Mechanisms of exercise-based rehabilitation following intracerebral hemorrhage: insights from preclinical evidence

Background: Intracerebral hemorrhage (ICH) is associated with high mortality and disability. Although acute treatment has improved, effective interventions for long-term functional recovery remain limited. Exercise rehabilitation is recommended in clinical guidelines to improve outcomes. However, its mechanisms in ICH have not been systematically summarized. This study aimed to review the current preclinical evidence, summarize the main mechanisms of exercise rehabilitation in ICH, and provide a reference for future mechanistic research and clinical translation. Methods: We searched PubMed, EMBASE, and Web of Science from database inception to November 2025. We included studies using rodent ICH models in which exercise or rehabilitation was the main intervention and mechanistic outcomes were reported. Two reviewers independently screened the studies and extracted data. Extracted information included model type, exercise protocol, functional assessment, tissue sampling site, and mechanistic markers. Results: A total of 23 studies were included. All used collagenase-induced striatal ICH models. Exercise protocols were heterogeneous. Treadmill training was the most common intervention (10 studies). Other interventions included skilled reaching training (4 studies), acrobatic or task-based training (3 studies), swimming (2 studies), voluntary wheel running (1 study), constraint-induced movement therapy (CIMT, one study), and one study combining treadmill and voluntary exercise. One study applied treadmill preconditioning before ICH induction. Most studies reported improvement in sensorimotor function. Mechanistic findings mainly clustered into four domains: reduction of neuroinflammation (e.g., decreased IL-1β and reduced glial activation); enhanced neurotrophin-related plasticity (e.g., increased BDNF/TrkB and GAP-43, with improved dendritic and structural markers); changes related to iron/ion toxicity and oxidative injury (e.g., reduced iron load, protein oxidation, and ion imbalance); epigenetic regulation (e.g., increased histone acetylation). Task-oriented training more often reported changes in synaptic and structural plasticity, while aerobic exercise more frequently focused on inflammation and neurotrophin-related markers. Conclusion: Preclinical evidence overall supports that exercise rehabilitation may regulate multiple pathways involved in injury and repair after ICH and is associated with functional recovery. However, there is large variation in exercise type, dose, and timing across studies. Most studies are based on a single animal model. Therefore, the optimal intervention protocol remains unclear. Future studies should standardize reporting of dose and time window and strengthen the link between mechanistic markers and clinical outcomes to improve translational value.