AIMS: Recent studies have suggested a possible association between maternal use of selective serotonin reuptake inhibitors (SSRIs) in early pregnancy and cardiovascular anomalies. The aim of the present study was to evaluate the teratogenic risk of paroxetine and fluoxetine. METHODS: This multicentre, prospective, controlled study evaluated the rate of major congenital anomalies after first-trimester gestational exposure to paroxetine, fluoxetine or nonteratogens. RESULTS: We followed up 410 paroxetine, 314 fluoxetine first-trimester exposed pregnancies and 1467 controls. After exclusion of genetic and cytogenetic anomalies, there was a higher rate of major anomalies in the SSRI groups compared with the controls [paroxetine 18/348 (5.2%), fluoxetine 12/253 (4.7%) and controls 34/1359 (2.5%)]. The main risk applied to cardiovascular anomalies [paroxetine 7/348 (2.0%), crude odds ratio (OR) 3.47, 95% confidence interval (CI) 1.13, 10.58; fluoxetine 7/253 (2.8%), crude OR, 4.81 95% CI 1.56, 14.71; and controls 8/1359 (0.6%)]. On logistic regression analysis only cigarette smoking of >or=10 cigarettes day(-1) and fluoxetine exposure were significant variables for cardiovascular anomalies. The adjusted ORs for paroxetine and fluoxetine were 2.66 (95% CI 0.80, 8.90) and 4.47 (95% CI 1.31, 15.27), respectively. CONCLUSION: This study suggests a possible association between cardiovascular anomalies and first-trimester exposure to fluoxetine.