Exploring the heterogeneity of the effects of SGLT-2 inhibitors in cardiovascular outcome trials

Background and aim: The interpretation of discrepancies across cardiovascular safety trials (CVOT) with SGLT-2 inhibitors in the incidence of major cardiovascular events (MACE) and mortality is complex, because of heterogeneity in trial protocols and baseline characteristics of patients enrolled. Aim of this analysis is the exploration of possible determinants of heterogeneity of relative risk reduction for major cardiovascular events (MACE) and all-cause mortality. Methods and results: Incidence of events (MACE and mortality) in intervention and control groups and baseline characteristics of patients were extracted for each trial (EMPAREG, CANVAS, and DECLARE). For studies including both primary and secondary prevention cohorts, those two subgroups were considered separately. Metaregression analysis was used to assess the association of relative risk reduction with baseline characteristics of patients, including observed incidence of events with placebo. The estimated reduction in the incidence of MACE associated with SGLT-2 inhibitors showed a significant association with the incidence of MACE in the control group, suggesting that a greater effect was observed in trials enrolling patients at higher risk (Slope −0.008 [−0.023; 0.007], p = 0.31). A higher proportion of patients treated with statins, β-blockers and insulin at baseline was associated with a greater reduction of MACE, but not of mortality. Conclusions: In CVOT trials, the magnitude of the effect of SGLT-2 inhibitors on MACE is driven by absolute risk of enrolled patients; as a consequence, the estimated risk reduction is lower in primary prevention cohorts, which have a lower risk. This result supports the hypothesis of a class effect of SGLT-2 inhibitors on MACE.