Functional neuroimaging of motor control in Parkinson's disease: a meta-analysis

Abstract

Functional neuroimaging has been widely used to study the activation patterns of the motor network in patients with Parkinson's disease (PD), but these studies have yielded conflicting results. This meta-analysis of previous neuroimaging studies was performed to identify patterns of abnormal movement-related activation in PD that were consistent across studies. We applied activation likelihood estimation (ALE) of functional neuroimaging studies probing motor function in patients with PD. The meta-analysis encompassed data from 283 patients with PD reported in 24 functional neuroimaging studies and yielded consistent alterations in neural activity in patients with PD. Differences in cortical activation between PD patients and healthy controls converged in a left-lateralized fronto-parietal network comprising the presupplementary motor area, primary motor cortex, inferior parietal cortex, and superior parietal lobule. Both, increases as well as decreases in motor cortical activity, which were related to differences in movement timing and selection in the applied motor tasks, were reported in these cortical areas. In the basal ganglia, PD patients expressed a decrease of motor activation in the posterior motor putamen, which improved with dopaminergic medication. The likelihood of detecting a decrease in putaminal activity increased with motor impairment. This reduced motor activation of the posterior putamen across previous neuroimaging studies indicates that nigrostriatal dopaminergic denervation affects neural processing in the denervated striatal motor territory. In contrast, fronto-parietal motor areas display both increases as well as decreases in movement related activation. This points to a more complex relationship between altered cortical physiology and nigrostriatal dopaminergic denervation in PD.

Keywords: Parkinson's disease; functional magnetic resonance imaging; meta analysis; motor; positron emission tomography.

Copyright © 2013 Wiley Periodicals, Inc.

Figures

Figure 1

Convergence of activation maxima for the group comparison between PD patients and healthy controls. A: Significant results for the contrast PD‐OFF vs. Controls (i.e., “PD‐OFF > Controls” or “Controls > PD‐OFF”). B: Decreased activation in PD‐OFF compared to Controls. C: Increased activation in PD‐OFF compared to Controls. D: Decreased activation in PD‐ON compared to Controls. Significant activation maxima for the contrast “PD‐ON vs. Controls” are omitted, since they are identical to results of the contrast “Controls > PD‐ON” shown in 1D. L, left; R, right. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 2

Convergence of activation maxima for the comparison between PD patients ON and OFF medication. A: Significant results for the contrast PD‐ON vs. PD‐OFF (i.e., “PD‐ON > PD‐OFF” or “PD‐OFF > PD‐ON”). B: Increased activation in PD‐ON compared to PD‐OFF. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 3

Voxel‐wise Spearman correlations between activation likelihood and UPDRS‐III scores. There was a significant correlation between likelihood of decreased activity in the right Putamen in PD patients (“Control > PD‐OFF”) and motor impairment (rho = 0.63, P < 0.05). Studies that do not report activations of the respective region have an MA‐value of 0. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]