Subdivision of essential tremor patients according to physiologic characteristics

OBJECTIVE: To test the effect of different forearm postures on tremor characteristics. We hypothesized that changes in tremor characteristics in relation to posture can subclassify essential tremor (ET) patients. METHODS: Fourteen ET patients were tested while seated and holding a full cup in three well-defined arm postures: 15 cm above the armrest, elbow at 90 degrees flexion and when the cup was near the mouth. Hand movements were recorded using a triaxial accelerometer. Concomitantly, we recorded surface electromyogram (EMG) signals from the wrist extensor muscles. Hand acceleration and the primary tremor frequency from the power spectrum were calculated for each posture in two independent trials. The coherence at the primary tremor frequency between the EMG and the accelerometry signals was calculated. RESULTS: ET patients could be classified into two groups: ET-1 (nine patients) had position-dependent peak frequencies while ET-2 (five patients) had position-independent peak frequencies. The latter group had significantly higher coherencies between EMG and accelerometry of tremor and insignificantly higher tremor amplitudes compared with ET-1. In both groups tremor amplitude increased when the hand was near the mouth. CONCLUSION: The results suggest that ET patients can be classified based on the position dependence of frequency and other physiologic properties.