Evaluating risks and benefits of Newer Diabetes medications when Used in Routine care (ENDURe)

IntroductionSodium glucose co-transporter 2 inhibitors (SGLT2i) and glucagon-like peptide 1 analogues (GLP1a) improve clinical outcomes (e.g., myocardial infarction, stroke, death from CV causes) for adults with type 2 diabetes mellitus. The majority of available data from clinical trials and observational studies are from outpatients. Data on their effectiveness among hospitalized patients, however, are lacking. MethodsWe conducted a multicentre, retrospective, cohort study of adults aged 65 years and older with type 2 diabetes mellitus hospitalized between 2017 and 2023 in Ontario. We compared adults newly prescribed an SGLT2i or GLP1a in hospital to adults newly prescribed a DPP4i in hospital. In a sensitivity analysis, new use of sulfonylurea was the comparator. Our primary outcome was the 1-year risk of a composite of all-cause mortality, hospitalization with myocardial infarction, stroke, heart failure, or renal failure. Secondary outcomes included components of the composite, short-term (i.e., 30 day) risk of hypoglycemia, and the 1-year risk of DKA. ResultsWe identified 6,713 older adults with diabetes who were newly prescribed one of the following medications during an inpatient hospitalization: SGLT2i (N=1520), GLP1 (N=90), DPP4i (N=3726), sulfonylurea (N=1377). Because new use of GLP1 was rare in hospital, we updated our exposure group to SGLT2i alone. Adults who received an SGLT2i were typically younger, and more likely to have heart failure or coronary artery disease compared to adults who received a DPP4i or sulfonylurea. Among adults who received an SGLT2i, at 1-year the primary outcome occurred in 26% compared to 31% who newly received a DPP4i (adjusted hazard ratio [HR] 0.82 95% Confidence Interval [CI] 0.69,0.97). In our sensitivity analysis using sulfonylurea as the comparator group, the hazard ratio for the primary outcome was 0.97 (95% CI 0.80, 1.18). We did not identify an increased risk of DKA or hypoglycemia for SGLT2i compared to DPP4i, though patients receiving an SGLT2i did have a lower rate of 30-day readmission (HR 0.73, 95%CI 0.58-0.92). ConclusionAmong older adults with type 2 diabetes mellitus, newly prescribing an SGLT2i or GLP1 during a hospitalization was uncommon. New use of an SGLT2i was associated with improved outcomes compared to DPP4i but this finding was not robust when new use of a sulfonylurea was the comparator. Future larger studies are needed to provide more definitive results.