Utility of evoked EMG monitoring to improve bone screw placements in the cervical spine

EMG threshold testing as a guide to accurate screw placement was examined during posterior cervical spine instrumentation. The accuracy of screw placements was compared with the surgeon blinded and unblinded to EMG thresholds. To determine the utility of EMG threshold testing in improving screw placements in the lateral mass and pedicles of the cervical/upper thoracic spine. EMG threshold testing in the lumbar spine is generally thought to improve the accuracy of pedicle screw placements. These results may not generalize to the cervical spine, where smaller pedicles, bicortical pilot holes, and the orientation of lateral mass screws away from midline may result in different alert thresholds. Triggered EMG thresholds were obtained from pilot holes in 244 lateral mass and 113 pedicles from 32 patients. Thresholds were compared with the accuracy of screw placements as determined from postoperative computed tomography scans. The percentage of inaccurate and potentially dangerous (IAPD) screws with the surgeon blinded and unblinded to EMG thresholds was determined. EMG threshold testing was more accurate in predicting IAPD screw trajectories in the pedicles (likelihood ratio 5.1) as compared with the lateral mass (likelihood ratio 2.9). In the pedicles, the number of IAPD screws decreased from 4.5% in the blinded controls to 0% in the unblinded group. In the lateral mass, there were no IAPD screw placements in the blinded control group, whereas 2% of the screws in the unblinded group were IAPD. EMG threshold testing in the cervical spine (C3, T1) is a moderately accurate diagnostic test and more predictive of potentially dangerous screw trajectories in the pedicles (C7, T1) as compared with the lateral mass (C3-C6). EMG threshold testing may decrease potentially dangerous screw placements in the pedicles, but may have less utility in the case of the lateral mass because of less reproducible placement of the stimulating probe.