The impact of exercise on inflammatory biomarkers: an umbrella review of systematic reviews and meta-analyses
TL;DR
It is suggested that exercise is associated with changes in circulating inflammatory biomarkers across diverse populations, though the certainty of the evidence is generally low to very low.
Abstract
Background Chronic low-grade inflammation is a pathophysiological hallmark of various chronic diseases, including cardiovascular diseases, metabolic disorders, and cancers. Regular physical exercise is recognized as an effective strategy to mitigate systemic inflammation; However, existing evidence is primarily derived from systematic reviews focused on specific populations or single biomarkers, resulting in a fragmented evidence base with inconsistent methodological quality. A comprehensive evaluation of the overall quality and certainty of the evidence remains lacking. Methods This umbrella review was conducted in accordance with standard guidelines. A systematic literature search was performed in PubMed, Embase, Web of Science, and the Cochrane Library up to January 2026 to identify meta-analyses examining the effects of exercise interventions on inflammatory biomarkers. Two independent reviewers performed literature screening and data extraction. Methodological quality was assessed using the AMSTAR 2 tool, and the certainty of evidence was graded according to the GRADE framework. The review focused on populations with diverse health statuses, providing a comprehensive evaluation of the impact of exercise on multiple inflammatory biomarkers. Results A total of 61 meta-analyses were included, encompassing 318 meta-analytic associations (unique exercise-biomarker-population combinations). Of these, 145 (45.6%) showed statistically significant effects of exercise on inflammatory markers. The overall certainty of evidence was generally low: only 10 (3.1%) were rated high quality, 89 (28.0%) moderate quality, and the remaining 219 (68.9%) low or very low quality. Significant heterogeneity was observed in the inflammation-modulating effects of exercise across populations with different health statuses. In most chronic diseases, exercise demonstrated a distinct anti-inflammatory trend. However, in patients with chronic obstructive pulmonary disease and fibromyalgia, exercise may transiently elevate certain inflammatory markers. Conclusions This umbrella review suggests that exercise is associated with changes in circulating inflammatory biomarkers across diverse populations, though the certainty of the evidence is generally low to very low. In chronic disease populations, reductions in CRP, IL-6, and TNF-α are generally interpreted as reflecting an attenuation of systemic inflammatory burden, though this interpretation should not be automatically extended to acute post-exercise changes in healthy individuals. Certain exercise modalities showed larger effect sizes in specific populations, but these associations should not be interpreted as evidence of superiority, as the comparisons were indirect, effect sizes may be inflated by the small-study effect, and circulating biomarkers cannot distinguish modality-specific neuromuscular adaptations. Caution is warranted for COPD and fibromyalgia, where exercise may paradoxically trigger acute inflammatory responses. Given the low certainty of evidence and heterogeneity in exercise protocols, the findings should be interpreted as hypothesis-generating rather than definitive, and are intended to inform future research rather than to serve as practice guidelines. Systematic review registration https://www.crd.york.ac.uk/PROSPERO/, identifier, CRD420251236842.