|Home | About | Journals | Submit | Contact Us | Français|
Pulmonary embolism is the leading cause of maternal mortality in developed countries and accounts for 20% of pregnancy related deaths in the United States.1 2 The risk of pulmonary embolism and deep vein thrombosis, collectively known as venous thromboembolism, is increased during pregnancy and is further increased by the presence of inherited or acquired thrombophilias. We summarise the epidemiology and diagnosis of venous thromboembolism in pregnancy and discuss the anticoagulant management of women with inherited thrombophilia who are at risk of, or who develop, venous thromboembolism during pregnancy and the postpartum period.
A 30 year old woman, a known heterozygote for the factor V Leiden mutation, presents at eight weeks' gestation in her first pregnancy wondering whether she should receive prophylactic anticoagulation to prevent recurrent venous thrombosis during pregnancy. Several years ago she developed a deep vein thrombosis of the left leg after an ankle fracture and prolonged immobilisation and was found to have the factor V Leiden mutation. The deep vein thrombosis was treated with anticoagulants for three months, and the woman has had no recurrent thromboembolic events since stopping warfarin. Her mother developed a deep vein thrombosis after surgery but did not undergo testing for thrombophilia.
We searched Medline and the Cochrane database of systematic reviews for studies evaluating the epidemiology, diagnosis, prevention, and treatment of venous thromboembolism during pregnancy and the postpartum period in women with inherited thrombophilia, using the key words “venous thrombosis”, “deep vein thrombosis”, “pulmonary embolism”, “pregnancy (complications)”, “thrombophilia”, and “anticoagulants.”
Venous thromboembolism occurs in 10 per 100000 women of childbearing age and affects 100 per 100000 pregnancies.3 Inherited thrombophilia is present in 30%-50% of women with pregnancy associated venous thromboembolism,3w1 with factor V Leiden being the most frequently identified inherited thrombophilia in the white population (table 11).).
Inherited thrombophilia may be associated with an increased risk of adverse events during pregnancy, including early pregnancy loss (odds ratio, range 1.40-6.25), late pregnancy loss (1.31-20.09), pre-eclampsia (1.37-3.49), placental abruption (1.42-7.71), and intrauterine growth restriction (1.24-2.92).w8 The risk varies depending on the specific thrombophilia. The risk is greater in women with more than one thrombophilic polymorphism.4
Women with no history of venous thromboembolism who show an increased risk of the condition include those with known thrombophilia and those with a strong family history of venous thromboembolism.5
The role of thrombophilia testing in pregnancy is controversial, and recommendations for testing are based on expert opinion; the lowest quality of evidence.6 In general, testing pregnant women without risk factors for thrombophilia is not indicated and the decision to carry out testing is individualised. Testing may be considered for women with a personal or family history (first degree relative before the age of 50) of venous thromboembolism, and possibly in women with a history of certain obstetric complications.
In considering whether to recommend thromboprophylaxis to women with thrombophilia, the risks of venous thromboembolism and of bleeding and the potential effects of venous thromboembolism and anticoagulant prophylaxis on the pregnancy should all be taken into account.w9 Currently no published randomised trials examine antithrombotic prophylaxis in pregnancy. Recommendations are therefore based on descriptive studies.
Options for anticoagulant prophylaxis in the antepartum period include low molecular weight heparin and unfractionated heparin, given subcutaneously in prophylactic doses (table 22).). Optimal dosing is controversial and some clinicians prefer to use intermediate or full therapeutic doses of either agent in patients who are considered to be at very high risk of venous thromboembolism. When prophylactic doses are used, laboratory monitoring of anticoagulant intensity is sometimes carried out but has not been shown to be beneficial.7
Two major factors influence the risk of venous thromboembolism during pregnancy in women with thrombophilia: the type of thrombophilia and the circumstances of any previous venous thromboembolism—that is, provoked or unprovoked. Thrombophilias that are associated with a high risk of venous thromboembolism during pregnancy include antithrombin deficiency,w1 w10 w11 protein C or S deficiency,w10 w11 compound heterozygosity for factor V Leiden and prothrombin gene mutation (G20210A)8w10 or other combinations of thrombophilia, and homozygosity for these conditions.8 9 10w12 Lower risk thrombophilias include heterozygosity for factor V Leiden or the prothrombin gene mutation (table 11).11w1 w10-w12
Women without inherited thrombophilia and those whose previous episode of venous thromboembolism was associated with a temporary risk factor (that is, provoked) have a low risk of recurrence during pregnancy,w13 w14 and antepartum prophylaxis is generally not warranted. Other factors, such as a family history of venous thromboembolism in pregnancy, may provide support for thromboprophylaxis even in the absence of a detectable thrombophilia.
Women with high risk thrombophilias have a high risk of pregnancy induced venous thromboembolism and should receive antepartum prophylaxis.w9 Women with lower risk thrombophilias who have no history of venous thromboembolism have a low risk of venous thromboembolism in pregnancy (relative risk 0%, 95% confidence interval 0% to 2.7%)12w15 and antepartum prophylaxis is not routinely recommended.
