Effects of a Highly Challenging Balance Training Program on Motor Function and Brain Structure in Parkinson's Disease

Abstract

Background: Parkinson's disease (PD) is characterized by motor deficits and brain alterations having a detrimental impact on balance, gait, and cognition. Intensive physical exercise can induce changes in the neural system, potentially counteracting neurodegeneration in PD and improving clinical symptoms.

Objective: This randomized controlled trial investigated effects of a highly challenging, cognitively demanding, balance and gait training (HiBalance) program in participants with PD on brain structure.

Methods: 95 participants were assigned to either the HiBalance or an active control speech training program. The group-based interventions were performed in 1-hour sessions, twice per week over a 10-week period. Participants underwent balance, gait, cognitive function, and structural magnetic resonance imaging assessments before and after the interventions. Voxel-based morphometry was analyzed in 34 HiBalance and 31 active controls. Additionally, structural covariance networks were assessed.

Results: There was no significant time by group interaction between the HiBalance and control training in balance, gait, or brain volume. Within-HiBalance-group analyses showed higher left putamen volumes post-training. In repeated measures correlation a positive linear, non-significant relationship between gait speed and putamen volume was revealed. In the HiBalance group we found community structure changes and stronger thalamic-cerebellar connectivity in structural covariance networks. Neither brain volume changes nor topology changes were found for the active controls after the training.

Conclusion: Thus, subtle structural brain changes occur after balance and gait training. Future studies need to determine whether training modifications or other assessment methods lead to stronger effects.

Keywords: Randomized controlled trial; gait; gray matter; idiopathic Parkinson’s disease; magnetic resonance imaging; physical exercise; putamen.

Fig. 1

Flow chart of stepwise exclusion of participants for the longitudinal sMRI cohort of the EXPANd trial. PD, Parkinson’s disease; sMRI, structural magnetic resonance imaging.

Fig. 2

Increased gray matter volume in the HiBalance cohort compared before and after the training. Cluster is uncorrected for multiple comparisons, p < 0.001 with a cluster forming threshold of k = 100 voxels. L left.

Fig. 3

Nodal degree in the HiBalance cohort. Left side: changes in degree significant in three nodes (turquoise). In red are the connections shown that differed between before and after the HiBalance training program. Right side: diagram showing number of connections of the significant nodes before and after the training program. R, right; L, left.

Fig. 4

Graph theory measures in the HiBalance training program cohort (left and middle column) and active control group (right column). Changes in community structure from pre to post- training program within the HiBalance cohort (left and middle column). There were no changes in the active control group (right column). The red dot with a black border shows the left putamen, which showed increased volume after training in the HiBalance cohort. Communities are depicted in different colors. Red dots show the community the left putamen belongs to (switching from community 2 pre-training to community 1 post-training). Community 1 is shown in purple pre-training and in red post-training. Community 2 is red pre-training and blue post-training. Community 3 is depicted in green. Community 4 is shown in blue pre-training and purple after-training. L left.

Fig. 5

Repeated measures correlation for the HiBalance training (left) and active control (right) program. Correlations of pre and post-training measures for balance performance as measured by MiniBESTest scores and gait speed (m/s). Gray matter volumes were extracted from the cluster obtained by the comparison of HiBalance pre < post for all participants pre and post-training.

Fig. 6

Partial Pearson’s correlation for the HiBalance training (left) and active control (right) program. Correlation between post-training gray matter values extracted at the left putamen and gait speed (m/s) as measured after training controlled for total intracranial volume.