OBJECTIVE To assess prospectively pregnancy outcome associated with first-trimester exposure to alprazolam in order to monitor for early signals of potential drug-related risk to the fetus. METHODS Reports of first-trimester exposure to alprazolam from patients, physicians, or pharmacists were registered. Information regarding drug exposure, risk factors, and pregnancy outcome was requested at 4.5 and 9 months after registration. RESULTS From June 1982 through December 1990, 542 pregnancies had been registered. Follow-up data were obtained on 411 registrants. There were 13 live births with congenital anomalies, 263 live births with no congenital anomalies, 47 spontaneous fetal losses, and 88 elective abortions. CONCLUSIONS No pattern of defects or excess of defects or spontaneous abortions was apparent. Although these findings may provide some reassurance in counseling women following inadvertent prenatal exposure, the cases represent a sample of insufficient size for reaching reliable and definitive conclusions about the safety of alprazolam for pregnant women and their developing fetuses. The cases accumulated in this report represent a sample of insufficient size to confirm or refute previous reports of specific congenital anomalies linked to exposure to benzodiazepines.