SodiUm Glucose cotransport 2 and combined 1/2 inhibitors: a systematic Review and network meta-analysis of efficacy and safety in Heart Failure across the range of ejection fraction and with or without diabetes multimorbidity (SUGAR-HF)

Background Variations in pharmacokinetics and pharmacodynamics among sodium-glucose co-transporter 2 inhibitors (SGLT2i) and SGLT1/2i have not been explored adequately. This gap in knowledge about the differential efficacy and safety is exacerbated by an absence of head-on randomized controlled trials (RCTs). This necessitates a network meta-analysis (NMA) to inform comparisons of SGLT2i and SGLT1/2i among each other. Hence, we reviewed the efficacy and safety of SGLT2i and SGLT1/2i in heart failure (HF). Methods This systematic review and network meta-analysis was conducted according to PRISMA-NMA and registered in PROSPERO (CRD42023435066). PubMed, Embase, Cochrane CENTRAL, and Scopus were searched until 2025/04/10, for RCTs of SGLT2i or SGLT1/2i in HF. We addressed a single question, analyzing efficacy separately within two ejection fraction (EF) strata (< 40% and > 40%) and pooling safety across the EF spectrum. The primary outcome was all-cause mortality; other outcomes included cardiovascular death, hospitalization, NT-proBNP, health status, and adverse events. We used frequentist random-effects NMA with a Bayesian sensitivity analysis. We explored heterogeneity through prespecified subgroup analyses for those with and without type 2 diabetes. Risk of bias was assessed using RoB 2, and certainty of evidence was assessed with CINeMA. Results We included 46 RCTs (37,068 participants), mostly at low risk of bias, of which 44 contributed to the network meta-analysis. All effects reported here are network meta-analysis estimates for each drug versus placebo. In HF with EF < 40%, dapagliflozin may have reduced all-cause mortality (RR 0.72, 95% CI 0.53–0.98). The composite of HF hospitalisation or cardiovascular death may be reduced by sotagliflozin (RR 0.70, 0.62–0.78), empagliflozin (RR 0.78, 0.69–0.88), and dapagliflozin (RR 0.74, 0.59–0.94). In HF with EF > 40%, no SGLT2i showed a clear effect on all-cause mortality. However, empagliflozin (RR 0.81, 0.72–0.91) and dapagliflozin (RR 0.84, 0.76–0.93) may have reduced the composite of HF hospitalization or cardiovascular death. For safety, pooled across the EF spectrum, empagliflozin reduced serious adverse events (RR 0.86, 0.80–0.93), while canagliflozin may have reduced all-cause adverse events (RR 0.74, 0.66–0.82). No clear evidence of publication bias was detected in the analyses where it could be evaluated. Conclusion The available evidence suggests possible differences in effect estimates between individual agents, but comparative superiority remains uncertain in the absence of head-to-head trials. This underscores the need for randomized controlled trials directly comparing these agents.