Antidepressant use during pregnancy and risk of stillbirth: a systematic review and meta-analysis

Introduction
To evaluate the current evidence estimating the association between antidepressant use during pregnancy and stillbirth.

Search Strategy
MEDLINE, EMBASE, and PsychINFO for studies investigating antidepressant use during pregnancy and risk of stillbirth, from inception until 21 January 2022.

Selection Criteria
Studies including pregnant women exposed to antidepressants during pregnancy investigating stillbirth were eligible, compared with either unexposed, indicated pregnant women or unexposed women in the general obstetric population.

Data Collection and Analysis
Data extraction and quality assessment were performed by two authors independently. Meta-analysis was used to generate pooled-effect estimates, and the ROBINS-I tool was used to assess risk of bias for individual studies.

Main Results
Seventeen studies were eligible. Although estimates from meta-analysis models suggest a small increased risk of stillbirth, summary effect estimate 1.19 (95% confidence interval [CI] 1.06, 1.34) between those individuals taking antidepressants during pregnancy and all other pregnant women, confounding control is likely inadequate. The risk of bias assessment showed most studies were low quality, with no studies scoring low risk; in a meta-analysis of studies with moderate risk of bias (n = 2), no association was noted, summary effect estimate 1.17 (95% CI 0.97, 1.41). Only six studies adjusted for confounding by indication, the findings of which were summarised narratively.

Conclusions
Although the overall meta-analysis found a small association between antidepressant use during pregnancy and stillbirth, this result was likely due to the overall low quality of studies included and by confounding in the underlying studies. Future studies must adequately address potential confounding by indication.