Clinical Oncology.
1988; 6: 813–818.
5. Holm K, Nysom K, Brocks V et al. Ultrasound B-mode
changes in the uterus and ovaries and Doppler changes
in the uterus after total body irradiation and allogeneic
bone marrow transplantation. Bone Marrow
Transplantation. 1999; 23: 259–263.
6. Meirow D. Reproduction post-chemotherapy in young
cancer patients. Molecular Cell Endocrinology. 2000;
169: 123–131.
7. Koyama H, Wada T, Nishizawa Y et al.
Cyclophosphamide-induced ovarian failure and its
therapeutic significance in patients with breast cancer.
Cancer 1977; 39: 1403–1409.
8. Salooja N, Szydlo RM, Socie G et al. Pregnancy
outcomes after peripheral blood or bone marrow
transplantation: a retrospective study. The Lancet 2001;
358: 271–276.
9. Callis L, Nieto J, Vila A, Rende J. Chlorambucil
treatment in minimal lesion nephrotic syndrome: a
reappraisal of its gonadal toxicity. Journal of Pediatrics
1980; 97: 653–656.
10. Freckman HA, Fry HL, Mendez FL et al.
Chlorambucil–prednisolone therapy for disseminated
breast carcinoma. Journal of the American Medical
Society 1964; 189: 23–26.
11. Chemaitilly W, Mertens AC, Mitby P et al. Acute
ovarian failure in the Childhood Cancer Survivor

252

Study. Journal of Clinical Endocrinology and
Metabolism 2006; 91: 1723–1728.

15. Dann EJ, Epelbaum R, Avivi I et al. Fertility and
ovarian function are preserved in women treated with
an intensified regimen of cyclophosphamide,
adriamycin, vincristine and prednisone (Mega-CHOP)
for non-Hodgkin lymphoma. Human Reproduction
2005;20; 2247–2249.
16. Meirow D. Reproduction post-chemotherapy in young
cancer patients. Molecular Cell Endocrinology 2000;
169:123–131.
17. Steffens M, Beauloye V, Brichard B et al. Endocrine
and metabolic disorders in young adult survivors of
childhood acute lymphoblastic leukemia (ALL) or
non-Hodgkin lymphoma (NHL). Clinical
Endocrinology 2008; 69: 819–827.
18. Green DM, Whitton JA, Stovall M et al. Pregnancy
outcomes in female survivors of childhood cancer: a
report from the Childhood Cancer Survivor Study.
American Journal of Obstetrics and Gynecology 2002;
187: 1070–1080.
19. Oktay K, Cil AP, Bang H et al. Efficiency of oocyte
cryopreservation: a meta-analysis. Fertility and
Sterility 2006; 86: 70–80.
20. Yang D, Brown SE, Nguyen K et al. Live birth after the
transfer of human embryos developed from
cryopreserved oocytes harvested before cancer
treatment. Fertility and Sterility 2007; 87: 1469.
e1–4.
21. Meirow D, Baum M, Yaron R et al. Ovarian tissue
cryopreservation in hematologic malignancy: ten
years’ experience. Leukemia and Lymphoma 2007; 48:
1569–1576.

Index

Note: page numbers in italics refer to figures and tables
abdomino-pelvic radiotherapy 244
ABO blood group
incompatibility 86
red cell products 166
abortion, recurrent spontaneous
antiphospholipid syndrome
133
see also miscarriage
ABVD drug regimen 247–248
acidosis, prevention in obstetric
hemorrhage 160, 159–160
activated partial thromboplastin time
(APTT) 5
disseminated intravascular
coagulation 212
factor XI deficiency 189–190
activated protein C (APC)
resistance in pregnancy 7
sensitivity ratio test
acute chest syndrome
blood transfusion 36
sickle cell disease 36
prevention 37
acute fatty liver of pregnancy 214,
223–224
clinical signs 223
diagnosis 223
pathogenesis 223–224
acute lymphocytic leukemia
(ALL) 246–247
acute myeloid leukemia
(AML) 246–247
ADAMTS 13 219–221, 222
adjunctive ante-natal treatments 83
adult respiratory distress syndrome
(ARDS), catastrophic
antiphospholipid
syndrome 134
alpha gene deletion 39
alpha thalassemia 38

carrier 38
distribution 29
amniocentesis 78–79
disadvantages 79
hemophilia 197–198
mutation detection 198
miscarriage risk 197–198
amniotic fluid embolism, cell
salvage 161
anemia
Fanconi’s 47
isoimmunized pregnancy 78–79
megaloblastic 23
microangiopathic hemolytic 214
postpartum 21
pregnancy 57
see also autoimmune hemolytic
anemia (AIHA); iron
deficiency
anesthesia
cesarean section
epidural catheter 122–123
full anticoagulation 125–126
epidural 122, 122–123
factor XI deficiency 191
labor
full anticoagulation 124
heparin use 122
obstetric hemorrhage
management 158–164
single shot subarachnoid 122,
122–123
spinal block 122
timing 123
see also general anesthesia; regional
anesthesia
analgesia
hemophilia 188
labor 123
full anticoagulation
124–125
von Willebrand disease 181
angiogenic factors, pre-eclampsia
205

antepartum hemorrhage 151
diagnosis 153
management 153
anti-␤2 glycoprotein I antibodies 131,
135
procoagulant effects 132
anticardiolipin antibodies (aCL) 131,
135
anticoagulation
cardiopulmonary bypass 126
delivery 120–126
full anticoagulation 123–126
life-saving treatment 126
management during
pregnancy 110–112
prosthetic heart valves 109
cesarean section 116
labor induction 116
management 111, 114
maternal/fetal outcomes 112
maternal/fetal risk 114
monitoring 115, 115
anti-D immunoglobulin 50, 76
administration 86–87
feto-maternal hemorrhage
prevention 76
immune/idiopathic
thrombocytopenic purpura
treatment 50, 51–52
miscarriage 76
prophylaxis
refusal 77
routine ante-natal 76–77
recombinant monoclonal
antibodies 86–87
mutated 87
safety 86–87
anti-D isoimmunization
measurement 78
antifibrinolytics, obstetric
hemorrhage 167–168
antihypertensive drugs,
pre-eclampsia 211
prevention 209

253

Index

antiphospholipid antibodies
(aPL) 131
acquired abnormalities 141–142
␤2 glycoprotein I-dependent 132
laboratory evaluation 135
management of pregnancy
with 136–137
pre-eclampsia 208
systemic lupus erythematosus 225
venous thrombosis 133