Among women with lower risk thrombophilias and previous venous thromboembolism, the role of antepartum prophylaxis remains controversial. A retrospective study of pregnant women with previous venous thromboembolism found that the rate of recurrence during pregnancy was 7.5%, irrespective of whether the first episode was unprovoked, pregnancy related, or occurred during use of oral contraceptives.w14 By contrast no recurrences occurred among women with a previous provoked venous thromboembolism. Consensus guidelines recommend that women with unprovoked or oestrogen related venous thromboembolism receive antepartum prophylaxis,w9 whereas prophylaxis may not be necessary in women with provoked venous thromboembolism. In a small prospective study of 23 women with thrombophilia who became pregnant after an episode of venous thromboembolism, the rate of recurrence was 8.3% (1 of 12 pregnancies) in women treated with prophylactic heparin compared with 25% (7 of 28 pregnancies) in women who did not receive prophylaxis.w15 The recurrences occurred mainly in women with antithrombin deficiency or who were homozygous for factor V Leiden.
Bleeding complications are minimally increased or not increased when using prophylactic dose low molecular weight heparinw16 or unfractionated heparin. Unfractionated heparin may be associated with heparin induced thrombocytopenia and osteoporosis; both of which are less common with low molecular weight heparin.
Low molecular weight heparin and unfractionated heparin do not cross the placenta and are considered safe for use during pregnancy.w16
Women who are at risk for venous thromboembolism during pregnancy should be educated about the signs and symptoms of the condition and should seek prompt medical attention if these develop.w9 Women who are of childbearing age and are identified with an inherited thrombophilia or have a history of venous thromboembolism should be educated about the potential need for antithrombotic prophylaxis during pregnancy. Consultation with a doctor with expertise in maternal-fetal medicine is recommended if a pregnancy is planned.
Women who develop venous thromboembolism during pregnancy are treated with therapeutic doses of low molecular weight heparin or unfractionated heparin for at least the same duration as if the event had occurred outside pregnancy. Although level 1 evidence is lacking, it is logical to presume that whatever aspect of pregnancy that potentially contributed to the occurrence of thrombosis will persist throughout pregnancy. For this reason, anticoagulation should be maintained for the duration of pregnancy and for at least six weeks post partum. Less commonly, warfarin may be used after the first trimester. Treatment should be started as soon as possible and in consultation with a specialist (table 22).). Therapeutic doses of low molecular weight heparin are based on body weight, and should be adjusted accordingly as pregnancy progresses. Anti-Xa level monitoring of low molecular weight heparin may be carried out to ensure that therapeutic levels of anticoagulation are achieved. Therapeutic dose unfractionated heparin is given as a twice daily subcutaneous injection and requires laboratory monitoring to achieve a mid-interval activated partial thromboplastin time 2.0-2.5 times that of the baseline time.
The frequency of major bleeding associated with therapeutic dose unfractionated heparin or low molecular weight heparin during pregnancy is about 2%. Bleeding is seen most commonly in relation to delivery, manifesting as wound haemorrhage or pelvic haematoma after caesarean section, or bleeding of the vulva or vaginal wall after vaginal delivery.w16 Vitamin K antagonists (for example, warfarin) cross the placenta and can cause fetal microcephaly, nasal hypoplasia, and stippled epiphyses when given between the sixth and 12th weeks of gestation.13 They can be associated with fetal haemorrhage. Vitamin K antagonists are thus generally avoided in pregnancy since low molecular weight heparin is an effective and safe alternative for prophylaxis and treatment of venous thromboembolism in pregnancy.w16 Use of warfarin in pregnancy is generally limited to the second and third trimesters in women with prosthetic heart valves, in whom there is uncertainty about the efficacy of low molecular weight heparin.
In women who have been treated with anticoagulant drugs, vaginal delivery and caesarean section increase the risk of bleeding, and use of regional anaesthesia is associated with a risk of epidural haematoma. Thus anticoagulants should be discontinued for a sufficient period before delivery to allow their effect to wear off. A safe interval between stopping anticoagulants and delivery can be achieved by scheduling elective induction of labour or caesarean section at least 12 hours after prophylactic dose low molecular weight heparin or 24 hours after therapeutic dose low molecular weight heparin.6w9
Restarting anticoagulation using low molecular weight heparin or unfractionated heparin may be considered within 12 hours of delivery in women with an uncomplicated delivery and in whom haemostasis is secure.6w9 In the postpartum period, low molecular weight heparin is usually given concurrently with warfarin until the international normalised ratio reaches at least 2.0 on two consecutive days.
Since the absolute risk of venous thromboembolism in the postpartum period is higher than during the pregnancy and because bleeding is less of a concern after labour and delivery, postpartum thromboprophylaxis is recommended for women with known inherited thrombophilia and for women with a history of venous thromboembolism. Warfarin is usually given for 6-8 weeks, targeting an international normalised ratio of 2.0-3.0. Alternatively, therapeutic dose unfractionated heparin or low molecular weight heparin may be given. Although these agents are detectable in breast milk, all are safe for use during breast feeding because warfarin metabolites are inactive and heparin in not absorbed through the gastrointestinal tract.14 15
*The left leg is affected in up to 90% of patients with deep vein thrombosis during pregnancy16
We thank Barry Walters for his critical review of the manuscript.
Contributors: WL prepared the first draft of the manuscript. WL, JWE, and JSG jointly revised and finalised the manuscript. JWE is guarantor.
Competing interests: None declared.
Provenance and peer review: Commissioned; externally peer reviewed.