Venous thromboembolism is an uncommon occurrence in pregnancy but can be fatal in some cases. Women with a previous history of deep venous thrombosis (DVT) or pulmonary embolism (PE) are considered to be at a higher risk of recurrence during pregnancy. The absolute risk is unknown but is reported in the literature to range from zero to 13 percent. Because of this presumed risk, some experts recommend that women with a history of a thromboembolic event receive prenatal and postpartum anticoagulant prophylaxis. However, warfarin is contraindicated in pregnancy, and heparin therapy is expensive and logistically difficult, and may cause bleeding problems and osteoporosis. Brill-Edwards and colleagues performed a prospective cohort study to evaluate the safety of withholding antepartum heparin prophylaxis in women with a history of venous thromboembolism.

Women eligible for the study had a history of a single episode of DVT of the leg or PE, and presented for prenatal care at fewer than 20 weeks' gestational age. Excluded were women with a known history of protein S, protein C or antithrombin deficiency, and women with the presence of antiphospholipid antibodies, factor V Leiden mutation or the prothrombin G20210A mutation. All eligible women had bilateral compression ultrasonography of both legs on entry into the study. In approximately two thirds of the women, blood samples were obtained to test for an underlying thrombotic disorder.

Regular prenatal follow-up care was performed until delivery. Any patient who presented with signs or symptoms of a DVT or PE before her due date underwent prompt diagnostic testing to rule out either condition. Within 24 hours after delivery, all women were given 5,000 to 7,500 U of subcutaneous heparin twice daily until discharge. They were sent home on warfarin therapy for four to six weeks, and a target International Normalized Ratio of 2.0 to 3.0 was maintained. The study organizers planned to enroll 250 patients but stopped after 125 patients were enrolled because a single interim analysis of the data ruled out a true recurrence rate of 10 percent.

Among the women enrolled in the study, there were 92 previous episodes of DVT and 39 previous cases of PE. During the prenatal period, there were three cases of recurrent venous thromboembolism. This finding corresponds to a recurrence rate of about 4 percent per patient-year. There were two cases of DVT, one at 28 and one at 29 weeks' gestation, and one case of PE at nine weeks' gestation. These patients were given low-molecular-weight heparin for the remainder of their pregnancies. There were also three cases of postpartum DVT.

Forty-eight of the 95 women (51 percent) from whom blood samples were obtained at baseline had at least one abnormality related to thromboembolic disease. Most common was a low plasma concentration of protein S (30 patients) and factor V Leiden mutation (11 patients). All three women who had an postpartum event had at least one abnormal laboratory result or a previous idiopathic event None of the 44 women whose laboratory evaluations were normal and whose prepregnancy thrombotic event was associated with a temporary risk factor (such as surgery) had a recurrent event during pregnancy.

The authors conclude that the risk of recurrent venous thromboembolism in pregnancy is quite low, and that routine prophylaxis with heparin is not necessary. However, they recommend that postpartum therapy with warfarin be employed in these women. They also note that all pregnant patients with a history of venous thromboembolism should undergo evaluation for thrombophilia, especially factor V Leiden.