Real-world safety profile of bisoprolol: signal detection and demographic stratification using the FAERS database.

Abstract

Bisoprolol, a selective β₁-blocker, is widely prescribed for cardiovascular disease. Real-world pharmacovigilance can clarify adverse drug reactions (ADRs) across demographic strata and identify signals not fully characterized in trials. We analyzed bisoprolol-associated reports in the FDA Adverse Event Reporting System. Disproportionality was assessed using complementary algorithms-reporting odds ratio (ROR), proportional reporting ratio (PRR), information component (IC with IC025), and empirical Bayes geometric mean (EBGM with EBGM05). To stabilize sparse strata, we required a ≥ 3 and multi-algorithm concordance (EBGM05 > 2, IC025 > 0, lower95_ROR > 1). Multiple testing was controlled using Benjamini-Hochberg FDR and Bonferroni adjustments. Signals were summarized overall and stratified by sex and age (< 40, 40-80, ≥ 80 years). Class-expected cardiovascular ADRs showed robust signals, including bradycardia, sinus bradycardia, bradyarrhythmia, atrioventricular block, hypotension, syncope/presyncope (all meeting stability and multiplicity criteria). Several hypothesis-generating signals with strong effect sizes were detected: cardiospasm (ROR≈285; EBGM05≈178), palmoplantar keratoderma (ROR≈217; EBGM05≈137), and hyperkalemia (ROR≈16.7; EBGM05≈13). Sex- and age-stratified analyses revealed clinically relevant patterns: in younger patients (< 40), bradyarrhythmia-type signals were most pronounced; in middle-aged adults (40-80), cardiospasm and palmoplantar keratoderma ranked among the top signals; in older adults (≥ 80), conduction-slowing events remained prominent. Overall reporting skewed female (≈55%) and older age. A qualitative EudraVigilance cross-check identified parallel case clusters for cardiospasm and palmoplantar keratoderma, corroborating cross-jurisdictional presence. FAERS data reaffirm bisoprolol's known bradycardic and hypotensive risks and highlight novel, biologically plausible signals-particularly cardiospasm and palmoplantar keratoderma-that vary by demographic subgroup. These findings support targeted clinical vigilance (heart-rate/conduction, electrolytes, dysphagia/chest pain, palmoplantar skin changes) and motivate confirmatory studies (prospective cohorts, nested case-controls). While robust to multiplicity control, signals remain hypothesis-generating given spontaneous-reporting biases; cautious interpretation and validation are warranted.