A Network Meta-Analysis of the Efficacy and Safety of Empagliflozin at Different Doses in Patients with Diabetes Mellitus Based on Randomized Controlled Trials

Abstract Background and objective: As an oral hypoglycemic drug that significantly reduces cardiovascular risk, empagliflozin is used in patients with type 2 diabetes. However, the dosage and administration of empagliflozin are still controversial clinically. To determine the appropriate treatment, we performed this network meta-analysis.Methods: We identified randomized controlled trials (RCTs) about empagliflozin from databases including PubMed, Ovid MEDLINE, Embase, ScienceDirect, Web of Science, the Cochrane Library, Scopus and Google Scholar. We analyzed the pharmacodynamics, adverse effects (AEs), and pharmacokinetics of empagliflozin at different doses.Results: We identified 8264 articles, of which 26 RCTs with 11796 patients were included. Regarding hemoglobin A1c (HbA1c ) and fasting plasma glucose (FPG), high doses (10, 25, 50 mg) were significantly better than low doses (1, 2.5, 5 mg). For total AEs, there was a dose-response trend in which safety decreased with increasing doses. According to SUCRA sequencing, the order for lowering HbA1c was 25 > 50 > 10 > 2.5 > 5 > 1 mg, for lowering FPG was 50 > 25 > 10 > 5 > 2.5 > 1 mg and for safety was 1 > 2.5 > 5 > 25 > 10 > 50 mg. When considering HbA1c, FPG and total AEs, we performed a hierarchical cluster analysis and network meta-analysis to find that 25 mg performed best among different doses, which was more significant after long-term use (≥ 12 weeks). Pharmacokinetic parameters exhibited significant dose-response relationships .Conclusions: High doses (10, 25, 50 mg) had better efficacy than low doses (1, 2.5, 5 mg). When considering HbA1c, FPG and total AEs, 25 mg performed best among the different doses. More RCTs exploring unconventional doses are needed to confirm these conclusions.