What the researchers compared
The review brings together 48 randomized trials, representing 56 comparisons between a regular exercise program and a non-exercising control group. The researchers analyzed changes in circulating BDNF, measured in serum or plasma.
The source material provides access only to the PubMed abstract, not the full text. It therefore does not allow a trial-by-trial breakdown of the populations, programs and their duration, or verification of all methodological choices.
A positive signal, but highly variable results
The average effect on circulating BDNF is positive: the standardized mean difference is 0.41, with a 95% confidence interval of 0.25 to 0.57. This measure allows results expressed on different scales to be pooled. It represents neither a 41% increase nor a measured cognitive improvement.
Results vary substantially between studies. The prediction interval ranges from −0.54 to 1.36: a positive average effect therefore does not guarantee an increase in another training context. The authors also identify a statistical signal of small-study effects, adding further uncertainty.
The effect remains positive when the analysis is restricted to samples collected approximately 24 hours or more after exercise. However, this is not enough to establish how long the change lasts after a program ends.
More exercise or higher intensity does not mean more BDNF
In the joint dose analysis, covering 29 trials, neither total prescribed exercise volume nor vigorous intensity compared with moderate intensity was statistically significantly associated with the effect on BDNF. The other analyses did not establish a dose-response relationship.
This finding does not demonstrate that all doses are equivalent. It means these data cannot be used to choose a volume or intensity specifically to maximize this biomarker.
A blood measurement is not a brain measurement
BDNF is involved in brain plasticity, but its concentration in the blood is not a validated proxy for its concentration in the brain. This review therefore does not demonstrate an improvement in memory or other cognitive functions.
The practical conclusion remains limited: exercise training appears to increase circulating BDNF on average, without justifying a prescription intended to optimize this marker. Funding and conflicts of interest have not been verified in the supplied source material.
Limitations to keep in mind
Certainty of evidence rated very low, substantial variability between studies, methodological limitations and a signal of small-study effects. Only the PubMed abstract was consulted. Funding and conflicts of interest have not been verified.
What we can take away
This review suggests an average increase in blood BDNF with exercise training. It does not establish a brain benefit or identify an optimal exercise dose.
Sources and provenance
- Dose-Response Effects of Exercise Training on Circulating Brain-Derived Neurotrophic Factor (BDNF): A Systematic Review and Meta-analysis of Randomised Controlled Trials.
Source date: 2026-10-26.
Journal Article, Meta-Analysis, Systematic Review, Review · Résumé PubMed uniquement, texte intégral non consulté