Bisphosphonates and atrial fibrillation risk: a final word

Abstract Background Bisphosphonates (BPs) are widely prescribed drugs that decrease bone fracture risk in osteoporosis patients. Nevertheless, the class has been associated with a plethora of adverse effects, including incidental atrial fibrillation (AF). This epidemiologic link has, however, been met with skepticism by some authors. Purpose To perform a meta-analysis aimed at ascertaining the extent to which BPs might increase the odds of AF. Methods We systematically searched MEDLINE, Embase, Web of Science, Cochrane Library and Google Scholar, from inception to the first of March, 2021, for randomized controlled trials comparing oral or intravenous BPs with placebo or a no-treatment control, in what concerns AF risk. In order to be included in the quantitative analysis, studies were required to feature a minimum patient follow-up of 6 months. De novo AF diagnoses served as the primary endpoint. Data related to individual BPs were further investigated separately, with respect to this outcome. Study-specific Mantel-Haenszel odds ratios (ORs) were pooled using traditional meta-analytic techniques, under a random-effects model. Results 42 RCTs, encompassing 52.436 patients (32.071 randomized to BPs), were regarded as eligible for quantitative synthesis. Of note, 2 pooled analyses, one of 4 trials with ibandronate and the other of 6 trials with risedronate, were included. Individual BP representation may be depicted as follows: Alendronate, 23 trials, with 14.599 patients; Risedronate, 7 trials, with 15.350 patients; Zoledronic acid, 7 trials, with 13.059 patients; Ibandronate, 4 trials, with 8.754 patients; and Minedronate, 1 trial, with 674 patients. 748 de novo AF diagnoses were reported, in total. In the main analysis, BPs were not found to be significantly associated with an increase in AF odds (OR 1.10, 95% CI 0.95–1.28, P 0.21, i2 0%). As for individual BPs, Alendronate (OR 1.09, 95% CI 0.82–1.45, P 0.55, i2 0%), Risedronate (OR 0.81, 95% CI 0.35–1.86, P 0.61, i2 31%), Ibandronate (OR 0.89, 95% CI 0.52–1.52, P 0.67) and Minedronate (0 AF events reported, both in the active and in the control group) were also not shown to meaningfully enhance AF risk. On the contrary, Zoledronic acid utilization was associated with a significant, though small, increase in new AF cases (OR 1.29, 95% CI 1.01–1.64, P 0.04, i2 0%). Conclusion The professed BP-driven increase in AF odds is not apparent in a fairly populated randomized setting. In fact, a barely significant increment in AF risk seems only to occur with the most potent BP (Zoledronic acid). Therefore, AF development concerns should not refrain doctors from prescribing this highly effective pharmacological class. Funding Acknowledgement Type of funding sources: None.