A network meta-analysis 1 included 453 trials assessing 21 antidiabetic interventions. 296 trials assessed add-on to metformin-based therapies. At low cardiovascular risk; insulin regimens and specific glucagon-like peptide-1 receptor agonists (GLP-1a) produced the greatest reductions in HbA1c level, and there were no clinically meaningful differences for mortality and vascular outcomes. In patients at increased cardiovascular risk receiving metformin-based background treatment, oral semaglutide, empagliflozin, liraglutide, extended-release exenatide, and dapagliflozin reduced all-cause mortality. Oral semaglutide, empagliflozin, and liraglutide also reduced cardiovascular death. Sodium-glucose cotransporter-2 inhibitors (SGLT-2i) reduced heart failure hospitalization and end-stage renal disease.
A study 2 from Danish national registries included 46 986 type 2 diabetes patients at moderate cardiovascular risk previously on metformin monotherapy, who started an additional glucose-lowering agent. Patients with a history of cardiovascular events were excluded. Patients were followed up to 2 years. DPP-4 inhibitors were used as reference.During follow-up, 1.3% of patients (range 0.8-2.1%) were hospitalised for heart failure: hazard ratios (HR) were 1.11 (95% CI 0.89 to 1.39) for GLP-1a, 0.84 (0.52 to 1.36) for SGLT-2i, 0.98 (0.77 to 1.26) for sulphonyl ureas and 1.54 (1.25 to 1.91) for insulin. The composite endpoint for major cardiovascular events occurred in 2.5% of patients (range 1.5-3.6%), yielding HRs of 0.82 (0.69-0.97) for GLP-1a, 0.79 (0.56-1.12) for SGLT-2i, 1.22 (1.03-1.49) for sulphonyl ureas and 1.23 (1.07-1.47) for insulin. 1865 (3.9%, range 1.9-9.0%) died from any cause, HRs for all-cause mortality were 0.91 (0.78-1.05) for GLP-1a, 0.79 (0.58-1.07) for SGLT-2i, 1.13 (0.99-1.31) for sulphonyl ureas and 2.33 (2.08-2.61) for insulin
Another network meta-analysis 3 included 43 RCTs (16 590 participants) comparing glucose-lowering agents after dual therapy failure. When added to metformin and sulphonyl ureas, after 24 to 36 weeks, the reduction in HbA1c was with GLP-1a 1.0%, basal insulin 0.8% and SGLT-2 inhibitors (0.7%), with no difference between GLP-1a and SGLT-2i; body weight increased with insulin treatment (~3 kg vs placebo), while the greatest reduction was observed for SGLT-2i compared with all other therapies.
A systematic review with meta-analysis 4 included 26 studies with 24 348 patients comparing insulin to GLP-1 analogs, and evaluated add-on therapy with metformin. Insulin was not associated with increased all-cause mortality (RR 0.99; 95% Cl 0.92 to1.06), cardiovascular mortality (RR 1.01; 95% CI 0.91 to1.13), myocardial infarction (RR 1.02; 95% CI 0.92 to1.15), or stroke (RR 0.87; 95% CI 0.68 to 1.12). Insulin treatment increased severe hypoglycemia risk (RR 2.98; 95% CI 2.47 to 3.61).
Primary/Secondary Keywords