Abstract
BACKGROUND: Several studies have evaluated maternal first trimester paroxetine use and the prevalence of congenital defects, particularly cardiac defects. To synthesize current epidemiologic information, a meta‐analysis was conducted. METHODS: A systematic literature search was conducted for original research published from January 1, 1992, through September 30, 2008. Results were extracted using predefined criteria, and authors were contacted for additional information when necessary. Compiled results were evaluated for funnel plot asymmetry, heterogeneity, and study characteristic associations. Where appropriate, fixed‐effect summary estimates were calculated and sensitivity analyses performed. RESULTS: Twenty reports (11 including results for aggregated congenital and combined cardiac defects, six for aggregated congenital defects only, and three for combined cardiac defects only) met prespecified inclusion criteria. There was little evidence of funnel plot asymmetry or overall heterogeneity. Summary estimates were produced for combined cardiac defects (prevalence odds ratio [POR], 1.46; 95% confidence interval [CI], 1.17–1.82) and aggregated congenital defects (POR, 1.24; 95% CI, 1.08–1.43) and first trimester paroxetine use. Some study characteristics may be associated with differential POR estimates for paroxetine and either combined cardiac or aggregated congenital defects. CONCLUSIONS: This meta‐analysis found little evidence of publication bias or overall statistical heterogeneity and only weak evidence of associations with some study characteristics. Although subject to limitations, the summary estimate indicates an increased prevalence of combined cardiac defects with first trimester paroxetine use. The summary estimate also indicates an increased prevalence of aggregated congenital defects with paroxetine; however, this association may be explained, in part, by the increased prevalence of combined cardiac defects. Birth Defects Research (Part A), 2010. © 2009 Wiley‐Liss, Inc.