Effects of Aerobic Exercise and Mind-Body Exercise in Parkinson's Disease: A Mixed-Treatment Comparison Analysis

Abstract

Background/Objectives: Aerobic exercise and mind-body exercise, are vital for improving motor and non-motor functional performance of Parkinson's disease (PD). However, evidence-based recommendations on which type of exercise is most suitable for each individual are still lacking. Therefore, we conduct a network meta-analysis to assess the relative efficacy of aerobic and mind-body exercise on motor function and non-motor symptoms in Parkinson's disease and to determine which of these therapies are the most suitable. Design: A network meta-analysis and dose-response analysis. Setting and Participants: Medline, Embase (all via Ovid), and the Cochrane Central Register of Controlled Trials were comprehensively searched for related trials through April 2021. Measurements: Study quality was evaluated using the Cochrane Risk of Bias Tool. The effect sizes of continuous outcomes were calculated using mean differences (MDs) or standardized mean differences (SMDs). A network meta-analysis with a frequentist approach was conducted to estimate the efficacy and probability rankings of the therapies. The dose-response relationship was determined based on metaregression and SUCRA. Results: Fifty-two trials with 1971 patients evaluating six different therapies were identified. For the UPDRS-motor score and TUG score, yoga all ranked highest (SUCRA = 92.8%, 92.6%, respectively). The SUCRA indicated that walking may best improve the BBS score (SUCRA = 90.2%). Depression, cognitive and activities of daily living scores were significantly improved by yoga (SUCRA: 86.3, 95.1, and 79.5%, respectively). In the dose-response analysis, 60-min sessions, two times a week might be the most suitable dose of yoga for reducing the UPDRS-motor score of PD patients. Conclusion: Yoga and walking are important options for increasing functional mobility and balance function, and yoga might be particularly effective for decreasing depressive symptoms and cognitive impairment and improving activities of daily living in PD. The potential optimal dose of yoga for enhancing motor ability in PD patients is 60-min sessions, two times a week. Registration: PROSPERO CRD42021224823.

Keywords: Parkinson’s disease; aerobic exercises; dose response; mind-body exercise; network meta-analysis.

FIGURE 1

Pairwise meta-analysis of aerobic and mind-body therapies on motor outcomes. (A) UPDRS-motor outcome. (B) TUG outcome. (C) BBS outcome. Meta-analysis results for pair-wise comparisons represented by MD and 95% credible interval (CrI).

FIGURE 2

Network meta-analysis of the efficacy of exercise therapies on motor outcomes. (A) UPDRS-motor outcome. (B) TUG outcome. (C) BBS outcome. MD and 95% credible interval (CrI) estimations were calculated as column-defining interventions compared with row-defining interventions. Significant results were labeled with bold, red and underlined.

FIGURE 3

The rank probability of various interventions based on the SUCRA. (A–C) Rank probability and mean rank on UPDRS-motor outcome. (D–F) TUG outcome. (G–I) BBS outcome. The greater the SUCRA value is, the better the rank.

FIGURE 4

Dose response of yoga using meta-regression and SUCRA on UPDRS-motor outcome. (A–D) The potential optimal time of yoga per session in UPDRS motor scale; (E–H) The potential optimal frequency of yoga each week in UPDRS motor outcome. (A,E) Dose-response curve for yoga using metaregression. (B–D,F–H) Dose response of yoga based on the rank probability and mean rank. The greater the SUCRA value is, the better the rank.

FIGURE 5

Clustered ranking plot of the network. (A) For UPDRS-motor&TUG outcomes. (B) For UPDRS&BBS. (C) For TUG&BBS outcomes. The plots are based on cluster analyses of SUCRA values. Each plot shows the SUCRA values for two different outcomes. The therapies in the upper right corner are more effective than the other therapies.