The Association of Background Medications on Initial eGFR Change and Kidney Outcomes in Diabetic Patients Receiving SGLT2 Inhibitor

Abstract

Background To determine whether background medications modify the effects of sodium-glucose cotransporter-2 inhibitor (SGLT2i) on the eGFR and kidney outcomes among patients with type 2 diabetes. Methods We used medical data from a multicenter health care facility in Taiwan and included 10,071 patients who received SGLT2i treatment from June 1, 2016, to December 31, 2018. Direct comparisons for use versus no use of specific background drugs were conducted after adjusting for baseline characteristics through propensity score matching. Patients were followed up until the occurrence of composite kidney outcomes (two-fold increase in the serum creatinine level or the development of end-stage kidney disease), mortality, or the end of the study period. Results Patients exhibited an initial mean (SEM) decline of −2.72 (0.10) ml/min per 1.73 m2 in eGFR dip from baseline to a mean treatment duration of 8.1±3.1 weeks after SGLT2i initiation. The eGFR trajectory stabilized 24 weeks after SGLT2i treatment with a mean (SEM) slope of −1.36 (0.25) ml/min per 1.73 m2 per year. Compared with no drug use, the use of background renin-angiotensin inhibitor (n=2073), thiazide diuretics (n=1764), loop diuretics (n=708), fenofibrate (n=1043), xanthine oxidase inhibitor (n=264), and insulin (n=1656) was associated with a larger initial decrease in eGFR, while background metformin treatment (n=827) was associated with a smaller initial decrease in eGFR after SGLT2i treatment. The only drugs associated with the long-term composite kidney outcome during SGLT2i treatment were renin-angiotensin inhibitor (hazard ratio [HR], 0.61; 95% confidence interval [CI], 0.40 to 0.95) and loop diuretics (HR, 1.88; 95% CI, 1.19 to 2.96). Conclusions Several background medications were associated with the initial eGFR dip after SGLT2i initiation. Most drugs were not associated with long-term composite kidney outcomes among patients treated with SGLT2i, except for renin-angiotensin system inhibitor associated with favorable outcomes and loop diuretics associated with worse composite kidney outcomes.