Cardiovascular outcomes and mortality after initiation of canagliflozin: Analyses from the EASEL Study

INTRODUCTION: In the EASEL study of patients with type 2 diabetes and high cardiovascular risk, initiation of sodium glucose co‐transporter 2 inhibitors (SGLT2i) was associated with lower risk of cardiovascular events and mortality and higher risk of below‐knee lower extremity (BKLE) amputation versus non‐SGLT2i therapies. This analysis further examined risk of cardiovascular events, cardiovascular and noncardiovascular death and BKLE amputation with the SGLT2i canagliflozin versus non‐SGLT2i. METHODS: New user cohorts were constructed from Department of Defense Military Health System patients initiating canagliflozin or non‐SGLT2i (4/1/2013‐12/31/2016). Propensity score matching (1:1) controlled for imbalances in baseline covariates. Incidence rates, hazard ratios and 95% confidence intervals for time to first composite outcome of all‐cause mortality (ACM) and hospitalization for heart failure (HHF), composite major adverse cardiovascular events (MACE) and individual components were evaluated using conditional Cox models. The National Death Index was used to differentiate cardiovascular from noncardiovascular death. The exploratory safety end‐point was BKLE amputation. RESULTS: After propensity matching, 15 394 patients with well‐balanced baseline covariates were followed for a median of 2.03 years (intent‐to‐treat). Canagliflozin showed significant benefit for ACM and HHF (P < .0001), MACE (P = .0001), cardiovascular death (P < .0001) and noncardiovascular death (P = .0018). No significant difference in risk of BKLE amputation was observed (P = .20), though few events were observed. Results were generally consistent in on‐treatment analyses. CONCLUSIONS: In this high cardiovascular risk cohort studied in routine clinical practice, canagliflozin was associated with lower risk of cardiovascular events, cardiovascular death and all‐cause mortality with no significant increase in BKLE amputation risk versus non‐SGLT2i.