Associations between incremental exercise capacity and multimodal brain characteristics across training levels

Previous work has shown that endurance training is associated with structural and functional plasticity in prefrontal, hippocampal, cerebellar and network level regions using both cross sectional and longitudinal designs. The present study extends this foundation by examining how key endurance indicators relate to multimodal neural characteristics across different training levels.

Thirty-two male participants were classified into high level endurance athletes, moderately trained runners and healthy active controls. All participants completed structural MRI and resting state functional MRI together with graded exercise testing to assess maximal oxygen consumption, relative maximal oxygen consumption, lactate threshold and individual lactate threshold (ILT). Gray matter volume, fractional amplitude of low frequency fluctuations and degree centrality were extracted from predefined regions of interest. Correlation analyses were first conducted in the full sample and within subgroups, followed by false discovery rate correction within imaging modalities. Associations surviving FDR correction were further evaluated using Spearman rank correlations, bootstrap confidence intervals for subgroup findings, and covariate-adjusted partial correlations as sensitivity analyses.

Across all participants, incremental exercise-capacity indicators were associated with multimodal neural characteristics in prefrontal, premotor, precuneus, temporal, and posterior cerebellar regions, with additional negative associations observed in the thalamus and inferior cerebellum in the uncorrected analyses. After FDR correction, the surviving associations became more selective. Subgroup analyses suggested descriptively different association patterns across training levels, but these subgroup findings were less stable in sensitivity analyses and should therefore be interpreted cautiously.

Incremental exercise-capacity indicators were associated with distributed multimodal brain characteristics, and several core associations remained robust after multiple-comparison correction and supplementary analyses. The subgroup results suggested stage-related differences in brain-physiology coupling, particularly a broader pattern in moderately trained individuals and a more focal prefrontal-associated profile in high-level athletes, although these findings still require cautious interpretation.