Efficacy and safety of a sodium-glucose co-transporter-2 inhibitor versus placebo as an add-on therapy for people with type 2 diabetes inadequately treated with metformin and a dipeptidyl peptidase-4 inhibitor: a systematic review and meta-analysis of randomised controlled trials

Aims: To conduct a systematic review and meta-analysis to assess the efficacy, safety and tolerability of sodium-glucose co-transporter-2 inhibitors vs placebo as add-on therapy after metformin and dipeptidyl peptidase-4 inhibitor dual therapy in type 2 diabetes. Methods: This systematic review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (PROSPERO registration number: CRD42018099398). A search was conducted via PubMed, www.clinicaltrials.gov and Cochrane Central Register of Controlled Trials of relevant randomised controlled trials up until 14 August 2020 that compared sodium-glucose co-transporter-2 inhibitors vs placebo as add-on therapy after metformin and dipeptidyl peptidase-4 inhibitor therapy. A random-effects model was used. Results: Six randomised controlled trials (1661 participants) met the inclusion criteria. Compared with placebo, sodium-glucose co-transporter-2 inhibitor treatment, as add-on to metformin and dipeptidyl peptidase-4 inhibitor therapy, was associated with a significant reduction in HbA_1c level [mean difference –8 mmol/mol, 95% CI –10, –6 (–0.7%, 95% CI –0.9, –0.6); P < 0.00001], in fasting plasma glucose level [mean difference –1.70 mmol/l, 95% CI –1.91, –1.49; P < 0.00001], in weight (mean difference –1.76 kg, 95% CI –2.04, –1.48; P < 0.00001) and in blood pressure (systolic blood pressure: mean difference –3.6 mmHg, 95% CI –4.8, –2.4; P < 0.00001; diastolic blood pressure: mean difference –1.5 mmHg; 95% CI –2.4, –0.6; P = 0.002). Genital mycotic infections (odds ratio 7.37, 95% CI 3.06, 17.76; P < 0.00001) were more common with sodium-glucose co-transporter-2 inhibitors, but there was no significant statistical difference in urinary tract infections (odds ratio 1.16, 95% CI 0.63, 2.13; P = 0.64), in hypoglycaemia (odds ratio 1.36, 95% CI 0.61, 3.04; P = 0.45), or in discontinuation rates due to adverse events (odds ratio 1.52, 95% CI 0.78, 2.97; P = 0.22) between the two groups. Conclusions: In comparison with placebo, add-on therapy with a sodium-glucose co-transporter-2 inhibitor is significantly more efficacious in lowering HbA_1c, fasting plasma glucose and weight in people with type 2 diabetes following inadequate glycaemic control with metformin and a dipeptidyl peptidase-4 inhibitor. The rate of discontinuation due to adverse events was similar despite higher risk of genital mycotic infections.