254

antiphospholipid syndrome 131–139
etiology 131–132
anti-␤2 glycoprotein I
antibodies 131, 135
anticardiolipin antibodies 131, 135
arterial thrombosis 133
aspirin 137
␤2 glycoprotein I
lupus anticoagulant 135
procoagulant effects 131–132
catastrophic 134
classification criteria 133
clinical features
CNS 134
heart valve defects 134
thrombocytopenia 134
thrombosis 132–134
CNS effects 134
heart valve defects 134
heparin replacement of
warfarin 137
lupus anticoagulant 135
management 135–139
dilemmas 138–139
pharmacological 138
plan 136
pregnancy morbidity 137–139
pre-pregnancy 136
women with previous venous
events 137
miscarriage 133–134, 138, 144
neonates 139
obstetric complications
early 133
late 134
morbidity 132
pathophysiology 131–132
placental dysfunction 134
pre-eclampsia 134, 208
preterm delivery 134
prevalence 131
primary 131
recurrent spontaneous
abortion 133
secondary 131
thrombocytopenia 134, 138
thrombosis 132–134
management 136
ultrasonography 137–138

warfarin replacement with
heparin 137
antiplatelet agents
delivery 120–121
pre-eclampsia prevention 209–210
see also aspirin
antithrombin
pre-eclampsia 210
pregnancy levels
antithrombin concentrates,
pre-eclampsia prevention 210
antithrombotics
pre-eclampsia prevention 209–210
pregnancy loss 146–147
thrombophilia 146–147
aprotinin, obstetric hemorrhage 167
arterial thrombosis, antiphospholipid
syndrome 133
artificial reproductive technologies
(ART) 249–250
ethics 250–251

fetus 58
hemolysis treatment 58
laboratory studies 57
management 57–58
neonates 58
pathogenesis 56–57
thromboprophylaxis 58
warm type 56, 58
autoimmune neutropenia (AIN) 45,
54–56
diagnosis 54–55
granulocyte colony-stimulating
factor 55–56
history taking 55
incidence 54
laboratory assessment 55
management 55–56
pathogenesis 54
postpartum period 56
pregnancy 56
sepsis 55
azathioprine, immune/idiopathic
thrombocytopenic purpura 52

ascites, fetal 74
aspirin
antiphospholipid syndrome 136,
137
delivery 120–121
labor 120–121
myeloproliferative disorders 236,
238
pre-eclampsia prevention 209–210
prosthetic heart valves 113,
114–115
thrombotic thrombocytopenic
purpura 222
venous thromboembolism
treatment 102
von Willebrand disease 181
autoantibodies
autoimmune hemolytic
anemia 57–58
maternal autoimmune
cytopenias 45
autoimmune cytopenias, maternal 58
autoantibodies 45
see also named conditions and
diseases
autoimmune hemolytic anemia
(AIHA) 45, 56–58
autoantibodies 57–58
blood transfusion 57–58
cold type 56, 58
diagnosis 57
epidemiology 56–57
examination 57

B cells 4
balloon tamponade 156
Bart’s hydrops 38–39
Bernard–Soulier syndrome 183–184
neonates 184
beta thalassemia carrier 39
beta thalassemia intermedia 39
beta thalassemia major 28, 39
screening 28–30
␤2 glycoprotein I 131
lupus anticoagulant 135
procoagulant effects 131–132
bilirubin 73, 78
serum levels 84
blood component therapy, obstetric
hemorrhage 160–161, 167
blood group systems 74–75
blood patch, epidural 125
blood transfusion
acute chest syndrome 36
autoimmune hemolytic
anemia 57–58
complications 84
fetal 79–80
historical landmarks 79
hematocrit 80
hemolytic disease of the
newborn 79–80
indicators 84

Index

iron deficiency 19
obstetric hemorrhage 160–161
sickle cell disease 36
prophylaxis 37–38
thalassemia 38, 39
pregnancy 41
ultrasound-guided direct
intravascular transfusion 79
B-Lynch brace suture 156
bone, thalassemia
complications 40–41
bone marrow
folate deficiency 23
immune/idiopathic
thrombocytopenic
purpura 46, 49
breastfeeding
glucose-6-phosphate dehydrogenase
deficiency 43
heparin use
low molecular weight heparin 122
busulfan, infertility incidence 246
cesarean section
anesthesia
epidural catheter 122–123
full anticoagulation 125–126
single shot
subarachnoid 122–123
anticoagulant management with
prosthetic heart valves 116
bleeding risk with LMW
heparin 125
factor XI deficiency 190
heparin use 122–123
immune/idiopathic
thrombocytopenic purpura
53
life-saving treatment 126
postpartum hemorrhage 173
sickle cell disease 37
thromboprophylaxis 122–123
carboprost, obstetric hemorrhage 159
cardiac problems
pre-eclampsia 207
thalassemia 41–42
cardiopulmonary bypass 126
cardiovascular disease,
pre-eclampsia 207
cauda equina syndrome 125, 125–126
cell salvage 162
machine 163
obstetric hemorrhage 161
principles 161–162

central nervous system (CNS),
antiphospholipid
syndrome 134
central venous catheterization,
obstetric hemorrhage 158
central venous pressure, monitoring in
obstetric hemorrhage 158
cephalhematoma,
hemophilia 187–188
chemoradiotherapy 243–251
fertility
effects 243
preservation 249–250
teratogenicity risk 248–249
teratogens 249
toxicity 248
chemotherapy 243–251
agents 245
infertility incidence 244–246
ovarian failure rate 245

see also coagulopathy, inherited
coagulation screen, obstetric
hemorrhage 169
coagulopathy
inherited 186–192
factor XI deficiency 189–191
genetic diagnosis 198
hemophilia 186–189
rare 191–192
prevention in obstetric
hemorrhage 160, 159–160
colloid fluids, obstetric
hemorrhage 159
competence 250–251
competitive heme oxygenase
inhibitors 87
compression stockings see venous
compression stockings
(TEDS)

children, consent 250–251

computed tomography pulmonary
angiography (CTPA)
pulmonary thromboembolism
radiation exposure

chlorambucil, infertility incidence 246

congenital heart defects, scanning 115

CHOP drug regimen 248

consent 250–251

chorionic villus sampling,
hemophilia 197

constitutional thrombocytopenia 47

chronic lymphocytic leukemia
(CLL) 247

cord traction, controlled 152–153

childhood cancer treatment,
pregnancy outcome 248

chronic myelogenous leukemia
(CML) 229, 247
circle of Willis 81
clopidogrel 120–121
clot lysis time 10
coagulation factor(s)
activity assays 196
hormonal influences 191
mutations 141–142
pregnancy 5, 6
loss 143
replacement in von Willebrand
disease 179–180
coagulation factor deficiencies
management
ante-natal 191
neonatal 192
pre-pregnancy 191
miscarriage 192
neonates 192
rare 191–192
thrombosis risk 192
treatment 191–192

contraception, sickle cell disease 30
corticosteroids 50
antiphospholipid
syndrome 135–136
autoimmune hemolytic anemia 58
autoimmune neutropenia 55
fetal and neonatal alloimmune
thrombocytopenia maternal
treatment 68
HELLP syndrome 213–214
immune/idiopathic
thrombocytopenic purpura
treatment 50, 51–52
thrombotic thrombocytopenic
purpura 221–222
counseling
genetic for hemophilia 194–197
pre-natal in immune/idiopathic
thrombocytopenic purpura 54
pre-pregnancy
antiphospholipid syndrome 136
coagulation factor
deficiencies 191
cranial irradiation, direct 244
Creutzfeldt–Jakob disease, variant
(vCJD) 182

