Abstract
BACKGROUND: Emerging evidence suggests that exercise is a simple and effective method for maintaining brain function. AIMS: This review evaluates the effects of five physical exercises, including aerobic training (AT), high-intensity interval training (HIIT), combined training (CT), resistance training (RT), and AT+RT, on the serum level of brain-derived neurotrophic factor (BDNF) in healthy and non-healthy populations. METHODS: We searched CNKI, PubMed, Embase, Scopus, Medline, Web of Science, and Cochrane Library databases to review randomized controlled studies on exercise interventions for BDNF. Quantitative merging analysis of the resulting data using Bayesian network meta-analysis. RESULTS: The screening and exclusion of the searched literature resulted in the inclusion of 39 randomized controlled trials containing 5 exercise interventions with a total of 2031 subjects. The AT, RT, AT+RT, HIIT, and CT groups (intervention groups) and the CG group (conventional control group) were assigned to 451, 236, 102, 84, 293, and 865 subjects, respectively. The Bayesian network meta-analysis ranked the effect of exercise on BDNF level improvement in healthy and non-healthy subjects as follows: RT > HIIT > CT > AT+RT > AT > CG. Better outcomes were observed in all five intervention groups than in the CG group, with RT having the most significant effect [MD = 3.11 (0.33, 5.76), p < 0.05]. CONCLUSIONS: RT at moderate intensity is recommended for children and older adults in the case of exercise tolerance and is effective in maintaining or modulating BDNF levels for promoting brain health. SYSTEMATIC REVIEW REGISTRATION: https://inplasy.com, INPLASY202250164.