Abstract

Compound muscle action potentials (CMAPs) of the lingual muscles were recorded by especially devised bipolar surface electrodes placed on the tongue. Distinct responses were evoked in the tongue muscles by peripheral electrical stimulation of the hypoglossal nerve medial to the angle of the jaw and by transcranial magnetic stimulation of the motor cortex. With cortical stimulation during voluntary activation of the tongue muscles it proved easy to obtain responses with the characteristics of centrally evoked responses allowing reliable measurements of latencies and amplitudes. By contrast, responses from magnetic stimulation of the intracranial segment of the hypoglossal nerve were more difficult to obtain and the reproducibility was often not satisfactory. In a group of 20 healthy subjects the average distal motor latency of both sides from peripheral stimulation was 2.4 ms and the corresponding amplitude was 9.3 mV on the left and 8.6 mV on the right side (range 5.1-16.0 mV). Cortical stimulation gave responses with an average onset latency of 8.6 ms and 8.8 ms and an average amplitude of 1.8 mV and 2.6 mV on the left and right sides of the tongue respectively (range 0.7-5.6 mV). From this mean conduction times of 6.2 ms on the left and 6.4 ms on the right side (SD 1.0 ms) between cortex and mandibular angle and relative amplitudes from cortical stimulation as compared with the peripheral CMAP of 29% on the left and 21% on the right side (range 7%-66%) were calculated. In 16 patients it was possible to differentiate between a central (supranuclear) and a peripheral (intranuclear) site for the lesions of the motors routes to the lingual muscles and to show subclinical lesions in some cases. With a recording arrangement allowing selective unilateral recording of muscle activity from both sides of the tongue the assumed bihemispheric motor representation of the lingual muscles was confirmed.