255

Index

crystalloid fluids, obstetric
hemorrhage 159
cyclophosphamide 248
dose and age relationship 246
infertility incidence 246
cytopenia see autoimmune cytopenias,
maternal
cytoreductive therapy,
myeloproliferative
disorders 236–238
postpartum 238–239
danaparoid 103
D-dimers, pregnancy 8–10
testing in venous
thromboembolism
deep vein thrombosis
diagnosis
epidemiology
management
post-thrombotic syndrome
deferasirox 41
deferiprone 41

256

delivery
anticoagulation 120–126
full 123–126
antiplatelet agents 120–121
aspirin 120–121
factor XI deficiency 190, 190
hemophilia 188, 187–188
hemorrhagic complications 124
heparin
low molecular weight 121–123
timing 123
immune/idiopathic
thrombocytopenic purpura
maternal considerations 52–53
mode 53
neonatal considerations 53
myeloproliferative disorders 238
pre-eclampsia 210–211
induced 210
prosthetic heart valves 115–116
sickle cell disease
management 36–37
thalassemia 41
thromboprophylaxis 124
unfractionated heparin
use 123–124
venous compression stockings
124
venous thromboembolism
management
von Willebrand disease
management 180–181

delta OD450 79, 78–79

erythroblastosis fetalis 84

desamino-8-D-arginine vasopressin
(DDAVP)
hemophilia 187, 188
obstetric hemorrhage 169
platelet function inherited
disorders 183
von Willebrand disease 179, 182
contraindications 180

erythrocytosis 235

desferrioxamine 41
diabetes mellitus type 1,
thalassemia 41
disseminated intravascular
coagulation (DIC) 225
activated partial thromboplastin
time 212
catastrophic antiphospholipid
syndrome 134
hypertensive thrombocytopenia 47
management 212
obstetric hemorrhage 169
placental abruption 214
pre-eclampsia 212
prothrombin time 212
dura mater puncture 125
dysfibrinogenemia 192
thrombosis risk 192
eclampsia 47, 203
elliptocytosis 42
embryo
cryopreservation 249–250
loss in thrombophilia 144, 143–144
embryopathy, warfarin 103, 112,
113
emergencies, obstetric, time to
death 152
endocrine conditions, thalassemia 41
endoglin 205, 208–209
endothelial cell activation,
pre-eclampsia 205
endothelial dysfunction,
pre-eclampsia 205–206
Entonox
epidural procedures
anesthesia for cesarean
section 122–123
dura mater puncture 125
vertebral canal hematoma risk 122
ergometrine, obstetric
hemorrhage 159

erythropoiesis
iron deficiency 14–15
megaloblastic 23
erythropoietin
recombinant 84–85
supplementation in iron
deficiency 18–19
essential thrombocythemia (ET) see
primary thrombocythemia
(PT)
ethics, artificial reproductive
technologies 250–251
evacuation of retained products of
conception (ERPC) 155–156
Evan’s syndrome 45
factor V Leiden
fetal 144
pregnancy loss 141–142, 143, 143
factor VIIa, recombinant
activated 183, 190, 192
factor VIII
ante-natal management 179
assays 177, 196
deficiency 186
hormonal influences in
pregnancy 177–178, 186, 187
mutations 196
postpartum levels 181
prophylaxis 186
replacement 179–180
thrombotic thrombocytopenic
purpura 219–220
von Willebrand disease 177
von Willebrand factor ratio 196
see also hemophilia
factor IX
assays 196
deficiency 186
mutations 196
prophylaxis 186
see also hemophilia
factor XI concentrate 190
factor XI deficiency 189–191
cesarean section 190
clinical features 189
complications 189
delivery 190, 190
incidence 189
labor 190, 190
management
ante-natal 190, 189–190

Index

intrapartum 190, 190
neonatal 191
postpartum 191
prepregnancy 189
neonates 191
peripartum bleeding risk 189
regional anesthesia 191
tranexamic acid 191
vaginal delivery 190
factor XIII
deficiency and miscarriage risk 192
maternal 192
Family Law Reform Act
(1969) 250–251
Fanconi’s anemia 47
fentanyl analgesia in labor 125
ferric carboxymaltose 18
ferritin
iron deficiency 14
pregnancy levels 14
ferrous salts, iron supplementation 17
fertility
artificial reproductive technologies
after
chemoradiotherapy 249–250
chemoradiotherapy effects 243
preservation with
chemoradiotherapy 249–250
restoration 250–251
thalassemia 39–40
fetal and neonatal alloimmune
thrombocytopenia
(FNAIT) 63–71
ante-natal management 70, 66–70,
71
optimal approach 71
ante-natal screening 65
clinical diagnosis 64
clinical significance 64
epidemiology 63
history in previous pregnancies 65
incidence 63–64
information for mother 66
intracranial hemorrhage 64, 65
intravenous immunoglobulin 66,
83
laboratory diagnosis 64
laboratory testing 65
management 65–71
maternal treatment 68–71
complications 68
non-responders 70
neonates 64
platelets 63, 65–66, 68
post-natal management 65
red cell antigens 86

severity prediction 65
ultrasound-guided fetal blood
sampling 66, 67, 67–68
fetal blood sampling (FBS),
ultrasound-guided 66, 67,
67–68
feto-maternal hemorrhage 73, 76
risk 74, 76
size quantification 76
fetus
abdomino-pelvic irradiation 244
autoimmune hemolytic anemia
58
blood transfusion 79–80
genes contributing to
pre-eclampsia 207–208
myeloproliferative disorder
monitoring 238
pericardial effusion 74
platelet transfusion
serial 66–67
in utero 66
thrombophilia 144, 144
warfarin effects 103, 112
dosage in pregnancy 113
see also teratogens
fibrin 10
fibrin degradation products (FDP) 8,
10
fibrinogen 167, 192
fibrinolysis 8–10
markers in pregnancy 7
fluid resuscitation
obstetric hemorrhage 159
warmer 160
folate deficiency 21–24
blood count 23
blood film 23
bone marrow 23
clinical signs/symptoms 25
diagnosis 23
epidemiology 21–23
hematinic assays 23
management 23–24
pathogenesis 23
prevention 24
prophylaxis 23
treatment 24
folic acid
autoimmune hemolytic anemia
treatment 58
food supplementation 24
preconceptual 23, 24
thrombophilia 146
requirements 24

fondaparinux 103
free fetal DNA (ffDNA) 80–81
hemophilia diagnosis 197, 198
fresh frozen plasma (FFP) 167
coagulation factor
deficiencies 191–192
factor XI deficiency 190
Gelfoam embolization 172, 173, 173
general anesthesia
cesarean section 125
obstetric hemorrhage 162
genetic counseling,
hemophilia 194–197
genetic diagnosis
hemophilia
carriers 196–197
first trimester 197–198
pre-natal 197–198
inherited coagulopathy 198
gestational hypertension 47, 203
gestational thrombocytopenia 46–47,
219
Gillick competence 250–251
Glanzmann’s thromabasthenia 183
glucose-6-phosphate dehydrogenase
deficiency 42–43
ante-natal management 43
breastfeeding 43
neonatal management 43
glycoprotein IIb/IIIa deficiency 183
graduated compression stockings see
venous compression stockings
(TEDS)
granulocyte colony-stimulating factor
(GCSF) 55–56
growth hormone (GH) deficiency 244
hematinic deficiencies 13–26
see also folate deficiency; iron
deficiency; vitamin B12
deficiency
hematocrit
blood transfusion 80
myeloproliferative
disorders 237–238
pregnancy 3
hematological malignancy
childhood 248
pregnancy outcomes 248–249
see also leukemia; lymphomas

