Ann N Y Acad Sci. 2020 Mar 24. doi: 10.1111/nyas.14332. [Epub ahead of print]
Bisphosphonates (BPs) are widely prescribed drugs used to treat osteoporosis, commonly arising in postmenopausal women and in chronic glucocorticoid use. Their mechanism of action is through inhibiting osteoclast-induced bone remodeling, and they also possess calcium sequestering properties. Common side effects involve the gastrointestinal system and rare but serious side effects, including osteonecrosis of the jaw. However, a link between BPs and atrial fibrillation (AF) has been proposed, with early clinical trials, such as the Fracture Intervention Trial and the HORIZON Pivotal Fracture Trial, reporting that BPs are associated with increased risk of AF. Nevertheless, subsequent studies have reported contrasting results, ranging from no effect of BPs to antiarrhythmic effects of BPs. Preclinical and electrophysiological studies on any proarrhythmic effect of BPs are limited in scope and number, but suggest possible mechanisms that include antiangionesis-related myocardial remodeling, calcium handling abnormalities, and inflammatory changes. Contrastingly, some studies indicate that BPs are antiarrhythmic by inhibiting fibrotic myocardial remodeling. In order to continue established clinical prescribing of BPs within absolute margins of safety, it will be necessary to systematically rule in/rule out these mechanisms. Thus, we discuss these studies and examine in detail the potential mechanistic links, with the aim of suggesting further avenues for research.
Discussion and conclusions
BPs are commonly used to treat conditions predominantly associated with age: postmenopausalor glucocorticoid-induced OP and bony metastases from cancer. Adverse effects lie on a spectrum of severity, but common side effects include gastrointestinal irritation, and less common but more severe rare side effects include osteonecrosis of the jaw. Clinical data suggest that there is a relationship between AF incidence and BP use, but studies report conflicting results both for and against a role of BPs in arrhythmogenesis, and some even suggest an antiarrhythmic effect. Perhaps, the effect size of BP use is small, and only subtly increases the risk of AF but does not always lead to it. Granted, older patients aremore likely to have comorbidities, such as hypertension, vascular disease, and diabetes mellitus, which increase the risk of AF regardless of BP use.However, there exist other confounding factors, such as medications, including statins, glucocorticoids, and antihypertensives, which are common in these patients and may be antiarrhythmic. These caveats mean that it is harder to tease out a small effect of BP, as these drugs may mask any proarrhythmic effects of BPs.
In conclusion, the retrospective clinical studies investigating the association between AF and use of BPs provide evidence which, on the whole, is neither confirmatory nor rejecting of a causative role. Several plausible mechanistic links have been proposed, including anatomical remodeling, alterations in cellular Ca2+ homeostasis, and inflammatory acute phase responses (Fig. 2). There is limited experimental evidence investigating these claims, with studies being limited themselves in terms of the experimental models used. If clinical practice is meant to operate in an entirely safe way, which is backed by evidence, then physiological studies looking to concurrently rule in/out the arrhythmogenic implications of BPs are vital. Such studies should include but not be limited to laboratorybased experiments using well-established animal models, such as murine exemplars, which can also be used to elucidate any effect of aging on these phenomena.