Glycemic Outcomes After Switching from SGLT2 Inhibitors to DPP-4 Inhibitors in Type 2 Diabetes, with a Comparison of Teneligliptin and Other Agents: A Retrospective Cohort Study

Background and Objectives: Sodium–glucose cotransporter-2 (SGLT2) inhibitors provide cardio-renal protection in type 2 diabetes, but tolerability-related discontinuation is common, creating a need for effective replacement therapy. Dipeptidyl peptidase-4 (DPP-4) inhibitors are widely used as follow-on agents, yet the clinical course after such a switch has not been systematically characterized, and whether individual DPP-4 inhibitors differ in efficacy in this setting is unknown. Materials and Methods: We conducted a single-center retrospective cohort study at Chungnam National University Hospital (Daejeon, Republic of Korea) between January 2013 and January 2025, including 117 adults with type 2 diabetes who switched from an SGLT2 inhibitor to a DPP-4 inhibitor and met pre-specified criteria for medication stability and follow-up. The primary outcome was the change in glycated hemoglobin (HbA1c) from baseline to 3 months; secondary outcomes were changes in body weight, body mass index, blood pressure, and renal parameters. A pre-specified subgroup analysis compared teneligliptin (n = 60) with other DPP-4 inhibitors as a class (n = 57). Results: In the overall cohort, body weight rose by 0.7 kg and systolic blood pressure by 4.3 mmHg at 3 months (both p < 0.05), whereas HbA1c was unchanged (p = 0.157). In the subgroup analysis, HbA1c fell significantly with teneligliptin (−0.36%; 95% confidence interval, −0.62 to −0.10; p = 0.009) but not with other DPP-4 inhibitors (+0.13%). The between-group difference was −0.49% (95% confidence interval, −0.83 to −0.15; p = 0.005) and persisted after adjustment for baseline HbA1c and estimated glomerular filtration rate (p = 0.007). Conclusions: Switching preserved overall glycemic control at 3 months but produced modest, anticipated increases in body weight and blood pressure. Teneligliptin was associated with a greater HbA1c reduction than the pooled group of other DPP-4 inhibitors; this association persisted after adjustment for the two available baseline covariates but could not be adjusted for diabetes duration, medication adherence, diabetic complications, or other unmeasured factors. Because confounding by indication and other residual confounding cannot be excluded in this short-term, single-center analysis, this finding is hypothesis-generating only and requires confirmation in adequately powered prospective head-to-head trials. Importantly, this study evaluated only the glucose-lowering effect of the switch; because the cardio-renal protection of SGLT2 inhibition is not reproduced by DPP-4 inhibitors, a DPP-4 inhibitor should be regarded as an unavoidable substitute when an SGLT2 inhibitor cannot be maintained rather than a therapeutically equivalent replacement, and preserved HbA1c at 3 months does not establish clinical equivalence between the two strategies.