257

Index

hematological variables in
pregnancy 237
hemoglobin Bart’s hydrops 38
hemoglobin H disease 38
hemoglobin levels
ante-natal management 22
iron deficiency 15
peripheral decrease 15
pregnancy 3
hemoglobin S 30
polymerization 30
hemoglobinopathies 28–43
ante-natal screening 28–30
neonatal screening 28–30
red cell membrane disorders 42–43
structural 28
unbalanced globin chain
production 28
see also named conditions and
diseases
hemolysis
autoimmune hemolytic anemia 56
causes 56, 57
immune-mediated 78
laboratory studies 57
sickle cell disease 30
treatment 58
see also HELLP (hemolysis, elevated
liver enzymes and low
platelets) syndrome
hemolytic disease of the
newborn 73–74
blood transfusion 79–80
indicators 84
intravenous immunoglobulin 84
light therapy 84
management 83–86
severe disease 83
routine ante-natal anti-D
prophylaxis 77
see also red cell alloimmunization
hemolytic uremic syndrome (HUS)
HELLP differential diagnosis 214
liver involvement 224–225
pre-eclampsia differential
diagnosis 214
pregnancy 47

258

hemophilia 186–189, 194
amniocentesis 197–198
mutation detection 198
analgesia 188
carrier status prediction 195, 195,
195
carriers 194–195
genetic detection 196

genetic diagnosis
limitations/hazards 196–197
laboratory detection 196
cephalhematoma 187–188
chorionic villus sampling 197
clinical features 186
complications in
pregnancy 186–187
DDAVP 187, 188
delivery 188, 187–188
fetal sexing 197
first trimester genetic
diagnosis 197–198
genetic counseling 194–197
heritability 194–195
incidence 186
intracranial hemorrhage 186–189
labor 188, 187–188
management
ante-natal 187, 187
intrapartum 188, 187–188
neonatal 188–189
postpartum 188
prepregnancy 187
maternal bleeding 186
mutations 196–197
neonatal risk 186–187
neonates 188–189
pedigree analysis 195, 195
pre-implantation diagnosis
198
pre-natal diagnosis 197–198
future techniques 198
severity 187
somatic mosaicism 196–197
sporadic 195, 195
tranexamic acid 188
transmission potential 195, 196
vitamin K 188
hemorrhage, obstetric 151–156
access 158
anesthetic management 158–164
general 162
regional 162
antepartum 151
management 153
antifibrinolytics 167–168
blood/blood component
therapy 160–161
cell salvage 161
principles 161–162
classification 160
coagulation screen 169
communication 158
definition 166
diagnosis 153
disseminated intravascular
coagulation 169
documentation 163–164

drills 164
early warning scoring system 159
epidemiology 151
factor XI deficiency 189
fibrinogen 167
fluid resuscitation 159
fresh frozen plasma 167
full blood count 169
general anesthesia 162
hemophilia 186
hemostatic replacement
therapy 166–167
hypothermia, acidosis and
coagulopathy ‘lethal triad’
prevention 160, 159–160
interventional radiology
elective management 172–173
emergency 172
emergency postpartum 172,
171–172
prophylactic
management 172–173
investigations 162
management 153–154, 169
anesthetic 158–164
hemostatic 166–169
practical points 169
monitoring 158, 169
mortality 158
oxytocics 158–159
pharmacological agents 167–169
placental abruption 214
platelet transfusion 166–167
post-hemorrhage care 163, 169
prevention 151–153, 155
protocols 164
radiological management
171–174
recognition 155
red cell products 166
regional anesthesia 162
time to death in emergencies 152
treatment 154
von Willebrand disease 178
see also postpartum hemorrhage;
primary postpartum
hemorrhage (PPH)
hemostasis, inherited
disorders 176–184
hemostatic markers, pregnancy 9
hemostatic replacement
therapy 166–167
blood/blood products 160–161, 167
desamino-8-D-arginine
vasopressin 169
fibrinogen 167
fresh frozen plasma 167
platelet transfusion 166–167

Index

prothrombinase
complexes 168–169
recombinant factor VIIa 168
red cell products 166
see also blood transfusion
headache, post-dural puncture 125
heart valve defects in antiphospholipid
syndrome 134
heart valves, prosthetic 109–117
ante-natal care 115
anticoagulation 109
aspirin 113, 114–115
cesarean section 116
heparin 109, 110–112,
113–116
labor induction 116
management 111, 114
maternal/fetal outcomes 112
maternal/fetal risk 114
monitoring 115, 115
warfarin 112, 115–116
aspirin 113, 114–115
biprosthetic 109–110
congenital heart defect
scanning 115
delivery management 115–116
infective endocarditis
prevention 116
labor management 115–116
management 111
replacement indications 109
thromboembolism
prevention 110
risk 110
valve thrombosis management
116
valve types 110, 109–110
warfarin anticoagulation 112
replacement with
heparin 115–116
HELLP (hemolysis, elevated liver
enzymes and low platelets)
syndrome 47, 224
acute fatty liver of pregnancy
differential diagnosis 214
antiphospholipid syndrome 134
clinical presentation 213
complications 213
corticosteroids 213–214
definition 212
diagnosis 213
differential diagnosis 213, 214
epidemiology 212–213
hypertension 213
management 213–214
pathophysiology 213
platelet transfusion 214
pre-eclampsia 212–214

thrombocytopenia 218
Hemocue

163

heparin
antiphospholipid syndrome 136,
137
breastfeeding
cesarean section 122–123
delivery 121–123
timing 123
labor 121–122
anesthesia 122
analgesia 123
induction 121
recommencing 121–122
low molecular weight 102–103
ante-natal management 103,
104–105
antiphospholipid syndrome
137
bleeding risk at cesarean
section 125
breastfeeding 122
cesarean section 122–123
delivery 121–123
labor 121–122
labor analgesia 123
myeloproliferative disorders 236,
238
post-natal management 105
postpartum treatment
pregnancy loss 147
prosthetic heart valves 109,
110–112, 113–116
thrombophilia 147
thromboprophylaxis 101
thrombotic thrombocytopenic
purpura 222
treatment monitoring
with warfarin 103
warfarin replacement 115–116,
123, 137
myeloproliferative disorders 236,
238
pre-eclampsia prevention 210
prosthetic heart valves in pregnant
women 109, 110–112,
113–115
warfarin replacement 115–116
side-effects 102–103
unfractionated 102
delivery management
prosthetic heart valves 109,
110–112, 113, 114–116
vaginal delivery 123–124
warfarin replacement 115–116
venous thromboembolism
treatment 102–103
ante-natal management 104–105

