Effects of aerobic exercise on learning and memory function and inflammatory markers in animal models of Alzheimer's disease: A meta-analysis

Background Alzheimer's disease (AD) is a progressive neurodegenerative disorder and the leading cause of dementia worldwide. Aerobic exercise (AE) may improve learning and memory function and modulate neuroinflammation in AD animal models, although evidence remains inconsistent. Objective To evaluate the effects of AE on learning and memory function and hippocampal inflammatory cytokines in AD animal models. Methods This systematic review and meta-analysis followed the PRISMA statement. We searched PubMed, Scopus, Embase, Web of Science, and the Cochrane Library, and included 19 animal studies. We used random-effects models to calculate standardized mean differences (SMD). We conducted subgroup analyses by AD modeling methods, exercise modality, AD pathological stage, and intervention duration. We also evaluated oxidative stress-related markers reported in the included studies as supplementary outcomes. Results AE significantly improved spatial memory (platform crossings, SMD = 1.82), spatial learning (escape latency, SMD = −1.48), and recognition memory (novel object recognition, SMD = 2.80). AE reduced hippocampal interleukin-1β (IL-1β) levels (SMD = −3.17); AE showed a trend toward reducing tumor necrosis factor-α (TNF-α) levels, but the effect did not reach statistical significance (SMD = −1.20, 95% CI: −2.44 to 0.04). Supplementary analyses showed that AE increased superoxide dismutase (SOD) activity (SMD = 2.05) and reduced malondialdehyde (MDA) content (SMD = −3.20). Moderate-to-high heterogeneity was observed across studies. Conclusions AE improves learning and memory function and modulates hippocampal inflammation in AD animal models, particularly IL-1β. Supplementary analyses suggest potential benefits for oxidative stress, with increased SOD and reduced MDA.