Real-world efficacy and safety of SGLT2 inhibitor add-on triple therapy in south indian patients with type 2 diabetes mellitus: a 24-week prospective observational cohort study.

Abstract

Background: Type 2 diabetes mellitus (T2DM) in South Asian populations is characterised by early-onset disease, visceral adiposity disproportionate to body mass index, and accelerated pancreatic β-cell decline which is collectively termed the Asian Indian Phenotype. This study evaluated the glycaemic, anthropometric, haemodynamic, hepatorenal, and mechanistic effects of adding a sodium-glucose cotransporter-2 (SGLT2) inhibitor to existing Metformin and Glimepiride dual therapy in South Indian patients with suboptimally controlled T2DM.

Methods: This prospective, single-arm observational cohort study (TREND statement for nonrandomised studies followed) enrolled 87 adult South Indian patients with T2DM inadequately controlled on Metformin (≥ 1500 mg/day) and Glimepiride (2-4 mg/day), with HbA1c 7.5-10.0% and eGFR ≥ 60 mL/min/1.73 m². Treatment was allocated at the discretion of the treating physician rather than by randomisation; patients received Dapagliflozin 10 mg (n = 52) or Empagliflozin 10-25 mg (n = 35) for 24 weeks. As an exploratory study, no a priori sample size calculation was performed and findings are considered hypothesis-generating. Continuous variables were assessed for normality (Shapiro-Wilk test and Q-Q plots); intra-group comparisons used the paired t-test with 95% confidence intervals (CIs). Subgroup, correlation, responder, and multivariable regression analyses were prespecified and are exploratory.

Results: HbA1c declined from 8.62 ± 0.74% to 7.71 ± 0.58% at Week 24 (mean reduction: -0.91%; 95% CI: -1.08 to - 0.74%; p < 0.001). By Week 24, 60 patients (69.0%) achieved HbA1c < 8.0%, 31 (35.6%) achieved < 7.5%, and 9 (10.3%) reached < 7.0%. Fasting plasma glucose fell by 35.6 mg/dL (95% CI: -42.8 to - 28.4; p < 0.001) and post-prandial glucose by 47.6 mg/dL (95% CI: -57.9 to - 37.3; p < 0.001). Body weight decreased by 2.27 kg (95% CI: -4.29 to - 0.25; p = 0.028) and waist circumference by 2.36 cm (95% CI: -4.19 to - 0.53; p = 0.012). Systolic blood pressure fell by 6.3 mmHg (95% CI: -9.9 to - 2.7; p = 0.001). A biphasic eGFR trajectory was documented: dip to 83.1 ± 11.9 mL/min/1.73 m² at Week 12, recovering to 85.6 ± 11.2 at Week 24 (p = 0.007 vs. baseline). ALP decreased significantly (- 11.7 U/L; 95% CI: -18.4 to - 5.0; p = 0.001). Baseline HbA1c was the only independent predictor of glycaemic response on multivariable regression (β=-0.47 per %; p < 0.001; adjusted R²=0.32). No between-drug difference was detected (p = 0.62), though the study was not powered to demonstrate equivalence. Patients with baseline HbA1c 8.5-10.0% achieved a greater absolute reduction (- 1.15%) than those with 7.5-8.5% (- 0.63%; p = 0.003), a gradient partly attributable to regression to the mean; the lower-baseline subgroup, however, more frequently reached glycaemic targets. Genital mycotic infections occurred in 10.3% of women versus 2.1% of men. No serious adverse events occurred.

Conclusion: In this uncontrolled real-world cohort, SGLT2 inhibitor triple therapy was associated with multidimensional metabolic improvement in South Indian T2DM. No difference between the two agents was detected, though the study was not designed to establish equivalence. Baseline HbA1c independently predicted the magnitude of glycaemic response. Because the design was observational and lacked a comparator, these associations are hypothesis-generating and support earlier intensification while awaiting confirmation in randomised trials.