delivery
maintenance
monitoring
post-natal management 105
postpartum
heparin-induced
thrombocytopenia 102, 103
hepatitis C, thalassemia 41
hereditary persistence of fetal
hemoglobin (HPFH) 29
hereditary spherocytosis 42
Hodgkin’s lymphoma 247–248
treatment 247
homocysteine
folate deficiency 23
pregnancy 11
Human Fertilisation and Embryology
Authority (HEFA) 250–251
human platelet alloantigens
(HPAs) 63
immunization against 64
hydrops fetalis 73, 83–84
hydroxycarbamide 37
hyperemesis gravidarum 222–223
hyperhomocystinemia treatment
24
thrombophilia 146
hypertension
chronic 203
gestational 47, 203
HELLP syndrome 213
sickle cell disease 34
thrombocytopenia 47
hypertension in pregnancy (HIP) 47
hyperuricemia, pre-eclampsia
marker 208
hypervolemia, physiological in
pregnancy 14
hypogonadotrophic hypogonadism,
thalassemia 39–40
hypothalamic–pituitary axis,
radiotherapy 244
hypothermia, acidosis and
coagulopathy ‘lethal triad’
prevention 160, 159–160
hysterectomy, primary postpartum
hemorrhage 156
imatinib 247, 249
immune response, maternal 83

259

Index

immune/idiopathic thrombocytopenic
purpura (ITP) 45, 54, 218
bone marrow 46, 49
clinical examination 48
delivery planning
maternal considerations 52–53
mode 53
neonatal considerations 53
diagnosis 46–48
epidemiology 46
evaluation of suspected disease 49
history 48
laboratory assessment 48–49
labor management 53–54
life-threatening bleeding 51–52
management 54
monitoring 50
neonates 53–54
pathogenesis 46
platelets 46, 48–49, 50
postpartum care 53–54
pregnancy monitoring 50
pre-natal counseling 54
refractory case management 52
splenectomy 52
thrombocytopenia 46–49
treatment 51, 50–51
immunoglobulin G (IgG)
maternal 73–74
maternal autoimmune
cytopenias 45
implantation, abdomino-pelvic
irradiation 244
infections
sickle cell disease 34
thrombocytopenia 47–48
transfusion transmitted in von
Willebrand disease 182
infective endocarditis, prevention with
prosthetic heart valves 116
infertility incidence
chemotherapy 244–246
radiotherapy 243–244
inflammatory bowel disease,
antiphospholipid
syndrome 131
internal iliac artery embolization
172
internal iliac vessel ligation 156

260

interventional radiology
indications for use 172
obstetric hemorrhage 171–174
postpartum hemorrhage
elective management 172–173
emergency 172, 171–172

intracranial hemorrhage
fetal and neonatal alloimmune
thrombocytopenia 64, 65
hemophilia 186–189
immune/idiopathic
thrombocytopenic purpura
delivery 53
intrahepatic cholestasis of
pregnancy 222–223
intrauterine growth restriction
(IUGR)
pre-eclampsia 203, 207
sickle cell disease 34
uterine artery abnormal
flow 218–219
intravascular transfusion,
ultrasound-guided direct
79
intravenous immunoglobulin
(IVIG) 50
autoimmune hemolytic anemia
58
autoimmune neutropenia 55
fetal and neonatal alloimmune
thrombocytopenia
treatment 66, 68, 83
hemolytic disease of the
newborn 84
immune/idiopathic
thrombocytopenic purpura
treatment 50, 51–52
Rhesus D antigen isoimmunization
prevention 83
serum plasmapheresis
combination 83
iron
absorption 13, 17
dietary 16–17
homeostasis 13–14
intramuscular 18
liquid form 17
overload in thalassemia 39, 39–40
parenteral 17–18
placental regulation of transfer to
fetus 14
requirements in pregnancy 13, 14
supplementation 16–18, 20
preventive 19–20
tablet form 17
iron chelation, thalassemia 38, 39–40,
41
iron deficiency 13–21
blood transfusion 19
clinical signs/symptoms 15, 15
diagnosis 14
effects 16

epidemiology 13
erythropoiesis 14–15
erythropoietin 18–19
hemoglobin levels 15
iron supplementation 16–18
laboratory investigations 16
management 16–19
maternal 14
pathogenesis 13–14
postpartum 21
prevention 19–20
screening 20–21
storage decrease 14
treatment 19
iron dextran 18
iron sucrose 18
Janus kinase 2 (JAK2) protein 233
mutations 233, 235
jaundice, ABO incompatibility 86
Kell antigen 85–86
kernicterus, risk 84
labetalol, pre-eclampsia treatment 211
labor
anesthesia
full anticoagulation 124
heparin use 122
analgesia 123
full anticoagulation 124–125
aspirin 120–121
factor XI deficiency 190, 190
hemophilia 188, 187–188
heparin
analgesia 123
induction 121
recommencing 121–122
immune/idiopathic
thrombocytopenic
purpura 53–54
induction
heparin use 121
with prosthetic heart valves 116
low molecular weight
heparin 121–122
prosthetic heart valves 115–116
sickle cell disease
management 36–37
third stage 124
active management 152–153
von Willebrand disease
management 180–181
warfarin 124
leukocyte filters 163

Index

leukemia
acute 246–247
chronic 229, 247
Libman–Sacks endocarditis 134
light therapy, hemolytic disease of the
newborn 84
liver, thalassemia complications 41
liver disease in pregnancy 222–223
acute fatty liver of pregnancy 214,
223–224
intrahepatic cholestasis of
pregnancy 223
see also hemolytic uremic syndrome
(HUS); HELLP (hemolysis,
elevated liver enzymes and low
platelets) syndrome;
pre-eclampsia

mirror syndrome 39
miscarriage
amniocentesis risk 197–198
anti-D use 76
antiphospholipid syndrome
133–134, 138, 144
coagulation factor deficiencies 192
definition 141
dysfibrinogenemia 192
factor XIII deficiency 192
sickle cell disease 34
see also pregnancy loss
misoprostol, obstetric
hemorrhage 159
monoclonal antibody-specific
immobilization of platelet
antigens (MAIPA) assay 64

