Abstract: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.

[Analysis Methods and Case Analysis of Effect Modification (1): Effect Modification in Epidemiology and Traditional Meta-analysis].

[Analysis Methods and Case Analysis of Effect Modification (2): Effect Modification in Network Meta-analysis].

[Analysis Methods and Case Analysis of Effect Modification (3): Effect Modification in Individual Patient Data Meta-analysis].

Meta‐epidemiological Study: a Step‐by‐step Approach by Using R

An Overview of Modern Machine Learning Methods for Effect Measure Modification Analyses in High-Dimensional Settings

Causally-interpretable meta-analysis: clearly-defined causal effects and two case studies

Effects Of Low-Fat Compared With High-Fat Diet On Cardiometabolic Indicators In People With Overweight And Obesity Without Overt Metabolic Disturbance: A Systematic Review And Meta-Analysis Of Randomised Controlled Trials

A Semiparametric Approach to Model Effect Modification

A Methodological Framework for Meta-analysis and Clinical Interpretation of Subgroup Data: The Case of Major Adverse Cardiovascular Events With GLP-1 Receptor Agonists and SGLT2 Inhibitors in Type 2 Diabetes

Efficacy and Safety of Pharmacological Combination Therapies for the Treatment of Type 2 Diabetes in China: A Systematic Review and Meta-Analysis

Effects of Pharmacist-Led Interventions on Glycaemic Control, Adherence, Disease Management and Health-Related Quality of Life in Patients with Type 2 Diabetes: a Protocol for a Network Meta-Analysis.

Change analysis for intermediate disease markers in nutritional epidemiology: a causal inference perspective

[Sensitivity analysis method for unmeasured confounding interference in observational study]

A powerful approach to the study of moderate effect modification in observational studies

Analyzing Treatment Effect by Integrating Existing Propensity Score and Outcome Regressions with Heterogeneous Covariate Sets

Outlier detection and influence diagnostics in network meta-analysis

Inappropriate Evaluation of Effect Modifications Based on Categorical Outcomes: A Systematic Review of Randomized Controlled Trials

Effectiveness and safety of different academic schools of traditional Chinese medicine in the treatment of obesity: A protocol for systematic review and meta-analysis

Effect of electronic health interventions on metabolic syndrome: a systematic review and meta-analysis

Effect of non-drug Intervention on Cognitive Function and Blood Sugar Control of Type 2 diabetes Patients with Mild Cognitive Impairment: A meta-analysis Protocol

A modified regression method to test publication bias in meta-analyses with binary outcomes