Statistical Analysis

We used linear mixed model analysis based on multiple imputations to evaluate the robustness of the results and address the missing data. Linear mixed model analyses for repeated measures were performed for each of the outcome measures, with group and time entered as fixed effects. Post hoc analyses with the Bonferroni correction for multiple comparisons were performed when statistical significance was observed in the repeated measures analysis. Two-sided P values less than 0.05 were considered statistically significant. Analyses were performed using the Statistical Package for the Social Sciences (SPSS) software version 25. A sample size was calculated to detect a between-group difference of 12 in the primary outcome, UPDRS with a standard deviation of 16.3, a type I error of 5% and a power of 80% [22]. At least 43 patients were needed in each group to allow a dropout rate of 25%.