lupus anticoagulant 135

monocytes 4

lymphocytes, pregnancy 4

myelocytes 4

lymphomas 247–248

myelofibrosis (PMF) 229, 232
case study 240
diagnosis 235

magnesium sulphate, pre-eclampsia
treatment 211
May Hegglin anomaly 47
mean corpuscular hemoglobin
concentration (MCHC), iron
deficiency 15
mean corpuscular volume (MCV)
iron deficiency 15
pregnancy 3
mechanical valves see heart valves,
prosthetic
megakaryocytes, immune/idiopathic
thrombocytopenic purpura 49
megakaryopoiesis, immune/idiopathic
thrombocytopenic purpura
46
megaloblastic anemia 23
megaloblastic erythropoiesis, folate
deficiency 23
metamyelocytes 4
methyldopa, pre-eclampsia
treatment 211
microangiopathic hemolytic
anemias 214
microangiopathies,
pregnancy-related 219
middle cerebral artery (MCA) 81
peak systolic flow velocity 81, 81,
82, 83

myeloproliferative disorders 229–240
aspirin 236, 238
case studies 239–240
clinical conditions 229–233
cytoreductive therapy 236–238
postpartum 238–239
delivery 238
diagnosis 233–235
epidemiology 229–233
familial 239
fetal monitoring 238
hematocrit 237–238
heparin 236, 238
high risk criteria 237
management 237–238
plan 235
postpartum 238–239
monitoring 238
myelofibrosis 229, 232
case study 240
diagnosis 235
pathogenesis 233
PCV 235–238
postpartum 238–239
Philadelphia-negative 229
placental infarction 233
platelet count 235–238
postpartum 238–239
polycythemia vera 229, 231
case study 239–240
diagnosis 234–235
treatment 236
postpartum
thromboprophylaxis 238

preconceptual meeting 237
primary thrombocythemia 229,
234
case study 239, 240
incidence 230
polycythemia vera 236
thrombosis 233
risk management 238
treatment 235–237
venesection 237
venous compression stockings 236,
238
MYH-9 disorders 47
myocardial infarction (MI), cardiac
compromise 126
natural anticoagulants, pregnancy 6,
7, 6–7
neonates
antiphospholipid syndrome 139
autoimmune hemolytic anemia 58
Bernard–Soulier syndrome 184
coagulation factor deficiencies 192
factor XI deficiency 191
fetal and neonatal alloimmune
thrombocytopenia 64
Glanzmann’s thromabasthenia 183
hemophilia 186–187, 188–189
immune/idiopathic
thrombocytopenic
purpura 53–54
von Willebrand disease 182, 183
neural tube defects
folic acid supplementation 24
preconceptual folic acid 23
neutropenia 54
severity 55
see also autoimmune neutropenia
(AIN)
NHS sickle and thalassemia
screening 28–30
nifedipine, pre-eclampsia
treatment 211
nitrogen mustard drug regimens 247
non-Hodgkin’s lymphoma 247–248
obesity, maternal and venous
thromboembolism 100–101,
120
obstetric shock, non-hemorrhagic
oocytes
freezing of unfertilized 250
production 243

261

Index

operative delivery
sickle cell disease 37
see also cesarean section
opioid analgesia in labor 124–125
oral anticoagulants (OACs) 110–112
labor/delivery
management 115–116
osteopenia, thalassemia
complications 40–41
osteoporosis
heparin-induced 102–103
thalassemia complications 40–41
ovarian function, chemoradiotherapy
effects 243
ovaries
abdomino-pelvic irradiation 244
failure rate with chemotherapy 245
laparoscopic transposition 249
tissue freezing 250
total body irradiation 244
oxytocic drugs, obstetric
hemorrhage 158–159
oxytocin 152–153, 158
packed cell volume (PCV),
myeloproliferative
disorders 235–238
postpartum 238–239
partial D alleles 85
partner testing, thalassemia 39–40
parvovirus B19 182
peak systolic flow velocity (PSV),
middle cerebral artery 81, 81,
82, 83
pedigree analysis, hemophilia carrier
prediction 195, 195
pelvic irradiation
fertility preservation 249
pregnancy outcome 248

262

placental abruption 214
placentation, normal pregnancy 142
plasma exchange, thrombotic
thrombocytopenic
purpura 221, 222
plasma products, von Willebrand
disease 182
complications 182
plasmin inhibitors 9–10
plasminogen activator inhibitors
(PAI) 8–10
platelet function inherited
disorders 183–184
platelet transfusion
HELLP syndrome 214
obstetric hemorrhage 166–167
serial fetal 66–67, 67
in utero 67
platelets
fetal and neonatal alloimmune
thrombocytopenia 63, 65–66,
68
function assays 177
gestational
thrombocytopenia 46–47
immune/idiopathic
thrombocytopenic
purpura 46, 48–49, 50
myeloproliferative
disorders 235–238
postpartum 238–239
pregnancy 5
thrombocytopenia in
pre-eclampsia 211–212
pneumatic compression boots, venous
thromboembolism 101–102
polycythemia vera (PV) 229, 231
case study 239–240
diagnosis 234–235
treatment 236
porphyrin ring 15

pericardial effusion, fetal 74

post-dural puncture headache 125

physiological changes during
pregnancy 222

postpartum hemorrhage
elective management 172–173
emergency
cesarean section 173
interventional
radiology 171–172
management protocol 172
interventional radiology 171–173
prophylactic management 172–173
see also primary postpartum
hemorrhage (PPH)

placenta
circulation in pregnancy loss 141
dysfunction in antiphospholipid
syndrome 134
high-risk pregnancy 218–219
infarction in myeloproliferative
disorders 233
manual removal 155
pathology in pregnancy loss 142
placenta accreta 172–173

postpartum period
autoimmune neutropenia 56
fetal and neonatal alloimmune
thrombocytopenia
management 66
immune/idiopathic
thrombocytopenic
purpura 53–54
myeloproliferative
disorders 238–239
sickle cell disease management 37
venous thromboembolism
management
post-thrombotic syndrome
pre-eclampsia 47, 203–215
angiogenic factors 205
antihypertensive drugs 211
antiphospholipid antibodies 208
antiphospholipid syndrome 134
antithrombin 210
cardiovascular risk 207
with chronic hypertension 203
complications 204
constitutional factors 207
delivery 210–211
induced 210
diagnosis 203–204
differential diagnosis 214
disseminated intravascular
coagulation 212
early onset 207
endoglin 205, 208–209
endothelial cell activation 205
endothelial dysfunction 205–206
endothelial markers 208
epidemiology 203
fetal genes 207–208
hematological
complications 211–214
HELLP syndrome 212–214
intrauterine growth restriction 203,
207
late onset 207
liver involvement 224
management 209–211
guidelines 211
microparticles 206
pathogenesis 206, 205–206, 208
placental markers 208
prediction 208–209
prevention 209–210
recurrence 203
regional anesthesia 210
risk factors 204
severe 204
sickle cell disease 34
signs/symptoms 204
thalassemia 39

