Objective To assess the efficiency and safety of a novel sodium-glucose co-transporter 2 (SGLT2) inhibitor—SGLT2 inhibitors, in combination with insulin for type 1 diabetes mellitus (T1DM). Methods We searched Medline, Embase, and the Cochrane Collaboration Library to identify the eligible studies published between January 2010 and July 2016 without restriction of language. The Food and Drug Administration (FDA) data and ClinicalTrials (http://www.clinicaltrials.gov) were also searched. The included studies met the following criteria: randomized controlled trials; T1DM patients aged between 18 and 65 years old; patients were treated with insulin plus SGLT2 inhibitors for more than 2 weeks; patients’ glycosylated hemoglobin (HbA1c) levels were between 7% and 12%. The SGLT2 inhibitors group was treated with SGLT2 inhibitors plus insulin, and the placebo group received placebo plus insulin treatment. The outcomes should include one of the following items: fasting blood glucose, HbA1c, glycosuria, or adverse effects. Data were analyzed by two physicians independently. The risk of bias was evaluated by using the Cochrane Collaboration’s Risk of Bias tool and heterogeneity among studies was assessed using Chi-square test. Random effect model was used to analyze the treatment effects with Revman 5.3.Results Three trials including 178 patients were enrolled. As compared to the placebo group, SGLT2 inhibitor absolutely decreased fasting blood glucose [mean differences (MD) -2.47 mmol/L, 95% confidence interval (CI) -3.65 to -1.28, P<0.001] and insulin dosage (standardized MD -0.75 U, 95%CI -1.17 to -0.33, P<0.001). SGLT2 inhibitors could also increase the excretion of urine glucose (MD 131.09 g/24 h, 95%CI 91.79 to 170.39, P<0.001). There were no significant differences in the incidences of hyperglycemia [odds ratio (OR) 1.82, 95%CI 0.63 to 5.29, P=0.27], urinary tract infection (OR 0.95, 95%CI 0.19 to 4.85, P=0.95), genital tract infection (OR 0.27, 95%CI 0.01 to 7.19, P=0.43), and diabetic ketoacidosis (OR 6.03, 95%CI 0.27 to 135.99, P=0.26) between the two groups.Conclusion SGLT2 inhibitors combined with insulin might be an efficient and safe treatment modality for T1DM patients.

Key words: sodium-glucose co-transporter 2, type 1 diabetes mellitus, meta analysis, fasting blood glucose, urine glucose, insulin dose, adverse reaction