ABSTRACT: Tremor is one of the cardinal symptoms of Parkinson's disease. Up to now, however, its pathophysiology remains poorly understood. Previously, oscillatory coupling at tremor frequency between the subthalamic nucleus und affected muscles was shown. In these studies, however, causality of coupling could not be demonstrated. Thus, we analyzed the statistical causality between intraoperatively recorded local field potentials in the subthalamic area and affected arm muscles during tremor episodes, using squared partial directed coherence, a recently developed causality measure. The analysis identified differential statistical causality patterns for Parkinson's disease patients of the akinetic-rigid subtype during tremor episodes (n=6) versus patients of the tremor-dominant subtype (n=8): for the akinetic-rigid Parkinson's disease patients significantly more cases of the subthalamic region were found to be statistically causal for electromyographic-tremor activity, a result in accordance with the standard basal ganglia model. In contrast, for the tremor-dominant patients, significantly more instances of electromyographic tremor activity turned out to be causal for activity of the subthalamic region. Furthermore, the clinical effective stimulation site coincided with the location of most input causalities from the periphery in seven out of eight tremor-dominant patients. The data suggest that, although tremor activity in tremor-dominant and akinetic-rigid Parkinson's disease patients was clinically similar, statistical causality between tremor electromyogram (EMG) and the subthalamic nucleus was fundamentally different. Therefore, we hypothesize different pathophysiological mechanisms to underlie the generation of tremor in the two subtypes of Parkinson's disease.
Neuroscience 03/2012; 210:353-62. · 3.38 Impact Factor