Chronic kidney disease (CKD) is associated with an elevated risk of sepsis and worse infectious outcomes compared with the general population. Glucagon‑like peptide‑1 receptor agonists (GLP‑1RA) have metabolic benefits in type 2 diabetes and have been proposed to exert protective effects against infection and sepsis. Whether these potential benefits extend to patients with CKD was unclear. The study summarized here evaluated the association between initiation of GLP‑1RA therapy and subsequent infectious outcomes in adults who have both CKD and type 2 diabetes mellitus.
The investigators performed a retrospective cohort study using the TriNetX electronic health‑record network. Eligible adult patients had diagnoses of CKD and type 2 diabetes and initiated either a GLP‑1RA or a dipeptidyl peptidase‑4 inhibitor (DPP‑4i) between January 2020 and May 2026. The primary outcome of interest was sepsis; several secondary infectious and mortality outcomes were predefined.
After applying inclusion criteria and propensity score matching to balance baseline characteristics between treatment groups, the analysis included 21,035 patients in the GLP‑1RA group and 21,035 patients in the DPP‑4i group. The abstract reports matched cohorts but does not provide the list of matching variables, baseline covariate balance metrics, or absolute event counts in the source abstract.
Primary outcome:
Secondary outcomes:
The abstract does not report follow‑up duration, censoring rules, or how each outcome was defined in the electronic record.
In the propensity‑matched cohorts, initiation of GLP‑1RA was associated with a lower hazard of sepsis compared with initiation of DPP‑4i (hazard ratio [HR] 0.72; 95% confidence interval [CI], 0.66–0.78). All‑cause mortality was also lower among GLP‑1RA users (HR 0.65; 95% CI, 0.61–0.70). These associations were observed in the matched analysis reported in the abstract.
GLP‑1RA use was associated with lower risks of several secondary infectious outcomes relative to DPP‑4i in the matched cohorts:
These hazard ratios indicate a consistent signal for reduced risk across multiple infection‑related endpoints in this retrospective comparison.
The hazard ratio for septic shock was 0.84 (95% CI, 0.73–0.96) favoring GLP‑1RA numerically; the reported P value was 0.014. The abstract notes that this result did not reach the Bonferroni‑corrected significance threshold applied by the authors, so septic shock was not considered statistically significant after multiple‑comparison adjustment.
In this large, matched retrospective cohort from the TriNetX network, initiation of GLP‑1RA versus DPP‑4i in adults with CKD and type 2 diabetes was associated with lower hazards of sepsis, all‑cause mortality, pneumonia, UTI, and COVID‑19. These findings suggest a potential infectious‑outcome benefit of GLP‑1RA in this high‑risk population. Clinicians should interpret these associations in the context of observational data and consider that randomized data or additional mechanistic evidence would be needed to support causal inference.
The source provided is the article abstract. Specific methodological details that are not reported in the abstract include the definitions and coding used for CKD and outcomes, the baseline covariates included in propensity score models, absolute event rates and event counts, median follow‑up time, sensitivity or subgroup analyses, and handling of competing risks. These details are necessary to fully evaluate residual confounding, ascertainment bias, and generalizability but were not available in the abstract.
According to the reported abstract (Yu‑Yang Tseng et al., Nephrol Dial Transplant, online ahead of print 2026), among adults with CKD and type 2 diabetes, initiation of GLP‑1RA compared with DPP‑4i was associated with lower risks of sepsis, mortality, pneumonia, urinary tract infection, and COVID‑19 in propensity‑matched analyses. Septic shock incidence was numerically lower but did not meet the authors' multiple‑comparison threshold for statistical significance. For full appraisal, readers should consult the full article for methodological details and absolute event data.