Abstract
INTRODUCTION: Overactive bladder (OAB) drugs are widely prescribed, yet the occurrence of atrial fibrillation (AF) after treatment initiation remains poorly characterized. METHODS: We evaluated reports of AF associated with OAB medications using two spontaneous reporting systems (SRSs): the Japanese Adverse Drug Event Report (JADER) database and the U.S. FDA Adverse Event Reporting System (FAERS). We screened eight agents and assessed signals using three disproportionality metrics: the reporting odds ratio (ROR), proportional reporting ratio (PRR), and Bayesian confidence propagation neural network (BCPNN). For drugs showing signals in both databases, we conducted stratified analyses by sex, age, and number of concomitant medications, and evaluated time-to-onset (TTO) using Weibull modeling. RESULTS: Consistent AF signals were identified for solifenacin succinate and mirabegron, whereas other agents did not meet the prespecified criteria. Solifenacin met the criteria in women and older adults in both JADER and FAERS. Mirabegron met the criteria across multiple strata in both datasets, indicating cross-stratum reproducibility. TTO was right-skewed, with most reports occurring within one year of initiation. Exploratory Weibull modeling, based on limited numbers of date-complete reports, suggested a wear-out pattern for solifenacin in JADER and an early pattern in FAERS, while mirabegron showed a random pattern in JADER and an early pattern in FAERS. These failure-type patterns should therefore be interpreted cautiously. DISCUSSION: These findings are hypothesis-generating, given the limitations of SRSs, such as underreporting, missing dates, and unknown exposure-and they reflect reporting patterns rather than causal risk. They outline strata and early treatment periods that may warrant clinical attention and help prioritize pharmacovigilance and targeted hypothesis-driven evaluation in routine OAB care.