Ondansetron Use in Pregnancy: Safety, Efficacy, and Guideline Recommendations

Treatment Algorithm

First‑line antiemetics for nausea and vomiting of pregnancy (NVP) or hyperemesis gravidarum (HG) should be antihistamines (e.g., promethazine, cyclizine), phenothiazines (e.g., prochlorperazine), or pyridoxine‑doxylamine combinations. The recommendation is based on BJOG guideline evidence. 1
Ondansetron is recommended as second‑line therapy when first‑line agents fail, with a safety profile supported by high‑quality BJOG data. This positions ondansetron after failure of antihistamines, phenothiazines, or pyridoxine‑doxylamine. 2
The stepwise treatment approach is driven primarily by safety considerations rather than proven superiority of any antiemetic class; no robust data demonstrate a clear efficacy advantage of one class over another. This reflects the consensus of BJOG analyses. 4

Robust evidence from BJOG shows no increased risk of overall congenital malformations with ondansetron exposure during pregnancy. Large‑scale cohort and meta‑analytic data underpin this conclusion. 1
More than 97 000 first‑trimester exposures have been examined across twelve studies, including a meta‑analysis, without detecting a rise in malformation rates. This extensive dataset provides high‑level confidence. 4
The largest individual study evaluated 88 000 first‑trimester exposures within a cohort of 1.8 million pregnancies and found no excess malformations. The sheer size confers strong statistical power. 2
Across the majority of studies—including disease‑matched and unexposed control groups—no increased malformation risk was identified. Consistency across diverse populations reinforces the safety signal. 1
The UK Teratology Information Service systematic review concluded that current data do not support an association between first‑trimester ondansetron use and a higher overall malformation rate. This expert review adds further credibility. 4

A single cohort study reported a modest increase in cardiac defects (adjusted OR 1.43, 95 % CI 1.28–1.61). However, this finding originates from a smaller dataset. 2
A larger, methodologically rigorous study of 88 467 first‑trimester exposures found no significant association after adjustment for confounders (adjusted RR 0.99, 95 % CI 0.93–1.06). The superior statistical power and comprehensive confounder control make this result the preferred evidence. 2

Ondansetron exposure is associated with a small relative increase in orofacial clefting (adjusted RR 1.24, 95 % CI 1.03–1.48). The relative risk is modest. 2
In absolute terms, this translates to approximately 3 additional cleft cases per 10 000 births (14 / 10 000 with ondansetron vs 11 / 10 000 without). The background risk is low, making the absolute increase minimal. 1

Studies that compare women with severe NVP/HG taking ondansetron to healthy, asymptomatic women are subject to confounding by indication; the underlying disease severity, not the drug, may drive observed outcomes. This limitation is highlighted in BJOG analyses. 4
When women with HG receiving ondansetron are compared to HG women using alternative antiemetics, no increase in major malformations is observed. This disease‑matched comparison mitigates confounding. 4
A comparative study of HG patients showed higher pregnancy‑termination rates among those not receiving ondansetron and higher live‑birth rates among those who did, suggesting a benefit of treatment in this high‑risk group. The finding underscores the importance of appropriate comparator groups. 4

The European Medicines Agency (EMA) issued warnings advising avoidance of ondansetron in the first trimester based on conflicting epidemiological data. This reflects regulatory caution despite mixed evidence. 4
The UK Medicines and Healthcare products Regulatory Agency (MHRA) did not issue a comparable warning, indicating divergent regulatory interpretations. This nuance is captured in BJOG reports. 4
The 2024 BJOG guideline explicitly states that ondansetron “should not be discouraged if first‑line antiemetics are ineffective.” This guideline endorsement supports its use when clinically indicated. 1