Oscillatory Communication between Cerebral Areas Associated with Essential Tremor

With an estimated prevalence of about 5%, essential tremor (ET) is the most frequent movement disorder. Nevertheless, the pathophysiological mechanisms giving rise to this usually bilateral 5–9Hz action tremor which mostly affects the upper extremities are still largely unknown. There is accumulating evidence that ET is at least partly of central origin. Animal models suggest an involvement of the inferior olive. PET and fMRI studies showed activation in several brain areas including cerebellum, thalamus and primary motor cortex, and an EEG-EMG coherence study demonstrated involvement of the primary sensorimotor cortex. In addition, the effect of deep brain stimulation in the thalamus of patients suffering from essential tremor reduces the action tremor, pointing towards an important involvement of the thalamus in the generation or mediation of tremor. We studied 8 patients with essential tremor by recording simultaneously neural activity with a whole-scalp neuromagnetometer (MEG) and peripheral tremor activity with surface electrodes (EMG). Subjects performed an isometric contraction of the left forearm for a total duration of about four minutes with interleaved rest periods of 30 seconds. All subjects showed a tremor at a frequency of 5–7Hz. Tremor frequency and its first harmonic were clearly evident in power spectra of the EMG recordings. We used the localization technique DICS to identify cerebral areas with significant coherence to EMG at tremor frequency and its first harmonic. Coherence is a correlation measure in the frequency domain which is normalized between 0 and 1. All subjects showed significant coherence between EMG of the wrist extensor muscle and contralateral primary motor cortex. All but one subject showed significant coherence between EMG and ipsilateral cerebellum. Phase delays between cerebral areas and EMG at the tremor frequency and its first harmonic were computed and indicated the existence of both afferent and efferent components in the coupling. In a further step DICS was used to identify areas of significant cerebro-cerebral coherence. The analysis revealed a network of areas consisting of contralateral primary motor cortex, premotor cortex, thalamus, brainstem and ipsilateral cerebellum which showed strongest coupling at twice the tremor frequency. Partial coherence analysis excluded the possibility that one central driving oscillator simply entrains oscillations in these areas leading to spurious coherences. These results are consistent with the view that in ET patients a brainstem-cerebello-thalamo-premotor-motor cortical network shows oscillatory interactions which lead to a rhythmic modulation of muscle activity becoming apparent as tremor.