Index

thrombocytopenia 211–212, 218
thrombophilia 208
trophoblasts 207–208
two stages model 206, 205–206
uterine artery abnormal
flow 218–219
pregnancy
hematological variables 237
physiological changes 222
pregnancy loss
antithrombotics 146–147
clotting factors 143
epidemiology 141
factor V Leiden 141–142
incidence 143
management 145–147
dilemmas 147
evidence requirement 147
pathogenesis 141–142
placental circulation 141
placental pathology 142
prevalence 142
recurrent 141
thrombophilia 141–147
diagnosis 145
heritable 142–144
very early 143–144
see also miscarriage
preterm delivery
antiphospholipid syndrome 134
sickle cell disease 34
primary postpartum hemorrhage
(PPH) 151
diagnosis 153
management 153–156
surgical 155–156
volume maintenance 154
pathogenesis 153
prevention 151–153
risk factors 152
primary thrombocythemia (PT) 229
case study 239, 240
diagnosis 234
incidence 230
polycythemia vera 236
prostaglandin E1 analogue 154
obstetric hemorrhage 159
protamine sulphate 124, 126
protein C
deficiency 141–142
pregnancy 6–7
protein S
deficiency 141–142
pregnancy loss 143
pregnancy 6–7

prothrombin complex, coagulation
factor deficiencies 191
prothrombin fragments 8
prothrombin G20210A mutation 143
fetal 144
pregnancy loss 143
prothrombin time (PT)
disseminated intravascular
coagulation 212
pregnancy 5
prothrombinase complexes 168–169
protoporphyrin 15
pulmonary thromboembolism
cardiac compromise 126
diagnosis
life-saving treatment 126
management
massive life-threatening
thrombolytic therapy
radiology see interventional radiology
radiotherapy
abdomino-pelvic 244
direct cranial irradiation 244
hypothalamic–pituitary axis 244
infertility incidence 243–244
pelvic irradiation, pregnancy
outcome 248
sites 244
total body irradiation 244
recombinant factor VIIa, obstetric
hemorrhage 168
red cell(s)
fetal leakage into maternal
circulation 76
folate levels 23
pregnancy 3, 4
red cell alloimmunization 73–87
cell salvage 161
genotype 74–75
intervention timing 78
management
recent advances 80–83
traditional 77–80
pathogenesis 73–74
phenotype 74–75
risk 74
red cell disorders, inherited 28–43
red cell membrane disorders 42–43
red cell products, hemostatic
replacement therapy 166
regional anesthesia
cesarean section 125–126

factor XI deficiency 191
obstetric hemorrhage 162
pre-eclampsia 210
remifentanil 125
RHCE gene 75
RHD gene 75
DNA sequences 85
RHD pseudogene 85
RHD/CE hybrids 85
Rhesus D antigen 74–75
isoimmunization prevention 75,
75–76
IVIG use 83
negativity 77–78
positivity 77–78
red cell products 166
status determination 78, 85
non-invasive fetal testing 80–81
variants 85
see also anti-D immunoglobulin
Rhesus D disease
blood transfusion 79–80
intervention timing 78
rheumatic fever 109, 116
prevalence 110
rheumatic heart disease 109, 116
rituximab
autoimmune hemolytic anemia 58
immune/idiopathic
thrombocytopenic purpura
52
routine ante-natal anti-D prophylaxis
(RAADP) 76–77
dosing schedule 77
scalp edema 74
sedation, maternal for fetal blood
transfusion 80
sepsis, autoimmune neutropenia 55
shock
hemorrhagic classification 160
obstetric non-hemorrhagic
sickle cell crisis 34
management in pregnancy 35–36
sickle cell disease
acute chest syndrome 36
prevention 37
blood transfusion 36
prophylaxis 37–38
booking time management 34–35
compound heterozygous 30
contraception 30

263

Index

sickle cell disease (cont.)
delivery management 36–37
fetal complications 30–34
hemolysis 30
homozygous (HbSS) 30
hydroxycarbamide 37
hypertension 34
infections 34
intrauterine growth retardation 34
labor management 36–37
management 34–36, 38
booking time 34–35
delivery/labor 36–37
postpartum 37
preconception 34
maternal complications 30–34
maternal mortality 33
miscarriage 34
morbidity/mortality 28
operative delivery 37
pathogenesis 30
peri-natal mortality 33
postpartum management 37
preconception management 34
pre-eclampsia 34
pregnancy management 35–36
prematurity 34
prophylactic blood
transfusion 37–38
screening 28–30
thrombosis risk 34
vaso-occlusion 30
sickling disorders 30–34
spinal block 122
timing 123
splenectomy, immune/idiopathic
thrombocytopenic purpura
52
stillbirth, dysfibrinogenemia 192
streptokinase, pulmonary
thromboembolism
systemic lupus erythematosus (SLE)
antiphospholipid antibodies 225
antiphospholipid syndrome 131
exacerbation 225
systemic thromboembolism,
prevention with prosthetic
heart valves 110

thalassemia 38–42
blood transfusion 38, 39
pregnancy 41
cardiac problems 41–42
delivery 41
fertility 39–40
hepatitis C risk 41
hypogonadotrophic
hypogonadism 39–40
iron chelation 38, 39–40
during pregnancy 41
iron overload 39, 39–40
management 39–42
medical problems in
pregnancy 40–41
partner testing 39–40
pathogenesis 38–39
preconception management 40
pre-eclampsia 39
pregnancy management 40
pregnancy risks 40
risks to baby 40
thrombin 189
thrombin activatable fibrinolysis
inhibitor (TAFI) 189
thrombin time (TT) 5
thrombin–antithrombin II (TAT)
complexes 8
thrombocytopenia
antiphospholipid syndrome 134,
138
causes 47
constitutional 47
drug-induced 47–48
gestational 46–47, 219
heparin-induced 102, 103
hypertension 47
immune/idiopathic
thrombocytopenic purpura 46
pregnancy 46–49
infections 47–48
moderate to severe 218
pre-eclampsia 211–212
see also fetal and neonatal
alloimmune thrombocytopenia
(FNAIT)
thromboelastograph analyzer 8
thromboelastograph trace 9
thromboelastography 7–8

T cells 4
TEDS see venous compression
stockings (TEDS)

264

teratogens
chemoradiotherapy 248–249
hydroxycarbamide 37

thromboembolism
prevention with prosthetic heart
valves 110
risk with prosthetic heart valves 110
see also deep vein thrombosis;
pulmonary thromboembolism;
venous thromboembolism

thrombolytic therapy, pulmonary
thromboembolism
thrombophilia
antithrombotics 146–147
embryo loss 144, 143–144
fetal 144, 144
heparin use 147
heritable
incidence 143
pre-eclampsia 208
pregnancy loss 142–144
prevalence 142
management 145–147
dilemmas 147
evidence requirement 147
pathogenesis 141–142
placental pathology 142
pre-eclampsia 208
pregnancy loss 145
diagnosis 141–147
very early 143–144
prevalence rate
screening tests 145
testing 145
venous thromboembolism
risk 100
thromboprophylaxis 99–106, 120
autoimmune hemolytic anemia 58
cesarean section 122–123
delivery 124
myeloproliferative disorders 238
venous thromboembolism
101
recommendations 103–104
thrombosis
antiphospholipid
syndrome 132–134
coagulation factor deficiencies
192
dysfibrinogenemia 192
myeloproliferative disorders 233
risk management 238
prosthetic heart valves 116
sickle cell disease 34
von Willebrand disease 182
see also deep vein thrombosis;
pulmonary thromboembolism;
venous thromboembolism
thrombotic endocarditis,
non-bacterial 134
thrombotic microangiopathies 218
thrombotic storm, catastrophic
antiphospholipid
syndrome 134
thrombotic thrombocytopenic
purpura (TTP) 218–225

