Abstract 110: Hospitalizations for Heart Failure Among Patients With Reduced Renal Function Treated With Metformin or Sulfonylureas: A Retrospective Cohort Study

Background: Patients with type 2 diabetes are at risk for the development of secondary complications, including impaired renal function and heart failure (HF). Before 2016, safety concerns limited metformin use in patients with kidney disease. Metformin’s association with HF outcomes and type of HF exacerbation in those with reduced kidney function remains incompletely described. Objective: To compare HF hospitalization between patients treated with metformin versus sulfonylurea following development of reduced kidney function. Research Design/Methods: This national retrospective cohort combined data from Veterans Health Administration, Medicare, Medicaid and National Death Index from 2001 through 2016. We identified patients with new-onset type 2 diabetes by selecting those who were new users of metformin or a sulfonylurea. We followed these patients longitudinally and enrolled those who experienced a kidney function decline and persisted on metformin or sulfonylurea. Patients entered the study on the date of development of reduced kidney function (estimated Glomerular Filtration Rate <60 ml/min or serum creatinine >1.4 [female] or 1.5 [male] mg/dL) and were followed through the primary outcome: hospitalization for HF, or a competing risk: all cause death; drug non-persistence; or study end (December 31 st 2016). The primary analysis used a propensity score weighted model. Echocardiogram reports were obtained from 1 year prior to admission through 7 days after admission through natural language processing to determine the ejection fraction (EF) and HF type [reduced EF<40%; indeterminate EF 40-50% or preserved EF>50%] for that hospitalization. Results: The weighted cohort included 24,683 metformin and 24,802 sulfonylurea users who developed reduced kidney function. At cohort entry, median age was 70 years (63, 78), eGFR 55.8 ml/min/1.73m^2 (51.6, 58.2), and A1c 6.6% (6.1,7.2). Cardiac comorbidities were prevalent at cohort entry, as groups had 12.1% prevalence of CHF (2,987 of 24,683 metformin; 3,009 of 24,802 sulfonylurea), and 31.9% metformin (7,797 of 24,683) and 31.6% sulfonylurea (7,869 of 24,802) users had underlying cardiovascular disease. There were 775 and 991 HF hospitalizations among metformin and sulfonylurea users. HF hospitalizations were 16.9 (15.7, 18.1) vs. 20.7 (19.5, 22.0) per 1,000 person-years respectively (adjusted hazard ratio 0.85, 95% CI: 0.78, 0.93). Of the 775 metformin and 991 sulfonylurea users hospitalized for HF, 44% (339 of 775) vs 45% (447 of 991) did not have an echocardiogram; 28% (214 of 775) vs 31% (302 of 991) had reduced EF; 9% (74 of 775) vs 9% (85 of 991) had indeterminate EF; 19% (148 of 775) vs 16% (157 of 991) had preserved EF. Conclusions: Among patients with type 2 diabetes who develop worsening renal function, persistent metformin use is associated with reduced HF hospitalization compared to sulfonylurea.