Sodium-Glucose Transport Protein 2 Inhibitor Use for Type 2 Diabetes and the Incidence of Acute Kidney Injury in Taiwan

IMPORTANCE: The association between sodium-glucose transport protein 2 inhibitor (SGLT2i) use and the incidence of acute kidney injury (AKI) remains controversial. The benefits of SGLT2i use in patients to reduce AKI requiring dialysis (AKI-D) and concomitant diseases with AKI as well as improve AKI prognosis have not yet been established. OBJECTIVE: To investigate the association between SGLT2i use and AKI incidence in patients with type 2 diabetes (T2D). DESIGN, SETTING, AND PARTICIPANTS: This nationwide retrospective cohort study used the National Health Insurance Research Database in Taiwan. The study analyzed a propensity score–matched population of 104 462 patients with T2D treated with SGLT2is or dipeptidyl peptidase 4 inhibitors (DPP4is) between May 2016 and December 2018. All participants were followed up from the index date until the occurrence of outcomes of interest, death, or the end of the study, whichever was earliest. Analysis was conducted between October 15, 2021, and January 30, 2022. MAIN OUTCOMES AND MEASURES: The primary outcome was the incidence of AKI and AKI-D during the study period. AKI was diagnosed using International Classification of Diseases diagnostic codes, and AKI-D was determined using the diagnostic codes and dialysis treatment during the same hospitalization. Conditional Cox proportional hazard models assessed the associations between SGLT2i use and the risks of AKI and AKI-D. The concomitant diseases with AKI and its 90-day prognosis, ie, the occurrence of advanced chronic kidney disease (CKD stage 4 and 5), end-stage kidney disease, or death, were considered when exploring the outcomes of SGLT2i use. RESULTS: In a total of 104 462 patients, 46 065 (44.1%) were female patients, and the mean (SD) age was 58 (12) years. After a follow-up of approximately 2.50 years, 856 participants (0.8%) had AKI and 102 (<0.1%) had AKI-D. SGLT2i users had a 0.66-fold risk for AKI (95% CI, 0.57-0.75; P < .001) and 0.56-fold risk of AKI-D (95% CI, 0.37-0.84; P = .005) compared with DPP4i users. The numbers of patients with AKI with heart disease, sepsis, respiratory failure, and shock were 80 (22.73%), 83 (23.58%), 23 (6.53%), and 10 (2.84%), respectively. SGLT2i use was associated with lower risk of AKI with respiratory failure (hazard ratio [HR], 0.42; 95% CI, 0.26-0.69; P < .001) and shock (HR, 0.48; 95% CI, 0.23-0.99; P = .048) but not AKI with heart disease (HR, 0.79; 95% CI, 0.58-1.07; P = .13) and sepsis (HR, 0.77; 95% CI, 0.58-1.03; P = .08). The 90-day AKI prognosis for the risk of advanced CKD indicated a 6.53% (23 of 352 patients) lower incidence in SGLT2i users than in DPP4i users (P = .045). CONCLUSIONS AND RELEVANCE: The study findings suggest that patients with T2D who receive SGLT2i may have lower risk of AKI and AKI-D compared with those who receive DPP4i.