THE MYOCLONUS IN CORTICOBASAL DEGENERATION - EVIDENCE FOR 2 FORMS OF CORTICAL REFLEX MYOCLONUS

The clinical and physiological characteristics of myoclonus in 14 patients with corticobasal degeneration are described. The myoclonus was focal, confined to one limb (usually the arm) and was most prominent on voluntary action or in response to sensory stimulation. On clinical inspection, the myoclonus appeared to occur at rest but EMG recordings revealed that apparently spontaneous myoclonus occurred only on a background of move or less continuous muscle activity (responsible for the rigidity and dystonia). The jerks consisted of hypersynchronous short duration bursts of EMG activity coincident in agonists and antagonists. Reflex myoclonus in hand muscles, to stimulation of the median nerve at the wrist, had a latency of similar to 40 ms. In 13 of the 14 patients reflex myoclonus was not associated with enlargement of the cortical sensory evoked potentials (SEPs); the later components of the parietal SEP were poorly formed and dominated by a broad positive wave with a peak latency similar to 45 ms. Prefrontal components of the SEP were relatively preserved, but there were no significant differences between the SEPs evoked from myoclonic and non-myoclonic limbs. Action myoclonus was not preceded by an identifiable cortical wave in the electroencephalogram back-aver aged before each jerk. Magnetic, but not electric, brain stimulation evoked repetitive bursts of myoclonus suggesting enhanced cortical excitability. The combination of focal, predominantly distal, hypersynchronous jerks, evidence of enhanced cortical excitability, together with the known cortical pathology in corticobasal degeneration suggests that the myoclonus in these patients may be cortical in origin. Since the latency of reflex myoclonus in corticobasal degeneration is only 1-2 ms longer than the sum of the afferent and efferent times to and from the cortex, we propose the reflex myoclonus is mediated by direct sensory input to motor cortical areas that activate corticospinal tract output. Such myoclonus differs from the typical form of cortical reflex myoclonus in which reflex jerks have a longer latency (50 ms in hand muscles), cortical SEPs are enlarged and action myoclonus is preceded by a cortical discharge. It is proposed that these various forms of cortical myoclonus can be explained by the presence of different cortical relays of sensory information to cortical motor areas. The myoclonus of corticobasal degeneration may represent enhancement of a direct sensory input to the motor cortex. In contrast, the more widely recognized variety of cortical reflex myoclonus may involve abnormal relays through sensory cortex to motor cortex, either directly or via cerebellar-thalamo-cortical projections.