Abstract |
The emergence of tremor at rest or while a stable posture is maintained are both common signs of Parkinson‟s disease. These involuntary rhythmical oscillations of a body segment are exhibited in a frequency range between 4 to 8 Hz. The neural mechanisms underlying their generation have been studied extensively. The great majority of the researchers have focused their attention on corresponding oscillations at supraspinal sites, namely basal ganglia, motor cortex and thalamus. However, our understanding of these mechanisms remains limited.
In this study, we alternatively examined the firing patterns and synchrony of motor units during limb tremors. These behaviors determine the features of muscle force tremors and associated neurogenic components of limb tremors. They additionally carry information on the tremor-related synaptic input to the motoneurons. We recorded the instantaneous acceleration of the upper limb of Parkinson‟s disease patients and control subjects at rest and while they tried to hold a stable posture. Simultaneously, we recorded surface EMG and single motor unit discharges of upper-limb muscles. The firing patterns were studied via spectral and interspike interval analyses. The synchrony was measured using coherence and cross-correlation computations between motor unit discharges and surface EMG.
Our results demonstrated two auto-spectral components, ca. 1.5-Hz apart, for the patients‟ rest and postural tremor. These components represented different intervals of randomly interchanged activity epochs, termed I and II. Epoch-I resembled physiological tremor in that it was characterized by 6-10 Hz non-overt tremor, motor unit synchrony with normal strength, and rhythmical motor unit firing at the intrinsic rate of the unit. In contrast, epoch-II intervals were characterized by the presence of 4-8 Hz overt tremor, enhanced motor unit synchrony, and spike-doublets or triplets bearing a one-to-one relation to each tremor cycle. The frequency of non-overt and overt tremor, along with the remaining characteristics of epoch-I and epoch-II intervals, did not differ whether a patient was at rest or held a stable posture. Spike-doublets and triplets had incidences obeying a strict rule, and exhibited interspike intervals in the 30-50-ms (beta) range with mean values that remained roughly constant for any given patient, irrespective of motor unit rate and tremor type. On the contrary, these mean values of the interspike intervals were correlated only with the difference between the periods of overt and non-overt tremor.
Overall, our findings reveal shared features of motor unit firing for either of the two epochs whether a patient was at rest or held a stable posture, thus suggesting an encompassing model of alternating, frequency-linked neural generators of tremor. The abnormal motor unit firing in epoch-II intervals underlies the greatly enhanced auto-spectral component which corresponds to overt tremor. It also likely reflects distinct features of the tremor-related synaptic input to the motoneurons, thus providing a useful basis for studying the generator of such tremor. On this basis, we propose that the spinal stretch reflex loop, which in Parkinson‟s disease remains intact, acts as a two-state oscillator. This conception could be succeeded through intermittently descending, supraspinally produced, beta oscillations that interact with the spinal stretch reflex loop, inducing a transition to overt tremor. Other than that, our bottom-up approach could in any case facilitate tremor studies in other movement disorders
|