Patients with coexisting chronic obstructive pulmonary disease (COPD) and type 2 diabetes mellitus (T2DM) carry an increased susceptibility to infectious complications, including sepsis. As newer antihyperglycemic drug classes—specifically glucagon-like peptide-1 receptor agonists (GLP-1RAs) and sodium-glucose cotransporter 2 inhibitors (SGLT2is)—are adopted in clinical practice, comparative evidence on infection-related outcomes in this multimorbid population remains unclear. The authors aimed to evaluate associations between initiation of GLP-1RAs versus SGLT2is and infection outcomes, with sepsis defined as the primary endpoint.
The investigators conducted a retrospective cohort study using the TriNetX global federated database. They identified individuals aged 40 to 100 years with documented diagnoses of both COPD and T2DM who initiated either a GLP-1RA or an SGLT2i during the period 2021–2024.
A new-user, active-comparator design was employed. To address confounding by measured covariates, a 1:1 propensity score matching strategy was applied, producing matched cohorts of GLP-1RA and SGLT2i initiators. Time-to-event analyses were performed using Cox proportional hazards models, and results are presented as adjusted hazard ratios (HRs) with 95% confidence intervals (CIs).
Details on the specific covariates included in the propensity score, exact matching algorithm parameters, length of follow-up, or handling of missing data were not provided in the abstract and therefore are not reported here.
The pre-specified primary outcome was sepsis. Secondary outcomes included all-cause mortality and components of the primary outcome, with severe sepsis explicitly reported among secondary endpoints. The analytic approach used Cox models to estimate hazard ratios comparing GLP-1RA initiators with SGLT2i initiators.
After propensity score matching, the study included 45,491 matched pairs of patients who initiated a GLP-1RA or an SGLT2i.
Compared with SGLT2i initiators, GLP-1RA initiators experienced:
Lower risk of sepsis: adjusted HR 0.832 (95% CI, 0.781–0.887).
Lower risk of all-cause mortality: adjusted HR 0.642 (95% CI, 0.598–0.689).
Lower risk of severe sepsis: adjusted HR 0.841 (95% CI, 0.774–0.914).
These estimates reflect the relative hazards observed in the propensity-matched cohorts within the TriNetX database for the indicated study period.
In this large, matched retrospective cohort drawn from a global federated network, initiation of a GLP-1RA versus an SGLT2i among patients with COPD and T2DM was associated with significantly lower hazards of sepsis, severe sepsis, and all-cause mortality. These associations suggest that, in this population, choice of antihyperglycemic therapy may be linked to differential infection-related outcomes.
Clinicians treating patients with both COPD and T2DM may consider these observational findings when evaluating antihyperglycemic options, recognizing that the observed associations do not prove causation. Decisions should integrate individual patient characteristics, comorbidities, glucose-lowering needs, and established benefits and risks of each drug class.
The abstract reports the study design, data source, matching approach, cohort size, primary and secondary outcomes, and adjusted hazard ratios. However, the abstract does not report several methodological or contextual details that are important for interpretation, including:
Because these details were not reported in the abstract, they cannot be described here.
Among matched patients aged 40–100 years with both COPD and T2DM who initiated therapy between 2021 and 2024 in the TriNetX database, initiation of GLP-1RAs was associated with lower adjusted hazards of sepsis, severe sepsis, and all-cause mortality compared with initiation of SGLT2is. The authors conclude that GLP-1RA use corresponded to significantly lower risks of these outcomes in this observational analysis.
Further review of the full article is required to assess methodological details, absolute risks, potential residual confounding, and applicability to specific patient subgroups before translating these findings into practice changes.