Physical activity and oxidative stress biomarkers in humans: a systematic review and quantitative meta-analysis

Abstract

Abstract Background The effects of physical activity on oxidative stress biomarkers in humans remain heterogeneous. This systematic review and meta-analysis quantified exercise-induced changes in key redox biomarkers. Methods Following PRISMA 2020 (PROSPERO: CRD420251144711), a systematic search of PubMed, Scopus, and Web of Science (2000–2025) identified human studies assessing malondialdehyde (MDA), superoxide dismutase (SOD), catalase (CAT), glutathione (GSH), total antioxidant capacity (TAC), and total antioxidant status (TAS). Random-effects models estimated pooled standardized mean differences (SMD) with 95% confidence intervals (CI). Subgroup analyses, meta-regression, sensitivity analyses, publication bias, and risk-of-bias assessments were fully performed. Results Among 98 eligible studies (up to 31 meta-analyzed, n = 4,742), physical activity significantly reduced MDA (SMD = -1.02; 95% CI: -1.95 to -0.09; p = 0.032; I² = 94.6%) and increased TAS (SMD = 0.76; 95% CI: 0.38 to 1.15; p < 0.001; I² = 0.0%). GSH showed a non-significant trend toward reduction (SMD = -0.58; p = 0.075; I² = 88.5%), while CAT, SOD, and TAC demonstrated no significant pooled effects. Subgroup analyses identified effect modification by age, BMI, and duration, though meta-regression found no linear relationships with age or duration. Sensitivity analyses confirmed robustness for SOD, GSH, and TAS, while MDA, CAT, and TAC were influenced by influential studies. Publication bias was detected for GSH (Egger’s p = 0.006). Heterogeneity was substantial (I² = 87–97%), and most studies were of moderate methodological quality. Conclusion Physical activity selectively reduces lipid peroxidation and enhances non-enzymatic antioxidant capacity without uniformly upregulating enzymatic defenses. However, due to substantial heterogeneity, moderate methodological quality, and potential publication bias for GSH, these findings should be interpreted cautiously as exploratory. Trial registration PROSPERO CRD420251144711 .