FC 128: Novel Glucose-Lowering Drugs and the Risk of Acute Kidney Injury in Routine Care: The Stockholm Creatinine Measurements (SCREAM) Project

Abstract BACKGROUND AND AIMS Little is known about the comparative effects of dipeptidyl peptidase-4 inhibitors (DPP-4i), glucagon-like peptide-1 receptor agonists (GLP-1RAs) or sodium glucose cotransporter-2 inhibitors (SGLT2i) on risk of acute kidney injury (AKI) in routinely care, which may differ from the controlled setting of trials. METHOD We performed a population-based retrospective analysis using the Stockholm CREAtinine Measurements (SCREAM) database, with complete information on diagnoses, medication dispensation claims and laboratory test results for all Stockholm citizens accessing healthcare during 2008–2018. We used a new user active comparator design: new initiation of SGLT2i, GLP1-RA or DPP-4i. The outcome AKI was defined by ICD-10 diagnostic codes as well as KDIGO criteria for rapid creatinine elevations. A multinominal propensity score model with inverse probability treatment weights (IPTW) was used to adjust for 59 identified confounders. Weighted Cox regression was used to estimate adjusted hazard ratios and weighted Kaplan-Meier curves for adjusted absolute risks. RESULTS We included 17 043 participants who newly initiated DPP-4i (N = 10 345), GLP1-RA (N = 4384) or SGLT2i (N = 2314). Mean age was 63 years (38% women) and median eGFR 89 mL/min/1.73 m2. During a median follow-up of 2.5 years, 1361 participants experienced AKI. SGLT2i users had the lowest incidence rate of AKI, 17.4 per 1000 person years [95% confidence interval (95% CI) 13.3–22.4], followed by GLP1-RA users, 22.2 (95% CI 19.6–25) and DPP-4 users, 26.1 (95% CI 24.6–27.8). The adjusted 3-year absolute risk for AKI was 5.98 (95% CI 3.65–8.38) in the SGLT2i arm, compared with 6.45 (95% CI 5.26–7.62) in the GLP1-RA arm and 7.03 (95% CI 6.45–7.59) in the DPP-4i arm. The adjusted hazard ratio was 0.85 (95% CI 0.58–1.18) for SGLT2i versus DPP-4i, and 0.93 (95% CI 0.78–1.11) for GLP1-RA versus DPP-4i. CONCLUSION This study of routinely cared patients initiating novel glucose-lowering drugs observed similar occurrence of AKI between therapies.