Background: Chinese guideline recommends using combination therapy of antidiabetic-drug plus metformin for type 2 diabetes (T2DM) patients failed to control glucose on metformin. Treatment effect varies between antidiabetic drugs, but there is no direct evidence on comparing the effect of each alternative drug plus metformin. This study aims to estimate and compare the efficacy and safety of seven classes of antidiabetic-drug added onto metformin for Chinese T2DM patients failed on metformin. Methods: Systematic review was performed by searching databases including CNKI, WanFang, CQVIP, PubMed, EMBASE, Web-of-Science, ScienceDirect, Cochrane-Library, to identify randomized controlled trials(1990-2016) comparing the treatment effect of 'antidiabetic-drug + metformin vs metformin' or 'metformin vs placebo/lifestyle-intervention' in Chinese T2DM patients. Based on Chinese guideline, three-line eight-class of antidiabetic-drugs were included: 1st-line-drug(metformin), 2nd-line-drug/oral-drug(α-glycosidase inhibitor, sulfonylurea, glinide, DPP-4 inhibitor, thiazolidinedione), 3rd-line-drug/injectable-drug(insulin, GLP-1 receptor agonist). Using metformin as common comparator, meta-analyses and indirect-treatment-comparisons were used to estimate the effect of 'antidiabetic-drug + metformin vs placebo/lifestyle-intervention'. Findings: 325 studies were finally included. Overall, compared with placebo/lifestyle-intervention, 'oral-drug + metformin' and 'injectable-drug + metformin' reduced HbA1c level by 1·89% and 2·28%, BMI level by 2·55kg/m2 and 3·10kg/m2, TC level by 1·03mmol/l and 1·50mmol/l, SBP level by 5·59mmHg and 8·51mmHg, while increased HDL-C level by 0·31mmol/l and 0·18mmol/l, respectively. In 'oral-drug + metformin', the top two treatments for reducing HbA1c level were 'glinide + metformin' (WMD:-2·03,95%CI[-2·70,-1·36]) and 'sulfonylurea + metformin' (WMD:-2·00,95%CI[-2·64,-1·36]); for reducing BMI level were 'glinide + metformin' (WMD:-4·25,95%CI[-5·70,-2·80]) and 'DPP-4 inhibitor + metformin' (WMD:-2·81,95%CI[-4·22,-1·40]); for reducing TC level were 'α-glycosidase inhibitor + metformin'(WMD:-1·41,95%CI[-2·15,-0·68]) and 'DPP-4 inhibitor + metformin' (WMD:-1·22,95%CI[-1·71,-0·73]); for increasing HDL-C level were 'α-glycosidase inhibitor + metformin'(WMD:0·51,95%CI[0·27,0·75]) and 'thiazolidinedione + metformin' (WMD:0·35,95%CI[0·10,0·60]). In 'injectable-drug + metformin', 'GLP-1 receptor agonist + metformin' was superior to 'insulin + metformin' in reducing HbA1c level (WMD:-2·42vs-2·04) and BMI level (WMD:-3·27vs-2·05), while inferior to 'insulin + metformin' in reducing TC level (WMD:-1·43vs-1·79). All combination therapies had no significant difference on risks of hypoglycemia or gastrointestinal-reaction vs placebo/lifestyle-intervention. Interpretation: Pharmacological combination therapies had better efficacy than placebo/lifestyle-intervention in Chinese T2DM patients. Funding: No funding was received. Declaration of Interest: There are no relevant conflicts of interest to disclose.
Read full abstract