Practice Patterns and Clinical Outcomes of Thromboprophylaxis during Pregnancy

Background: Venous thromboembolism (VTE) is the fifth leading cause of pregnancy-related mortality in the United States. Thromboprophylaxis is recommended for those at highest risk (>2% antepartum and >1% postpartum), but practice patterns vary. We aimed to describe thromboprophylaxis practices and outcomes among pregnant women cared for at Johns Hopkins Health System.

Methods: We performed an observational, retrospective cohort study of consecutive pregnant patients treated with pharmacologic thromboprophylaxis at Johns Hopkins Health System from 2011 to 2022. Patients anticoagulated for other indications, including recurrent miscarriage, were excluded. Those with obstetric antiphospholipid antibody syndrome were included given high risk for thrombosis. VTE and bleeding events were identified by chart review. The primary efficacy outcome was imaging-confirmed VTE. The primary safety outcome was major bleeding using ISTH obstetric criteria.

Results: A total of 169 pregnancies were included in our analysis resulting in 158 (93.5%) deliveries including 157 (92.9%) live births and 127 (75.1%) full term births. Median maternal age was 33 years (IQR 29-36). Median BMI was 31.5 kg/m2 (IQR 27.4-36.6). Eighty-six (50.9%) women were non-Hispanic White, 62 (36.7%) were non-Hispanic Black, 12 (7.1%) were Hispanic, and 5 (3.0%) were Asian. Caesarean delivery occurred in 75 (44.4%) pregnancies.

The primary indication for thromboprophylaxis was a history of VTE in 134 (79.3%) pregnancies, of which 93 (69.4%) were provoked by hormonal therapy or pregnancy. The second most common indication was high-risk thrombophilia [27 (15.9%) of pregnancies]. Sixteen (9.5%) pregnancies were already on anticoagulation. Median gestational age at enoxaparin thromboprophylaxis initiation was 9 weeks (IQR 7-12). The majority started prophylactic intensity (101, 59.8%) followed by intermediate (37, 21.9%) or therapeutic intensity dosing (31, 18.3%). Of 158 deliveries, 117 (74.1%) were switched to unfractionated heparin (UFH) at approximately 36 weeks gestation. Postpartum, 151 of 158 (95.6%) were prescribed enoxaparin. Postpartum dosing was predominantly prophylactic intensity (91, 57.6%), followed by intermediate (32, 20.3%) or therapeutic intensity dosing (35, 22.2%).

Two (1.2%) antepartum and 7 (4.4%) postpartum VTE events occurred. All (100%) of the antepartum VTE events and 6 (85.7%) of the postpartum VTE events occurred in pregnancies with history of VTE. Four (57.1%) of the postpartum VTE events occurred after caesarean section deliveries, 2 (28.6%) occurred after vaginal delivery, and 1 (14.3%) occurred after fetal loss. One postpartum VTE event occurred on postpartum day 1 prior to thromboprophylaxis reinitiation. Otherwise, postpartum VTE events occurred at a median of 14 days (IQR 2-21) after delivery. Postpartum VTE events occurred in 5 (5.5%), 0 (0%), and 1 (2.9%) of pregnancies prescribed prophylactic, intermediate, and therapeutic intensity dosing respectively.

Neuraxial anesthesia was administered in 127 deliveries (80.4%). There were no epidural-related bleeding events. One (0.6%) antepartum and 9 (5.7%) postpartum major bleeds occurred. Eight (89%) of 9 postpartum major bleeds occurred within 24 hours of delivery, and of those 8 early major bleeds, 7 (87.5%) had concurrent obstetric reasons for major bleeding including uterine atony and retained placenta. Seven (77.8%) of the postpartum major bleeds occurred in those with caesarean deliveries. Reversal of anticoagulation was not used in any of the major bleed events.

Conclusion: In our retrospective cohort of 169 pregnancies at risk for VTE, thromboprophylaxis was associated with a low prevalence of antepartum major bleeding and thrombosis. VTE tended to occur in the postpartum setting among those with a prior history of VTE. There was high utilization of epidural anesthesia despite antepartum anticoagulation. Postpartum hemorrhage most often occurred within the first 24 hours of delivery, in those with caesarean deliveries, and in those with concomitant obstetrical reasons for major bleeding. Our data provides additional real-world evidence in a diverse cohort that thromboprophylaxis during pregnancy is well-tolerated and does not preclude use of neuraxial anesthesia. Analysis is ongoing on the clinical outcomes of those pregnancies on UFH and those on enoxaparin in late gestation and prior to delivery.