Bikash Bose

STUDY DESIGN: Retrospective review of 61 consecutive patients. OBJECTIVES: To determine the effectiveness of combining intraoperative monitoring of both spontaneous electromyographic activity and compound muscle action potential response to stimulation for detecting a perforation of the pedicle cortex irritation of nerve root during lumbar spine fusion surgery. SUMMARY OF BACKGROUND DATA: The complication rate from instrumentation used with lumbar spine fusion varies from 1 to 33%. To prevent neurologic complications, several monitoring techniques have been used to alert surgeons to possible neurologic damage being introduced during nerve decompression or placement of instrumentation with spine procedures. Because of different sensitivities, one monitoring technique may not be as effective for preventing complications as a combination of techniques. METHODS: Sixty-one consecutive patients who underwent instrumented posterior lumbar fusions received continuous electromyographic monitoring and stimulus-evoked electromyographic monitoring. A significant neurophysiologic event was signaled by sustained neurotonic electromyographic activity, prompting an alert and a pause in the surgical manipulations that precipitated the activity. After insertion of the transpedicular screws, the integrity of the pedicle cortex was tested by stimulating each screw head and recording compound muscle action potentials. In the presence of a pedicle breach, stimulus intensities below 7 mA were sufficient to evoke compound muscle action potentials from the muscle group innervated by the adjacent spinal nerve root, prompting a surgical alert and subsequent repositioning of the screw. RESULTS: Fourteen significant neurophysiologic events occurred in 13 of 61 patients (21%). Sustained neurotonic electromyographic discharges occurred in 5 of 40 patients during placement of interbody fusion cages, in 2 patients during placement of transpedicular screws, and in 1 patient during tightening of rods. On pedicle screw stimulation, breaches of the pedicle cortex were detected in 6 patients. After surgery, no new neurologic deficits were found in 60 of the 61 patients. One patient who experienced temporary paraparesis had sustained neurotonic electromyographic discharges during retraction of the thecal sac and distraction of the disc space before placement of the cage. CONCLUSION: These results suggest that intraoperative electromyographic monitoring provides a real-time measure of impending spinal nerve root injury during instrumented posterior lumbar fusion, allowing for timely intervention and minimization of negative postoperative sequela.

Abstract Early investigators indicated that conservative management was superior to operative intervention in the treatment of central cord injuries. Their clinical data suggested that operative treatment, in fact, worsened the condition. Recent experience with this clinical entity, however, indicates that in selected patients operative intervention may be of value in improving the rate and degree of motor recovery. A retrospective study of all individuals admitted to our hospital (Delaware Valley Spinal Cord Injury Center) with central cervical spinal cord injury was done (28 patients). One-half had been treated with medical therapy alone (Group I); the others were treated both medically and surgically (Group II). Medical therapy consisted of intravenous mannitol, dexamethasone, and sodium bicarbonate given during the acute phase of the injury. Both groups were immobilized using either a halo or a Philadelphia collar. Criteria for entry into the surgical group were one or both of the following: (a) failure to improve progressively after an initial period of improvement, with persistent compression of neural tissue visualized on myelography and (b) unacceptable instability of the spinal bony elements. The patients were given neurological scores based on the motor power of the major muscle groups. The stability of the spine was scored using the Panjabi-White scale. The two groups were compared using Student's t-test and the two-factor analysis of variance. There was no significant difference in initial neurological scores between the groups. The surgical groups had a higher incidence of instability of the bony elements of the cervical spine, as judged by the Panjabi-White scale. At the time of discharge, Group II had significantly improved motor scores when compared with Group I. The results indicate that operative intervention did not produce neurological worsening and may be safely applied when patients with central cord injuries meet the criteria used in the study. Although the two groups are not comparable, operative intervention provided statistically better motor recovery than did conservative therapy (P < 0.05).