Efficacy of multimodality spinal cord monitoring during surgery for neuromuscular scoliosis.

STUDY

DESIGN

This study determined the relative efficacy of somatosensory-evoked potentials and motor-evoked potentials in monitoring spinal cord function during surgery for patients with idiopathic versus neuromuscular scoliosis.

OBJECTIVES

To determine whether patients with idiopathic versus neuromuscular scoliosis demonstrate significantly different somatosensory-evoked potentials and motor-evoked potentials recorded during surgery.

SUMMARY OF BACKGROUND DATA: Ashkenaze et al (1993) and others have reported that cortical somatosensory-evoked potentials are unreliable when used to monitor spinal cord function in patients with neuromuscular scoliosis. It was recommended that other neurophysiologic tests be used.

METHODS

Somatosensory-evoked potentials and motor-evoked potentials were recorded from two groups of patients: those with idiopathic scoliosis and those with neuromuscular scoliosis. Somatosensory-evoked potentials were obtained before and during surgery. Motor-evoked potentials were obtained during surgery. Normal variability, as indicated from idiopathic scoliotic results, was compared with data obtained from patients with neuromuscular scoliosis. Motor-evoked potentials and somatosensory-evoked potentials were obtained sequentially during the duration of surgery.

RESULTS

Single-channel cortical somatosensory-evoked potentials demonstrated a 27% positive rate, which was consistent with results (28%) from Ashkenaze et al. The use of multiple recording sites for the somatosensory-evoked potentials and the addition of motor-evoked potential procedures indicated that a reliable response could be obtained in more than 96% of the patients. It also was found that cortical somatosensory-evoked potentials were more affected by anesthetic agents when recorded from patients with neuromuscular scoliosis compared with patients with idiopathic scoliosis.

CONCLUSIONS

Single-channel cortical somatosensory-evoked potentials demonstrated a high level of unreliability, which reduced their clinical effectiveness. However, by using multiple recording sites with the somatosensory-evoked potentials and by administering motor-evoked potential procedures, it was possible to monitor spinal cord function in neuromuscular patients and avoid postoperative neurologic deficits.