Association of Immediate Postpartum Etonogestrel Implant Insertion and Venous Thromboembolism

Abstract

OBJECTIVE: To estimate the rate of readmissions for postpartum venous thromboembolism (VTE) during the first 30 days postdelivery between women with and without the immediate postpartum insertion of the etonogestrel contraceptive implant. METHODS: The Nationwide Readmissions Database from 2016 was used to identify women with a singleton delivery, immediate postpartum insertion of the etonogestrel contraceptive implant, and readmission for VTE within 30 days of discharge. Those with a prior history of VTE or anticoagulant therapy were excluded. These women were compared with the number of women readmitted within 30 days for VTE who did not have the contraceptive implant placed during delivery admission. RESULTS: Of 3,387,120 deliveries, 8,369 women underwent etonogestrel contraceptive implant placement during the delivery admission. There was no difference identified in the rate of readmission for VTE between exposed and unexposed women. Of these, seven had received a postpartum etonogestrel contraceptive implant (0.85/1,000; 95% CI 0.22-1.45/1,000 deliveries), compared with 1,192 without an etonogestrel contraceptive implant (0.35/1,000; 95% CI 0.33-0.37/1,000 deliveries); odds ratio (OR) 2.41; 95% CI 0.58-9.89. The rates of diabetes, thrombophilia, systemic lupus erythematosus, and cesarean birth did not differ between groups. Women who underwent etonogestrel contraceptive implant placement were younger and were more likely to have government-sponsored health insurance, a smoking history, hypertension, peripartum infection, or postpartum hemorrhage than women who did not receive an etonogestrel contraceptive implant (P<.001). After adjusting for these confounders, there remained no difference in rates of VTE, adjusted OR 1.81; 95% CI 0.44-7.45. CONCLUSION: The immediate postpartum placement of the etonogestrel contraceptive implant was not associated with an increased rate of VTE; however, our sample size was underpowered to determine no difference.