Spinal-evoked potentials in dogs with acute compressive thoracolumbar spinal cord disease.

Spinal evoked potentials (SpEP) were recorded on an electromyograph from electrodes placed percutaneously in the ligamentum flava at the lumbosacral junction and between the 10th and 11th thoracic vertebrae following tibial nerve stimulation in 31 anesthetized dogs with acute compressive spinal cord injuries. The neurologic status of each dog was determined by clinical examination before SpEP recordings, and the neurologic status was monitored for 2 months in dogs that had surgical or conservative treatment. Two months after spinal injury, the response to treatment (outcome) of each dog was evaluated and graded as favorable (ambulatory and urinary continent) or unfavorable (nonambulatory, urinary incontinent, or euthanatized with confirmation of myelomalacia). Onset latencies, conduction velocities, amplitudes and durations of the wave forms, and the ratio of conduction velocity to combined durations of the first positive (P1) and first negative (N1) waves (CV/DPN index) were determined and were compared with reference data from clinically normal (control) dogs. Single SpEP recordings were of value in determining the prognosis for recovery. Significant differences were not found in the L7-S1 recordings between the reference (control) and spinal injury groups. Analysis of data from the T10-11 recordings indicated significant differences between the reference and spinal injury groups and between the favorable and unfavorable outcome groups within the spinal injury group. A CV/DPN index was less than 30 in dogs with unfavorable outcomes and greater than 30 in dogs with favorable outcomes. Stepwise discriminant analysis of data from the spinal injury group predicted outcome correctly in all dogs.