Sports Med. 2026 Apr 18. doi: 10.1007/s40279-026-02434-3. Online ahead of print.
ABSTRACT
BACKGROUND: The lasting effects of exercise on brain health in older adults are poorly documented.
OBJECTIVE: The aim of this study was to assess the long-term, post-intervention effects 4 years after a 5-year exercise intervention on brain volumes and cognitive abilities in older adults.
METHODS: A total of 106 older adults (70-77 years at baseline) from the randomized controlled trial (RCT) Generation 100 study randomized into high-intensity interval training (HIIT), moderate-intensity continuous training (MICT), or a control group following national physical activity guidelines for 5 years were included. Post-intervention assessment was performed 9 years after inclusion. Brain magnetic resonance imaging (MRI) and exercise characteristics were acquired at baseline and after 1, 3, 5, and 9 years. Cognitive abilities were assessed with verbal list learning, pattern separation, and odor identification tests at baseline and 9 years. Linear mixed models were used to examine the interaction between exercise group and time on brain volumes. The analyses were repeated, including self-reported exercise behaviors (intensity and duration) as covariates. To examine the general effect of exercise intensity independent of group, a separate model was run across all participants with self-reported exercise intensity as independent variable. For cognition, linear regressions examined associations between exercise group and self-reported exercise intensity on cognitive scores at 9 years, and hippocampal volume loss over time on cognitive scores at 9 years. Whether baseline peak oxygen uptake (VO2peak) predicted brain volumes and cognitive scores at 9 years was tested with linear regression.
RESULTS: The time × group interaction revealed greater hippocampal volume loss in the HIIT compared with the control group 4 years after the intervention ended, similar to what was found during the intervention. We did not find that self-reported exercise behavior was associated with hippocampal volume over time. Across all participants, greater hippocampal volume loss was associated with poorer verbal memory at 9 years. The test scores on verbal memory and pattern separation did, however, not differ between the groups, but the MICT, and trending for the HIIT group, had higher odor identification scores at 9 years. Higher baseline VO2peak predicted larger cortical volume and better pattern separation score at 9 years across intervention groups, in line with what was observed during the intervention.
CONCLUSIONS: The control group, which followed the national physical activity guidelines, had the lowest hippocampal volume loss over time, even 4 years after the end of a 5-year exercise intervention in fit older adults. Furthermore, higher baseline VO2peak provided protective effects on both cortical volume and pattern separation ability across almost a decade, independent of exercise group.
CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov NCT01666340 (registered 16 August 2012).
PMID:41999477 | DOI:10.1007/s40279-026-02434-3
