Optimal dose and type of exercise improve the overall balance in adults with Parkinson’s disease: a systematic review and Bayesian network meta-analysis

Abstract

Objective: To investigate the dose–response relationship between various exercise modalities and their effectiveness in enhancing balance in Parkinson's disease (PD) patients. Methods: Searches were conducted in PubMed, Medline, Embase, PsycINFO, Cochrane Library, and Web of Science from inception to July 23th, 2024. Data were analyzed using R software, with the MBNMA and RJAGS packages to compute standardized mean differences (SMD) and 95% credible intervals (95% CrI). The network's risk of bias was rigorously evaluated by two independent reviewers using the ROB2 tool. Results: The analysis included 73 studies with 3,747 PD patients, assessing the effects of different exercises on balance. The findings demonstrate a nonlinear increasing dose–response relationship. Dance was the most effective, with an optimal dose of 1000 METs-min/week, closely followed by aerobic exercise (AE) at 1200 METs-min/week. Aquatic exercise (AQE), multicomponent exercise (MulC), and sensory exercise (SE) exhibited an inverted U-shaped response, with peak improvements at 700 METs-min/week for AQE, 610 METs-min/week for MulC, and 620 METs-min/week for SE. Balance and gait training (BGT), mind–body exercise (MBE), and resistance training (RT) also significantly improved balance, with optimal dosages below 750 METs-min/week. Conclusion: Specific exercises, particularly dance at 1000 METs-min/week, significantly enhance balance in PD patients. Nonlinear dose–response patterns and inverted U-shaped curves were evident, supporting the necessity for personalized exercise prescriptions based on the minimal clinically important difference (MCID) identified for AE, AQE, MBE, MulC, and SE, to optimize therapeutic outcomes. PROSPERO registration number: CRD42024517241.