Real-world Use and Outcomes of Sodium-glucose Cotransporter-2 Inhibitors in Adults With Diabetes and Heart Failure: A Population-level Cohort Study in Alberta, Canada

Abstract

ObjectivesSince 2016, clinical guidelines have recommended sodium-glucose cotransporter-2 inhibitors (SGLT2is) for people with type 2 diabetes with heart failure. We examined SGLT2i dispensation, factors associated with dispensation, and heart failure hospitalization and all-cause mortality in people with diabetes and heart failure. MethodsThis retrospective, population-based cohort study, identified people with diabetes and heart failure between Jan 1, 2014 to Dec 31, 2017 in Alberta, Canada and followed them for a minimum of three years for SGLT2i dispensation and outcomes. Multivariate logistic regression assessed the factors associated with SGTL2i dispensation. Propensity scores were used with regression adjustment to estimate the effect of SGLT2i treatment on heart failure hospitalization. ResultsAmong 22,025 individuals with diabetes and heart failure (43.4% women, mean age 74.7±11.8 years), only 10.2% were dispensed an SGLT2i. Male sex, age <65 years, a higher baseline A1C, no chronic kidney disease, presence of atherosclerotic cardiovascular disease, and urban residence were associated with SGLT2i dispensation. Lower heart failure hospitalization rates were observed in those with SGLT2i dispensation (548.1 per 100 person years) vs those without (813.5 per 1,000 person years; p<0.001) and lower all-cause mortality in those with an SGLT2i than those without (48.5 per 1,000 person years vs 206.1 per 1,000 person years; p<0.001). Regression adjustment found SGLT2i therapy was associated with a 23% reduction in hospitalization. ConclusionsSGLT2is were dispensed to only 10% of people with diabetes and established heart failure, underscoring a significant care gap. SGLT2i use was associated with a real-world reduction in heart failure hospitalization and all-cause death. This study highlights an important opportunity to optimize SGLT2i use.