A systematic review and meta-analysis evaluating the effect of exercise on the development of cancer-related lymphedema

Abstract Background The purpose of this systematic review and meta-analysis was to (i) evaluate effects of exercise on cancer-related lymphedema (CRL) incidence, and (ii) explore whether effect differed according to patient and exercise intervention characteristics. Methods A search of six electronic databases was undertaken to identify intervention studies published up to May 2025. Studies included individuals at risk of, and with CRL, comparing exercise to no exercise, and reporting lymphedema outcomes. Meta-analyses using random effects models estimated the relative risk (RR) of exercise on CRL. Exploratory subgroup analyses were conducted for upper- versus lower-limb lymphedema, 5 or 5+ lymph nodes dissected, and exercise intervention characteristics including exercise mode and degree of supervision. Overall quality of evidence was assessed using the GRADE approach. Results Seventeen studies (published 2002-2024) involving 2739 individuals were included. Most (88%, n = 15) studies focused on upper-limb lymphedema post-breast cancer, and two studies investigated risk of lower-limb lymphedema. With low overall certainty, the RR of developing CRL for those in the exercise group compared with the non-exercise group was 0.71 (95% confidence interval (CI), 0.53 to 0.96). Majority of evidence is derived from studying those at high risk of breast cancer-related lymphedema, but subgroup analyses suggest benefit may extend outside the breast cancer setting. Subgroup analyses support participation in any/all exercise modes, even when unsupervised. Conclusion These findings underscore the promise of exercise for CRL risk reduction and the urgent need for rigorously designed trials to clarify effects across patient risk profiles, cancer types, and exercise approaches. PROSPERO registration number CRD42020196623