Sodium–glucose co-transporter-2 inhibitors for the prevention of atrial fibrillation: a systemic review and meta-analysis

Abstract

Abstract Aims Sodium–glucose co-transporter-2 (SGLT2) inhibitors are reported to have cardiac benefits. The effects of SGLT2 inhibitors on the prevention of atrial fibrillation (AF) remain inconclusive. We aimed to investigate whether SGLT2 inhibitors can prevent AF occurrence in patients with cardiometabolic diseases. Methods and results We searched MEDLINE, EMBASE, and the Cochrane CENTRAL database up to 1 July 2023. Randomized, placebo-controlled trials of SGLT2 inhibitors in patients with diabetes, heart failure, chronic kidney diseases (CKDs), or cardiometabolic risk factors were included. The primary outcome was AF occurrence. Relative risks (RRs) with 95% confidence intervals (CIs) were calculated in the overall population and selected subgroups. Forty-six trials comprising 101 100 patients were included. Overall, no significant risk reduction of AF occurrence was observed with SGLT2 inhibitors, although there was a favourable trend (RR 0.90, 95% CI 0.80–1.01). In trials with follow-up durations of over 1 year, a similar result was achieved (RR 0.90, 95% CI 0.80–1.01). The results were consistent across different SGLT2 inhibitors, with RRs (95% CIs) of 0.82 (0.60–1.12) for canagliflozin, 0.87 (0.73–1.03) for dapagliflozin, 0.97 (0.78–1.22) for empagliflozin, 0.99 (0.66–1.50) for sotagliflozin, and 0.87 (0.58–1.29) for ertugliflozin. Analyses in different doses of SGLT2 inhibitors yielded similar results. The associations between SGLT2 inhibitors and AF occurrence were also absent in patients with diabetes, heart failure, and CKDs. Conclusion For patients with cardiometabolic diseases or risk factors, SGLT2 inhibitors did not decrease the risk of AF occurrence, regardless of follow-up duration, type or dose of the drug, or the patient population. Lay summary The effects of sodium–glucose co-transporter-2 (SGLT2) inhibitors on the prevention of atrial fibrillation (AF) remain inconclusive.