Prevention of heart failure events with sodium–glucose co‐transporter 2 inhibitors across a spectrum of cardio‐renal‐metabolic risk

Aims Trials have tested the safety and efficacy of sodium–glucose co‐transporter 2 inhibitors (SGLT2i) across various disease states. We performed a meta‐analysis of randomized controlled trials (RCTs) to estimate the relative and absolute effects of SGLT2i in the prevention of heart failure (HF) events across different risk groups. Methods and results We conducted a systematic review and meta‐analysis of large, placebo‐controlled RCTs with >1000 participants evaluating HF hospitalization and the composite of cardiovascular (CV) death or HF hospitalization. Due to varying durations of therapeutic exposure and follow‐up, absolute risk reductions and number needed to treat were calculated based on incidence rates (per 100 patient‐years). Across 71 553 patients enrolled in 10 late‐phase RCTs, SGLT2i reduced the risk of HF hospitalization by 31% [hazard ratio (HR) 0.69, 95% confidence interval (CI) 0.64–0.74; I 2 = 0%] and the composite outcome of CV death or HF hospitalization by 24% (HR 0.76, 95% CI 0.72–0.80; I 2 = 1.4%) compared with placebo. The number of patient‐years of treatment exposure needed to prevent one CV death or HF hospitalization ranged from 19–26 (established HF) to 72–125 (chronic kidney disease) to 96–400 (high‐risk type 2 diabetes). In mixed‐effects meta‐regression analyses, the benefits of SGLT2i on HF hospitalizations or the composite outcome (CV death or HF hospitalization) were not influenced by age, sex, or change in intermediate markers (glycated haemoglobin, systolic blood pressure, and body weight) (all P ≥ 0.10). Conclusion Despite wide variation in baseline risks and disease states evaluated, SGLT2i demonstrated comparable relative risk reductions in preventing HF events. Patients at highest baseline risk derived the greatest absolute benefits in preventing HF events. These composite estimates may help guide targeted implementation of SGLT2i for the prevention of HF events in type 2 diabetes and chronic kidney disease and in the treatment of HF.