RESUMO
Background: The effects of metformin on atrial fibrillation (AF) in type 2 diabetes patients remain unclear. We aimed to explore the effects of metformin on AF, including new-onset AF and AF burden, in type 2 diabetes patients with pacemakers. Methods and results: This retrospective study included a total of 227 patients. Based on the presence of paroxysmal AF, the patients were divided into a paroxysmal AF group (n = 80) and a non-AF group (n = 147). In the non-AF group, a significant association was observed between metformin use and a lower risk of new-onset AF in multivariable Cox hazards models (hazard ratio [HR]: 0.36; 95 % confidence interval [CI]: 0.14-0.91; p = 0.0311*) when adjusted for age, sex, body mass index (BMI), drinking, smoking, left atrial dimension, creatinine, complications, and drugs. In the paroxysmal AF group, univariable analysis indicated no association between the AF burden and metformin use (p = 0.817). Furthermore, when adjusted for metformin use, age, sex, BMI, drinking, smoking, cardiovascular disease, myocardial infarction, heart failure, stroke, and ejection fraction in multivariable Cox hazards models, we found a lower proportion of major adverse cardiovascular events (MACEs) both in the total (HR: 0.28; 95 % CI: 0.1-0.82; p = 0.0202*) and the non-AF group (HR: 0.19; 95 % CI: 0.05-0.79; p = 0.0223*) compared to that in the AF group (HR: 0.31; 95 % CI: 0.02-4.41; p = 0.3879). Conclusion: In type 2 diabetes patients with pacemakers, metformin reduced the probability of new-onset AF instead of addressing the AF burden. Furthermore, metformin therapy decreased the incidence of MACEs in type 2 diabetes patients without AF.