Introduction Relationships between glycemic-lowering effects of sodium glucose co-transporter 2 inhibitors and impact on kidney and cardiovascular outcomes are uncertain.

Research design and methods We analyzed 4395 individuals with prebaseline and postbaseline hemoglobin A1c (HbA1c) randomized to canagliflozin (n=2193) or placebo (n=2202) in The Canagliflozin and Renal Events in Diabetes with Established Nephropathy Clinical Evaluation trial. Effects on HbA1c were assessed using mixed models. Mediation of treatment effects by achieved glycemic control was analyzed using proportional hazards regression with and without adjustment for achieved HbA1c. End points included combined kidney or cardiovascular death, end-stage kidney disease or doubling of serum creatinine (primary trial outcome), and individual end point components.

Results HbA1c lowering was modified by baseline estimated glomerular filtration rate (eGFR). For baseline eGFR 60–90, 45–59, and 30–44 mL/min/1.73 m2, overall HbA1c (canagliflozin vs placebo) decreased by −0.24%, −0.14%, and −0.08% respectively and likelihood of >0.5% decrease in HbA1c decreased with ORs of 1.47 (95% CI 1.27 to 1.67), 1.12 (0.94 to 1.33) and 0.99 (0.83 to 1.18), respectively. Adjustment for postbaseline HbA1c marginally attenuated canagliflozin effects on primary and kidney composite outcomes: unadjusted HR 0.67 (95% CI 0.57 to 0.80) and 0.66 (95% CI 0.53 to 0.81); adjusted for week 13 HbA1c, HR 0.71 (95% CI 0.060 to 0.84) and 0.68 (95% CI 0.55 to 0.83). Results adjusted for time-varying HbA1c or HbA1c as a cubic spline were similar and consistent with preserved clinical benefits across a range of excellent and poor glycemic control.

Conclusions The glycemic effects of canagliflozin are attenuated at lower eGFR but effects on kidney and cardiac end points are preserved. Non-glycemic effects may be primarily responsible for the kidney and cardioprotective benefits of canagliflozin.22