The role of transcranial motor evoked potentials in predicting neurologic and histopathologic outcome after experimental spinal cord ischemia

Background: Monitoring of myogenic motor evoked potentials to transcranial stimulation (tcMEPs) is clinically used to assess motor pathway function during aortic and spinal procedures that carry a risk of spinal cord ischemia (SCI). Although tcMEPS presumably detect SCI before irreversible neuronal deficit occurs, and prolonged reduction of tcMEP signals is thought to be associated with impending spinal cord damage, experimental evidence to support this concept has not been provided. In this study, histopathologic and neurologic outcome was examined in a porcine model of SCI after different durations of intraoperative loss of tcMEP signals. Methods: In 15 ketamine-sufentanil-anesthetized pigs (weight, 35-45 kg) the spinal cord feeding lumbar arteries were exposed. tcMEP were recorded from the upper and lower limbs. Under normothermic conditions, animals were randomly allocated to undergo short-term tcMEP reduction (group A, < 10 min, n = 5) or prolonged tcMEP reduction (group B, 60 min, n = 10), resulting from temporary or permanent clamping of lumbar segmental arteries. Neurologic function was evaluated every 24 h, and infarction volume and the number of eosinophilic neurons and viable motoneurons in the lumbosacral spinal cord was evaluated 72 h after induction of SCI. Results: In all animals except one, segmental artery clamping reduced tcMEP to below 25% of baseline. All but one animal in group A had reduced tcMEP for less than 10 min and had normal motor function and no infarction at 72 h after the initial tcMEP reduction. Seven animals in group B (70%) had reduced tcMEP signals for more than 60 min and were paraplegic with massive spinal cord infarction at 72 h. Two animals (one in both groups) had tcMEP loss for 40 min, with moderate infarction and normal function. In general, histopathologic damage and neurologic dysfunction did not occur when tcMEP amplitude recovered within 10 and 40 min after the initial decline, respectively. Conclusion: Prolonged reduction of intraoperative tcMEP amplitude is predictive for postoperative neurologic dysfunction, while recovery of the tcMEP signal within 10 min after the initial decline corresponds with normal histopathology and motor function in this experimental model. This finding confirms that intraoperative tcMEPs have a good prognostic value for neurologic outcome during procedures in which the spinal cord is at risk for ischemia.