Exercise, physical activity, physical fitness, and humoral responses to influenza vaccination in adults: a systematic review and meta-analysis

Background Exercise, habitual physical activity, and physical fitness have been proposed as behavioral adjuvants that could improve influenza vaccine immunogenicity, but individual studies are small and inconsistent. Objective To evaluate the effects of assigned acute and chronic exercise interventions, and the associations of habitual activity and fitness, with humoral responses to influenza vaccination in adults. Methods We searched PubMed, Web of Science, Embase, Cochrane Library, Scopus, and SPORTDiscus from inception to May 1, 2026. Randomized evidence was analyzed separately for acute exercise and chronic training; nonrandomized interventions and observed activity or fitness were considered separately. Risk ratios (RRs) and mean differences (MDs) in log 2 hemagglutination inhibition (HAI) titers were synthesized when compatible numerical data were available. Design-appropriate risk-of-bias tools and GRADE were applied. Results Twenty-two reports were included. Seven randomized reports provided primary quantitative estimates, and five contributed to at least one meta-analysis of two or more trials. Acute exercise did not show a clear effect on H1N1 seroprotection (RR 0.88, 95% CI 0.65–1.21; k = 3) or H3N2 seroprotection (RR 1.04, 0.91–1.20; k = 3), or on H1N1 seroconversion (RR 1.30, 0.58–2.92; k = 2) or H3N2 seroconversion (RR 0.84, 0.50–1.40; k = 2). Chronic-training titer estimates were imprecise for H1N1 (MD 0.50 log 2 units, −0.64 to 1.64; k = 2), H3N2 (0.27, −0.17 to 0.72; k = 2), and influenza B (−0.14, −0.64 to 0.36; k = 2). Other primary comparisons were supported by single trials. Certainty was low or very low. Conclusion The available randomized evidence does not establish an immunogenicity benefit of acute exercise or chronic training, but does not demonstrate equivalence or exclude benefit. Observational associations remain vulnerable to confounding. Larger trials with complete numerical reporting are needed; these antibody findings do not establish effects on clinical influenza. Systematic Review Registration https://www.crd.york.ac.uk/PROSPERO/view/CRD420261425732 , identifier CRD420261425732.