Zhou · Frontiers in aging neuroscience 2022 · systematic review and Bayesian network meta-analysis · n=39 studies (2,031 participants)

Effects of different physical activities on brain-derived neurotrophic factor: A systematic review and bayesian network meta-analysis.

Cited 55 times in the scientific literature.

Level 1 - systematic review of randomized trials

Systematic review and Bayesian network meta-analysis of 39 randomized controlled trials.

PubMed 36092802 · doi:10.3389/fnagi.2022.981002 · record verified 2026-08-29

What was done

The authors conducted a systematic review and Bayesian network meta-analysis across seven databases (CNKI, PubMed, Embase, Scopus, Medline, Web of Science, Cochrane Library) to compare the effects of five exercise interventions on serum brain-derived neurotrophic factor (BDNF) in healthy and non-healthy populations. Interventions comprised aerobic training (AT, 451 subjects), resistance training (RT, 236 subjects), AT+RT (102 subjects), high-intensity interval training (HIIT, 84 subjects), combined training (CT, 293 subjects), and conventional control groups (CG, 865 subjects) across 39 randomized controlled trials encompassing 2,031 participants.

What was found

All five exercise modalities yielded superior BDNF improvements relative to control. Network meta-analysis ranked the interventions in order of effect: RT > HIIT > CT > AT+RT > AT > CG. Resistance training exhibited the largest effect compared to control [MD = 3.11 (0.33, 5.76), p < 0.05]. Exact numerical effect sizes and confidence intervals for HIIT, CT, AT+RT, and AT were not reported in the abstract.

Why it matters

This review provides comparative rankings for physical activity modalities on circulating BDNF, suggesting resistance exercise may offer the strongest stimulus for neurotrophic upregulation.

Limits

The analysis pooled healthy and non-healthy cohorts together rather than isolating clinical subgroups. Intervention arm sizes were imbalanced (84 subjects in HIIT versus 451 in AT), and exact numerical effect sizes, intervention durations, and training parameters for most modalities are not provided in the abstract.

Cited by