Index

acquired 220, 222
previous idiopathic 220–221
acute presentation in
pregnancy 220
ADAMTS 13 219–221, 222
aspirin 222
clinical signs 219
complications 221
congenital 220, 222
diagnosis 219
epidemiology 219
factor VIII 219–220
HELLP differential diagnosis 214
low molecular weight heparin 222
pathology 219
plasma exchange 221, 222
pre-eclampsia differential
diagnosis 214
pregnancy 47
presentation during pregnancy 221
previous acquired
idiopathic 220–221
relapse risk 220–221, 222
treatment 221–222
von Willebrand factor 219–220
thyroid dysfunction, thalassemia 41
ticlopidine 120–121
tin-mesoporphyrin 87
tissue plasminogen activator
(t-PA) 8–10
total body irradiation 244
tranexamic acid
coagulation factor deficiencies 192
factor XI deficiency 191
hemophilia 188
immune/idiopathic
thrombocytopenic purpura 52
obstetric hemorrhage 167–168
platelet function inherited
disorders 183
von Willebrand disease 181–182
transferrin, iron deficiency 14
transferrin receptors 14–15
trophoblasts, pre-eclampsia 207–208
Tuohy needle 125
tyrosine kinase inhibitors 247
ultrasonography
antiphospholipid
syndrome 137–138
see also uterine artery, Doppler
screening
ultrasound-guided direct intravascular
transfusion 79

ultrasound-guided fetal blood
sampling 66, 67, 67–68
ursodeoxycholic acid 223
uterine artery
abnormal flow 218–219
Doppler screening 208, 218–219
myeloproliferative disorders 238
uterine artery embolization 156
complications 174
elective management 172–173
emergency postpartum 171–172
obstetric hemorrhage 171
uterine compression, bimanual 155
uterus
abdomino-pelvic irradiation 244
total body irradiation 244
vascular endothelial growth factor
(VEGF) inhibitors 205
venesection, myeloproliferative
disorders 237
venous compression stockings
(TEDS)
delivery 124
myeloproliferative disorders 236,
238
venous thromboembolism 101–102
venous thromboembolism 91–99
antiphospholipid syndrome 133
aspirin treatment 102
D-dimer testing
delivery management
diagnosis
radiation exposure
epidemiology
heparin treatment 102–103
ante-natal management 104–105
delivery
maintenance
monitoring
post-natal management 105
postpartum
management 101–105
ante-natal 104, 104–105
non-pharmacological 101–102
pharmacological 102–104
post-natal 105, 105
thromboprophylaxis 101
maternal obesity 100–101, 120
pathogenesis 99–101
pneumatic compression
boots 101–102
post-natal management 105, 105
risk
assessment 120

with autoimmune hemolytic
anemia 58
with thrombophilia
risk factors 100, 99–100, 100,
101
ante-natal 104
signs/symptoms
thrombophilia 100
testing
thromboprophylaxis 101
ante-natal 104
management strategy 101
post-natal management 105
recommendations 103–104
treatment
initial
maintenance
venous compression
stockings 101–102
Virchow’s triad 99
warfarin treatment 103
vertebral canal hematoma risk 122
Virchow’s triad, venous
thromboembolism 99
vitamin B see folic acid
vitamin B12 deficiency 23, 25–26
folic acid supplementation
contraindication 24
homocysteine levels 23
treatment 24
vitamin K
coagulation factor deficiencies
192
hemophilia 188
warfarin
in labor 124
in pregnancy 112–113
vitamin K antagonists
von Willebrand disease
176–182
analgesia 181
aspirin 181
classification 177
clinical features 176
coagulation factor replacement
179–180
complications
factor replacement 182
pregnancy 178
DDAVP 179, 182
contraindications 180
factor VIII 177
hormonal influences in
pregnancy 177–178
inhibitor formation 182
laboratory evaluation 177

265

Index

von Willebrand disease (cont.)
management 178
ante-natal 179, 179
intrapartum 181, 180–181
neonates 182, 183
postpartum 181, 181
pre-pregnancy 178
maternal bleeding 178
neonates 182, 183
plasma products 182
complications 182
pregnancy outcomes 178
prevalence 176–177
subtypes 177
thrombosis 182
tranexamic acid 181–182
transfusion transmitted
infections 182
von Willebrand factor (vWF)
activity 176

266

alloantibody formation 182
ante-natal management 179
antigen assays 177
deficiency 176
factor VIII ratio 196
gene mutations 176–177
hormonal influences in
pregnancy 177–178
multimers 177
postpartum levels 181
replacement 179–180
thrombotic thrombocytopenic
purpura 219–220
V/Q scan
pulmonary thromboembolism
radiation exposure

fetal effects 103, 112
labor 124
pregnancy outcome 112–113
dosage 113
prosthetic heart valves 112
replacement with
heparin 115–116
replacement with heparin
antiphospholipid syndrome 137
for delivery 123
prosthetic heart valves 115–116
thromboprophylaxis 101
venous thromboembolism
treatment 103
warfarin embryopathy 103, 112
dosage 113
weak D variants 85

warfarin
antiphospholipid syndrome 137

white blood cells 4

Source Exif Data:

File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.4
Linearized                      : Yes
Page Mode                       : UseOutlines
XMP Toolkit                     : 3.1-702
Producer                        : Acrobat Distiller 7.0.5 for Macintosh
Modify Date                     : 2010:09:24 11:12:39+07:00
Create Date                     : 2010:03:09 20:52:33+05:30
Creator Tool                    : dvips(k) 5.95a Copyright 2005 Radical Eye Software
Metadata Date                   : 2010:09:24 11:12:39+07:00
Document ID                     : uuid:adfb43b9-2b8f-11df-8860-0011247f73b2
Instance ID                     : uuid:e046da93-498f-4818-96b7-8df3d3835d19
Format                          : application/pdf
Creator                         : Sue Pavord, Beverley Hunt
Title                           : The Obstetric Hematology Manual.
Subject                         : 0521865646, 9780521865647
Has XFA                         : No
Page Count                      : 278
Page Layout                     : SinglePage
Author                          : Sue Pavord, Beverley Hunt
Keywords                        : 0521865646, 9780521865647

EXIF Metadata provided by EXIF.tools

The Obstetric Hematology Manual. Manual

Navigation menu

Versions of this User Manual:

Views

Navigation