4111Association of DPP-4 inhibitors, GLP-1 receptor agonists and SGLT-2 inhibitors with cardiovascular events stratified by renal function: a network meta-analysis

Abstract

Abstract Background Chronic kidney disease is a frequent co-morbidity in patients with diabetes and is associated with poor outcomes and elevated cardiovascular risk. Several novel oral antidiabetic therapies have been shown to be efficacious in reducing cardiovascular events. However, the impact of baseline renal function on cardiovascular efficacy is not well-established. Purpose To assess the cardiovascular efficacy of Dipeptidyl Peptidase-4 inhibitors (DPP-4i), Glucagon-Like Peptide-1 Receptor Agonists (GLP-1RA) and Sodium-Glucose Cotransporter-2 inhibitors (SGLT-2i) stratified by renal function. Methods Cardiovascular outcome trials enrolling at least 1000 patients with type 2 diabetes for >12 months' duration were included if they reported on a composite outcome of cardiovascular mortality, nonfatal myocardial infarction and nonfatal stroke and stratified outcome data by estimated glomerular filtration rate (eGFR) of >60 mL/min/1.73m2 or <60 mL/min/1.73m2. Studies enrolling patients with recent acute coronary events were excluded. Treatment and interaction effect estimates were aggregated by random-effects Bayesian network meta-analysis. Results 9 studies were included (2 DPP-4i; 4 GLP-1RA; 3 SGLT-2i), enrolling 92,626 patients accounting for 315,167 patient-years of follow-up. The eGFR was <60 mL/min/1.73m2 in 21,302 patients (23.0%) and >60 mL/min/1.73m2 in 71,324 patients (77.0%). The mean eGFR was 64.8 mL/min/1.73m2 in DPP-4i trials, 78.6 mL/min/1.73m2 in GLP-1RA trials and 78.6 mL/min/1.73m2 in SGLT-2i trials. Compared to placebo, DPP4i were not associated with reductions in the primary outcome (HR 1.00, 95% credible interval [CrI] 0.89–1.13). No reductions in the primary outcome were present for either strata of renal function (eGFR >60: HR 1.00, 95% CrI 0.77–1.29; and eGFR <60: HR 0.99, 95% CrI 0.81–1.20), with no significant interaction between renal function subgroups (interaction coefficient 1.00, 95% CrI 0.72–1.41). GLP-1RA were associated with reductions in the primary outcome overall (HR 0.85, 95% CrI 0.77–0.93). There was no significant interaction between renal function subgroups (eGFR <60: HR 0.87, 95% CrI 0.70–1.06; and eGFR >60: HR 0.83, 95% CrI 0.71–0.94; interaction coefficient 1.05, 95% CrI 0.83–1.35). Similarly, SGLT-2i were associated with reductions in the primary cardiovascular composite outcome (HR 0.88, 95% CrI 0.79–0.97), with no significant interaction between renal function subgroups (eGFR <60: HR 0.82, 95% CrI 0.65–1.05; and eGFR >60: HR 0.91, 95% CrI 0.78–1.06; interaction coefficient 0.92, 95% CrI 0.69–1.21). Conclusions Both SGLT-2i and GLP-RA are associated with reductions in a composite cardiovascular outcome with no evidence of interaction by renal function, while DPP-4i were not associated with reductions in events, with no evidence of interaction by renal function. These data suggest that cardiovascular efficacy of these medications is not affected by baseline renal status.