Systematic literature review and network meta-analysis of sodium-glucose co-transporter inhibitors vs metformin as add-on to insulin in type 1 diabetes

Aims To identify and synthesize phase 3 and phase 4 randomized controlled trials (RCTs) of sodium-glucose co-transporter (SGLT) inhibitors and metformin as adjuncts to insulin in type 1 diabetes (T1DM) using network meta-analysis (NMA). Materials and methods A systematic literature review (SLR) identified relevant RCTs of >= 12 Weeks duration. MEDLINE, Embase, the Cochrane Library and grey literature were searched through October 2018. NMAs indirectly compared SGLT inhibitors and metformin for change from baseline in HbA1c, weight, total daily insulin dose and systolic blood pressure at Week 24 to 26 and Week 52. Safety outcomes were also explored. Results Nine trials (N = 6780) were included in the SLR. NMAs indicated that all therapies performed better than placebo for the efficacy outcomes at both time points. Compared with metformin at Week 24 to 26, the SGLT inhibitors dapagliflozin (5 mg), sotagliflozin (200 mg) and empagliflozin (10 mg) had larger reductions in HbA1c (mean difference [MD] = -0.24, 95% credible interval [CrI], -0.41 to -0.07, MD = -0.23, 95% CrI, -0.39 to -0.08 and MD = -0.35, 95% CrI, -0.51 to -0.19, respectively) and in weight, which were sustained in sensitivity analyses. There were few differences observed in the results of safety outcomes, such as risk of diabetic ketoacidosis (DKA), which should be interpreted cautiously because of wide CrIs. Conclusions Adjunctive use of SGLT inhibitors in T1DM can improve glycaemic control compared with metformin while enabling weight loss, with consistent efficacy across the class. However, these results are based on indirect evidence so confirmation in a head-to-head study would be valuable.