Abstract: Objective To evaluate the effects of aerobic exercise, resistance exercise, and aerobic combined resistance exercise (combined exercise) on blood lipids for patients with type 2 diabetes with network meta-analysis, to develop exercise prescription. Methods Randomized controlled trials (RCTs) about the above exercises for blood lipid in patients with type 2 diabetes were searched in databases of PubMed, Ovid, Web of Science, Cochrane Library, Springer Link, Science Direct, CNKI, Wanfang data and CBM, and were supplied by retrospective references. After evaluating the methodological quality of the RCTs, the data were analyzed with R-Studio, Stata 14.0 and Review Manager 5.3. Results A total of 21 RCTs were recalled, with 1735 patients. Two poor quality literatures were excluded based on sensitivity analysis. Compared with the control group, all three kinds of exercise significantly reduced the level of fasting glucose, total cholesterol, triglyceride, low density lipoprotein, increased the level of high density lipoprotein, with good homogeneity. Rank probability graph showed that the combined exercise was the best for fasting glucose, total cholesterol triglyceride and high density lipoprotein, while aerobic exercise was the best for low density lipoprotein. Conclusion It is the best to select the combined exercise to improve fasting glucose, total cholesterol, triglycerides and high density lipoprotein, and aerobic exercise for low density lipoprotein.

Key words: aerobic exercise, resistance exercise, type 2 diabetes, randomized controlled trial, network meta-analysis