Does Exercise Attenuate Disease Progression in People With Parkinson's Disease? A Systematic Review With Meta-Analyses.

Only nine of the 40 trials measured anything after the exercise stopped, and the authors did not analyse follow-up results, so nothing here says whether any change lasts. Planned analyses of exercise dose and disease severity could not be done: dosages differed by exercise type, and trials mixed people with mild and moderate disease. Everyone in the review had mild to moderate disease, so it says nothing about people with advanced Parkinson's disease, where the authors themselves argue there is probably less value in trying to slow progression. Most programmes lasted twelve weeks or less (38 of 60 comparisons), so the long-term picture is unknown. Trials measured many different outcomes and few measured the same one, so most outcomes could not be pooled at all and were described in words instead. The reviewers could not tell whether the improvements in 'off' motor scores reflect a change in the disease itself or simply better motor performance from training. The certainty of the evidence was rated low for the motor score result and very low for the BDNF result.

The authors declared no conflicts of interest and said they received no financial support for the research, authorship or publication of the article.

Do not read this as proof that exercise slows Parkinson's disease. The pooled improvements in 'off'-medication motor scores and in BDNF were rated low and very low certainty, and exercise had no clear effect on inflammatory or oxidative markers or on brain volume. The positive result for aerobic exercise on its own comes from only 3 comparisons, and the authors themselves say it should be read with caution.