Comparative efficacy of different rehabilitation exercise interventions on upper limb function and lymphedema in postoperative breast cancer patients: a network meta-analysis

Background While surgery is a cornerstone of breast cancer (BC) treatment, it frequently leads to postoperative complications such as upper limb dysfunction and lymphedema(LE). These sequelae are prevalent and profoundly diminish patients’ recovery and quality of life (QoL). Although rehabilitation exercise has proven effective, the comparative superiority of different exercise protocols remains to be elucidated. Objective This study aimed to first systematically evaluate the aggregate efficacy of rehabilitation exercise interventions for postoperative breast cancer patients. Subsequently, we employed a network meta-analysis to delineate the comparative efficacy of different rehabilitation modalities, with the ultimate goal of informing the identification of an optimal protocol. Methods We systematically searched seven English and Chinese databases, including PubMed, Embase, and the Cochrane Library, from their inception to December 2025. Randomized controlled trials (RCTs) evaluating rehabilitation exercise interventions were included. The Cochrane Risk of Bias 2 (RoB-2) tool was used to assess bias risk, and R software was employed for the network meta-analysis. Continuous outcomes were expressed as mean difference (MD) or standardized mean difference (SMD), and dichotomous outcomes as odds ratios (OR). The Surface Under the Cumulative Ranking Area (SUCRA) was used to rank interventions. This study protocol was registered with PROSPERO (CRD420251243892). Results A total of 14 RCTs involving 2207 patients were included. For reducing lymphedema incidence, SUCRA-based ranking favored Structured Self-Management/Education (SSM/E, SUCRA = 98.5%), followed by Progressive Resistance Training (PRT, SUCRA = 75.2%) and Comprehensive Exercise (CE, SUCRA = 35.1%). Although SSM/E ranked highest according to SUCRA values, this result was based on evidence from only one small randomized controlled trial (RCT), and the certainty of the evidence for SSM/E was low. Network meta-analysis suggested that SSM/E improved upper limb function (DASH score), however, this finding was derived from only two studies. Due to substantial heterogeneity in measurement tools and reporting methods, other outcomes were synthesized qualitatively. The narrative evidence suggested that CE may improve shoulder range of motion, and that SSM/E, PRT, and CE may improve quality of life. The evidence regarding pain relief was inconclusive. Conclusion Rehabilitation exercise interventions have been associated with improved functional recovery, reduced risk of LE, and enhanced QoL in postoperative BC patients. SSM/E showed potential benefits in reducing LE incidence; however, this finding was based on only one RCT and should be interpreted with caution. PRT showed more consistent effects across the included comparisons. More high-quality RCTs with standardized intervention protocols and outcome assessments are required to establish the relative effectiveness of rehabilitation strategies. Systematic Review Registration https://www.crd.york.ac.uk/PROSPERO/ , identifier CRD420251243892.