JAMA Intern Med. 2026 Jan 20. doi: 10.1001/jamainternmed.2025.7409. Online ahead of print.
ABSTRACT
IMPORTANCE: No randomized clinical trial has directly compared the effectiveness of sodium-glucose cotransporter-2 inhibitor (SGLT2i) and glucagon-like peptide-1 receptor agonist (GLP-1RA) treatment in reducing acute and chronic kidney outcomes.
OBJECTIVE: To examine the comparative effectiveness of SGLT2i and GLP-1RA treatment for acute and chronic kidney outcomes in individuals with type 2 diabetes.
DESIGN, SETTING, AND PARTICIPANTS: This comparative effectiveness study with a target trial emulation design used nationwide, population-based data from Denmark. Participants were individuals with metformin-treated type 2 diabetes who initiated SGLT2i or GLP-1RA treatment from January 2014 to November 2020, with follow-up through October 2024.
EXPOSURE: Initiation of an SGLT2i or a GLP-1RA.
MAIN OUTCOMES AND MEASURES: The 2 coprimary outcomes were chronic kidney disease (CKD; 40% reduction in estimated glomerular filtration rate [eGFR], severe albuminuria, or kidney failure) and acute kidney injury (AKI). Secondary outcomes included the individual components of CKD, albuminuria, and death. Intention-to-treat effects were estimated using inverse probability of treatment weights, comparing risks for CKD assessed by the Aalen-Johansen estimator, and AKI burden by mean cumulative counts (MCCs; mean number of events per individual as multiple AKI events were possible). Subgroup analyses included stratification by preexisting cardiovascular or kidney disease.
RESULTS: The study included 36 279 individuals who initiated an SGLT2i and 18 782 who initiated a GLP-1RA (median [IQR] age, 63 [55-71] years vs 61 [52-70] years), with comparable diabetes duration, eGFR, and urine albumin-creatinine ratios. The weighted 5-year risk of CKD was 6.7% (95% CI, 6.4%-7.0%) for SGLT2i initiators and 8.2% (95% CI, 7.8%-8.6%) for GLP-1RA initiators (risk ratio: 0.81 [95% CI, 0.76-0.87]; risk difference: -1.5% [95% CI, -2.0% to -1.0%]). The 5-year MCC of AKI per 100 individuals was 25.2 (95% CI, 24.4-26.1) for SGLT2i initiators and 28.7 (95% CI, 27.4-30.0) for GLP-1RA initiators (MCC ratio: 0.88 [95% CI, 0.83-0.93]; MCC difference: -3.5 [95% CI, -5.0 to -2.0]). In contrast, the secondary outcomes of albuminuria and mortality were slightly reduced in GLP-1RA initiators. Results were consistent across subgroups, with the most pronounced CKD and AKI reductions with SGLT2i observed among individuals without preexisting kidney disease.
CONCLUSIONS AND RELEVANCE: This comparative effectiveness study found that initiation of SGLT2i vs GLP-1RA treatment in individuals with type 2 diabetes was associated with a lower 5-year risk of CKD and a lower 5-year count of AKI. These findings underscore the potential of SGLT2i treatment for primary prevention of kidney disease in individuals with type 2 diabetes.
PMID:41557360 | DOI:10.1001/jamainternmed.2025.7409