Ondansetron Use for Women With Hyperemesis Gravidarum and Risks of Spontaneous Abortion and Stillbirth: A Retrospective Cohort Study

BACKGROUND: Ondansetron, a selective 5-hydroxytryptamine type 3 serotonin receptor antagonist, is the most prescribed medication to treat hyperemesis gravidarum in the United States. However, certain characteristics of ondansetron, such as QT interval prolongation and its ability to cross the placenta, may suggest possible fetal harm following maternal exposure. OBJECTIVE: To assess the association between ondansetron use and risk for spontaneous abortion and stillbirth in pregnant women affected by hyperemesis gravidarum. DESIGN: Retrospective cohort study. SETTING: Electronic healthcare databases from over 500 healthcare facilities across the United States between 2001 and 2016. PARTICIPANTS: Pregnant women affected by hyperemesis gravidarum are eligible for evaluation of spontaneous abortion and stillbirth, respectively. MEASUREMENTS: The hazard ratios (HRs) and 95% confidence intervals (CIs) for spontaneous abortion and stillbirth were estimated for women exposed to ondansetron versus women not exposed to an antiemetic in a time-dependent manner, using a Cox proportional-hazards regression model. Dose-response relationships between ondansetron use and respective outcomes were assessed, with ondansetron doses categorized as no use, 1-4, 5-15, and ≥ 16 mg. We conducted post hoc analyses to compare the risks of spontaneous abortion and stillbirth among ondansetron users with those exposed to other antiemetics, as well as with those exposed only to the two most frequently used antiemetics, promethazine and metoclopramide. RESULTS: There were 34 273 and 50 862 pregnant women affected by hyperemesis gravidarum, eligible for evaluation of spontaneous abortion and stillbirth, respectively. Following propensity score matching on a 1:1 ratio between women solely exposed to ondansetron and women not exposed to antiemetics, a total of 10 238 and 13 585 pairs were included in the main analysis of spontaneous abortion and stillbirth, respectively. Exposure to ondansetron was associated with a decreased risk for spontaneous abortion compared with women not exposed to an antiemetic (adjusted HR 0.82, 95% CI 0.73, 0.91) under the time-dependent Cox model. There was no significant dose-dependent relationship between spontaneous abortion and ondansetron exposure. There was no association between ondansetron exposure and stillbirth. The adjusted HR for stillbirth with time-dependent defined exposure was 1.34 (95% CI 0.87-2.08) in mild cases and 0.72 (95% CI 0.23-2.26) in severe cases. A reduction in spontaneous abortion risk with ondansetron users compared to other antiemetic users was also observed (aHR 0.85, 95% CI 0.73-0.98). CONCLUSION: Obstetric use of ondansetron suggested a decreased risk of spontaneous abortion, with no increased risk of stillbirth among pregnant women affected by hyperemesis gravidarum.