Validation of Triggered EMGs by CT Scan for Thoracic Pedicle Screw Placement

Timothy R. Kuklo, MD, JD
Associate Professor
Orthopaedic Surgery and Neurological Surgery
Washington University School of Medicine
St. Louis, MO
David W. Polly, Jr., MD
Professor and Chief of Spine Service
University of Minnesota, Department of Orthopaedic Surgery
Minneapolis, MN
Ronald A. Lehman, Jr.
Washington University Medical School
St. Louis, MO
Betzaida Vazquez
Walter Reed Army Medical Center
Washington, DC
Exhibit from the SRS 2002 Annual Meeting
INTRODUCTION: Pedicle stimulation is an easily reproducible monitoring technique for lumbar screws. For thoracic pedicle screws, rectus abdominus triggered EMG (trEMG) thresholds are frequently used for T6-T12. However, no reliable method of obtaining triggered potentials for T1-T5 currently exists to identify medially placed screws.

PURPOSE: The purpose of this study was (1) to validate the trEMG thresholds in the thoracic spine (T6-T12), as recorded from the rectus abdominus, by postoperative CT scan and (2) to determine if a threshold effect exists for T1-T5 with this method.

METHODS: Triggered EMGs were recorded using an ascending method of stimulation for each thoracic pedicle screw in 18 consecutive spinal deformity patients (262 titanium screws) undergoing posterior spinal fusion. A postoperative fine-cut CT scan was obtained and evaluated by an independent observer to assess for possible medial pedicle violation(s). TrEMG recordings and CT findings were then analyzed.

RESULTS: The mean trEMG threshold was 13.3 mA for all thoracic screws. For T6-T12, the mean trEMG threshold was 13.6 mA (192 screws), and 12.4 mA for T1-T5 (70 screws) (p>0.05). There were 12 screws (4.5%) that were determined to have a medial violation by CT scan (T5-3, T6-2, T7-1, T8-2, T11-2, T12 –2). Of these, the average trEMG for was 11.1 mA for T6-T12 screws (10 screws)(82% of mean threshold) (p<.05), which was also <80% of the mean screw threshold for those individuals. The lowest measured trEMG was 6 mA (2 screws), both with a medial violation. However, two additional medial screws (on the curve convexity) could not be distinguished from the mean trEMG for that individual. For T1-T5 (2 medial screws on the curve convexity), the average trEMG was 12.5 mA (101% of mean threshold)(p>.05), which was > 113% of the mean screw threshold for each individual patient. These two screws (both at T5) were removed. There were no neurologic complications.

CONCLUSIONS: No absolute triggered EMG threshold could be established in this study, except for screws placed from T6-T12 with < 6 mA threshold (2 screws). Further, T1-T5 triggered EMGs are unreliable with this technique. For spinal deformity, convex screws appear to have higher threshold potentials, and therefore, this technique may be less reliable. Our findings could reflect a difference between titanium and stainless steel screws.
Last Updated: 04/26/2005