Exercise-Induced Modulation of Inflammatory Biomarkers After Stroke: A Systematic Review and Meta-Analysis

OBJECTIVE: The purpose of this study was to perform a systematic review of the literature on the effectiveness of physical exercise on cytokine levels after stroke. METHODS: The PubMed, LILACS, ScienceDirect, Web of Science, Scopus, and Embase databases were searched without language or date restriction through September 2025 using the terms "stroke" and "exercise" and "cytokines" and related terms. Eligible studies included patients with stroke (P), investigated aerobic or multimodal exercise interventions (I), compared them with a control group (C), and reported outcomes on cytokines or other systemic inflammatory biomarkers (O). To assess the risk of bias in each study, we used the criteria described in the Cochrane Handbook for Systematic Review of Interventions. Fixed-effect meta-analysis were performed for homogeneous outcomes. RESULTS: In total, 8 randomized clinical trials were included for qualitative analysis, and in the meta-analysis. The meta-analysis showed that aerobic training significantly reduced IL-6 (SMD -0.51; 95% CI -0.86 to -0.15; P = .006) and hs-CRP (SMD -1.46; 95% CI -2.18 to -0.75; P < .001), while no significant effect was observed for TNF-α (SMD -0.11; 95% CI -0.53 to 0.31; P = .60). Heterogeneity was high across studies. CONCLUSIONS: Exercise interventions were associated with reductions in cytokine levels, particularly IL-6 and hs-CRP. These effects were mainly observed after aerobic training. Given the limited number of trials and the substantial heterogeneity, these findings should be interpreted with caution.