OBJECTIVE: This systematic review and network meta-analysis aims to synthesize and compare the evidence on the effects of different exercise interventions on postural balance and activity ability in stroke survivors, and to determine their relative effectiveness through probability ranking, to provide to offer preliminary evidence for clinical practice in post-stroke rehabilitation. METHODS: A systematic literature search was performed across the following electronic databases from inception until January 2026, China National Knowledge Infrastructure, Wanfang Data, Chinese Scientific Journal Database, Cochrane Library, Embase, PubMed, Web of Science, and MEDLINE. Version 2 of the Cochrane risk-of-bias tool for randomized trials (RoB-2) was used to assess the quality of included studies. The study protocol was registered a priori with PROSPERO CRD420251031404. RESULTS: This study included a total of 24 randomized controlled trials, which compared the effects of six types of exercise interventions on the postural balance and activity ability of stroke patients: robot-assisted training, treadmill training, core stability training, resistance training, aquatic therapy, and overground gait. The analysis demonstrated that exercise interventions significantly improved postural balance (SMD = 0.27, 95% CI: 0.03, 0.51) and activity ability (SMD = 0.96, 95% CI: 0.15, 1.76) of stroke patients. Probability ranking further indicated that, compared to other exercise interventions, core stability training showed a superior potential in enhancing both postural balance (SUCRA = 75.4%; SMD = 0.50, 95% CI: 0.04, 0.96) and activity ability (SUCRA = 96.0%; SMD = 2.68, 95% CI: 0.66, 4.70). CONCLUSIONS: This network meta-analysis indicates that core stability training shows the greatest potential in improving postural balance and activity ability in stroke patients. In clinical rehabilitation practice, it may be considered as a preferred option; however, given the limitations of evidence certainty, these findings should be interpreted cautiously and further validated by high-quality studies.
Shi et al. (Sun,) studied this question.