Electrophysiological studies of long-term electrical stimulation of the cerebellum in monkeys.

Six rhesus monkeys were stimulated on the paravermal cortex for 205 hours (18 days) with different charge densities in order to determine the electrode performance and neural damage that may result from long-term cerebellar stimulation comparable to that being used in man. The electrode-tissue interface was relatively stable and no neural damage was found when the charge/phase (0.5 muC/ph) or charge density (7.4 muC/sq cm/ph) was very low. At all higher charge levels tested (2.4, 4.8, 10, and 22 muC/ph), changes in the electrode-tissue interface, meningeal encapsulation, and neural damage were directly related to the charge density delivered. Unstimulated electrodes on the opposite paravermal cortex exhibited mild tissue reactivity and cell damage, probably due to mechanical compression of the molecular layer and pial vessels. Motor cortex field potentials could be evoked by charges as low as 0.1 muC/ph delivered to paravermal cortex; for a given charge/phase longer pulses were more effective than short pulses. After neural damage resulting from 205 hours of 4.8 muC pulses at 10 per second (total charge 14.76 C), the threshold for the motor cortex evoked potential increased by a factor of four or more. With the charge held constant to different-sized electrodes placed bilaterally in the same monkey, damage was greater under the smaller electtrode. This finding suggests that the charge density to cerebellar cortex must be controlled to avoid neural damage.