Effectiveness of different exercise interventions on balance and cognitive functions in stroke patients: A network meta-analysis.

Abstract

Background: Exercise interventions are proven to improve functional outcomes in stroke patients, yet the optimal and safest exercise modalities remain uncertain. This network meta-analysis (NMA) aims to systematically compare the effects of various exercise interventions on balance and cognitive functions, providing robust evidence to guide clinical decision-making.

Methods: Web of Science, PubMed, Embase, and Cochrane Library were searched up to September 2024. Randomized controlled trials (RCTs) evaluating exercise interventions for balance and cognitive improvements in stroke patients were included. Quality assessment and data extraction were performed, followed by Bayesian NMA using Stata 15.0 and R 4.41.

Results: This study ultimately included 40 RCTs with 2,302 patients. Six commonly employed exercise interventions in clinical treatment were covered, including aerobic exercise (AE), core stability exercise (CSE), physical/mental exercise (PME), resistance training (RT), high-intensity interval training (HIIT), and mixed-component exercise (Mixed). According to the surface under the cumulative ranking curve (SUCRA), Mixed was the most effective intervention for improving Berg balance scale scores (SUCRA = 82.89%). AE was the most effective intervention for improving patients' performance on the timed up and go test (SUCRA = 88.46%). PME exhibited superior effectiveness in improving Montreal cognitive assessment scores (SUCRA = 86.43%).

Conclusions: Mixed and AE noticeably improves balance function in stroke patients, while PME and AE notably enhance cognitive function. The efficacy of other forms of exercise requires further validation. For patients whose primary objective is to improve balance, we recommend prioritizing Mixed. In cases of markedly impaired physical function, a single type of exercise should be selected. For patients aiming to enhance cognitive function, we recommend the selection of PME as the preferred option.

Trial registration: Registration date: 23 September 2024. PROSPERO registration number: CRD42024593741.