The purpose of this study was to provide clear evidence in support of the use of exercise to improve depressive symptoms in patients with Parkinson disease (PD) and to investigate whether this effect differs by exercise type and intensity.

Three independent reviewers searched for randomized controlled trials (RCTs) that applied exercise interventions with depressive symptoms as an outcome measure for patients with PD on PubMed and Web of Science up to February 28, 2022. Random-effects meta-analyses were performed, in which standardized mean differences (SMDs) between the effects of exercise and control interventions on depressive symptoms with 95% CIs were calculated.

A total of 19 RCTs including 1,302 patients with PD were eligible for meta-analysis, and we obtained 23 comparisons from the included studies for data synthesis. Physical exercise interventions showed significant effects on the reduction in depressive symptoms in patients with PD (SMD = 0.829; 95% CI = 0.516-1.142; < 0.001). Moderator analyses on exercise type revealed significant positive effects for combined exercise interventions (SMD = 1.111; 95% CI = 0.635-1.587; < 0.001), whereas aerobic training alone failed to show significant effects (SMD = 0.202; 95% CI = -0.045 to 0.449; = 0.108). Both light-to-moderate intensity exercises (SMD = 0.971; 95% CI = 0.521-1.421; < 0.001) and moderate-to-vigorous intensity exercises (SMD = 0.779; 95% CI = 0.407-1.152; < 0.001) significantly improved depressive symptoms with a small difference between the exercise intensities.

Our results suggest that physical exercise has significant antidepressant effects in patients with PD. These effects seemed to be more closely associated with exercise type than intensity. Different types of exercise interventions may result in greater benefit and require further investigation.