OBJECTIVES : There are multiple classes of antidiabetic drugs in China, which vary in efficacy and safety. However, no study exists comparing all the alternative classes of drugs at the same time. This study aims to estimate and compare the efficacy of alternative classes of antidiabetic drugs in the regimen of monotherapy for Chinese patients with type 2 diabetes (T2DM).

METHODS : Systematic literature review was conducted by retrieving databases including CNKI, WanFang, CQVIP, PubMed, EMBASE, Web-of-Science, ScienceDirect, and Cochrane-Library, to identify randomized controlled trials (1990-2016) comparing the efficacy of ‘antidiabetic drug monotherapy’ vs. ‘placebo/lifestyle-intervention (i.e., diet/exercise)’ in Chinese T2DM patients. Three-lines eight-classes of antidiabetic drugs were included referring to Chinese guideline for T2DM care: 1^st-line-drug (metformin), 2^nd-line-drug (α-glycosidase-inhibitor, sulfonylurea, glinide, DPP-4-inhibitor, thiazolidinedione), 3^rd-line-drug (insulin, GLP-1-receptor-agonist). Meta-analysis was used to estimate the efficacy of different classes of antidiabetic drug vs. placebo/lifestyle-intervention, using random- and fixed-effect models.

RESULTS : 99,072 records were identified through database search, and 46 studies were finally included. Compared with placebo/lifestyle-intervention, 1^st-line-drug (metformin) could change HbA1c by -1.04%, BMI by -1.77kg/m², TC by -0.64mmol/l, SBP by -3.13mmHg, HDL-C by 0.19mmol/l, and hypoglycemia by 2%; 2^nd-line-drug could change HbA1c by -1.19~-0.25%, BMI by -0.64~1.79kg/m², TC by -0.3~0.42mmol/l, SBP by -3.2~1.84mmHg, HDL-C by -0.11~0.22mmol/l, and hypoglycemia by -3~2%; 3^rd-line-drug could change HbA1c by -2.37~-1.28%, BMI by -1.11kg/m², TC by -0.67mmol/l, SBP by -3.00mmHg, HDL-C by 0.18 mmol/l, and hypoglycemia by 0. In the five 2^nd-line-drugs, sulfonylurea was optimal for decreasing HbA1c, α-glycosidase-inhibitor was optimal for decreasing BMI and SBP, DPP-4-inhibitor was optimal for decreasing TC, and thiazolidinedione was optimal for increasing HDL-C and decreasing hypoglycemia. In the two 3^rd-line-drugs, insulin was optimal for decreasing HbA1c, while no data supported for the comparison in BMI, TC, HDL-C, SBP and hypoglycemia.

CONCLUSIONS : This study shows that pharmacological monotherapies have better efficacy than placebo/lifestyle-intervention in Chinese T2DM patients.