Effects of Three Aerobic Exercise Modalities (Walking, Running, and Cycling) on Circulating Brain-Derived Neurotrophic Factor in Older Adults: A Systematic Review and Meta-Analysis

Yang Cheng¹Yuxiang Liu²Jing Ma¹Zhen Li³Enli Han^2*Shumin Bo^1*

• ¹School of Kinesiology and Health, Capital University of Physical Education and Sports, Beijing, China
• ²College of Physical Education and Sport, Langfang Normal University, Langfang, China
• ³School of Physical Education and Sport Science, Fujian Normal University, Fuzhou, China

Background: Recent studies indicate that older adults (aged 55 years and above) represent a critical period for changes in circulating BDNF levels and cognitive function. Regular aerobic exercise (AE) has been recognized as a promising non-pharmacological strategy to influence neuroplasticity and cognitive function, primarily through the regulation of brain derived neurotrophic factor (BDNF). However, inconsistent findings have been reported regarding the specific effects of three AE modalities—walking, running, and cycling—on circulating BDNF levels in older adults. This study aimed to systematically evaluate the effects of these AE modalities on BDNF levels in the elderly through meta-analysis, and to further compare the relative effectiveness of various exercise protocols. Methods: A systematic search was conducted from database inception to June 10, 2025, in PubMed, Web of Science, Embase, the Cochrane Library, and Scopus. The methodological quality and risk of bias of the included studies were assessed using the Cochrane Risk of Bias 2 tool. Meta-analysis and network meta-analysis were conducted using Stata. Results: A total of 17 studies involving 900 older participants were included. Meta-analysis indicated that three AE modalities significantly increased circulating BDNF levels (P = 0.03). Subgroup analysis revealed that the intervention effect was significantly influenced by participants' health status (P < 0.01). Specifically, the interventions had positive effects in healthy individuals and those with mild cognitive impairment. Exercise-related variables such as modality, intensity, and the interval were identified as potential moderators. Network meta-analysis demonstrated that protocols involving low-intensity short-duration walking (WLS) were superior to other exercise protocols (p < 0.05), and protocols involving moderate-intensity short-duration walking (WMS) were more effective in increasing BDNF levels than high-intensity long-duration walking (P < 0.05). Surface under the cumulative ranking curve results further supported the superiority of WLS (99.9%) and WMS (83.7%) over other exercise protocols. Conclusion: Walking, running, and cycling are effective for improving circulating BDNF levels in older adults; however, the magnitude of improvement depends on participants' health status and specific exercise prescription. Interventions involving walking at low to moderate intensity demonstrated favorable efficacy. This effect may be more favorable in healthy individuals and those with mild cognitive impairment.