Session VIII - Pelvis

Evaluation of Intraoperative Nerve Monitoring during the Insertion of Iliosacral Implants in an Animal Model

Berton R. Moed, MD; Michael J. Hartman, DVM, MD; B.K. Ahmad, MD; Dianna D. Cody, PhD; Joseph G. Craig, MD, Department of Orthopaedic Surgery, Wayne State University, Detroit, MI

Introduction: The purpose of this study was to evaluate somatosensory evoked potential (SSEP), spontaneous electromyographic (EMG) and Stimulus-Evoked (S-E) EMG monitoring of neural function during the placement of iliosacral implants.

Methods: While monitoring the first sacral (S1) nerve root function using SSEP, spontaneous EMG, S-E EMG monitoring techniques, a 2.0 mm stainless steel K-wire was progressively inserted into the S1 body of each of 17 dog hemipelves using a General Electric high speed computerized tomographic (CT) scanner capable of 1-mm slice thickness with a resolution of 0.45 mm. The K-wire was directed in an attempt to compromise the S1 canal and nerve root. A minimum of four data points per hemipelvis was planned. The endpoint was contact with the nerve. It was expected that this endpoint would be heralded by a burst of spontaneous EMG activity and an abnormal SSEP signal. Current thresholds required to evoke an EMG response (S-E EMG) were continuously recorded. A current threshold of < 4 milliamperes (mA) was expected as indicative of nerve contact and < 6 mA indicating broaching of the neural canal. A GE Advantage Windows workstation running version 1.2 software was used with a configuration calculated to provide distance measurements to the nearest tenth of a millimeter. Anatomical dissection at completion of the study documented final K-wire position. A power study had been prospectively performed to determine the number of hemipelves required to provide 90% power, assuming that measurements would be obtained from both sides of the same animal with a minimum of four data measurements per side.

Results: A total of 113 data points were obtained (>6 per hemipelvis). The correlation coefficient for the current threshold/distance S-E EMG relationship was found to be 0.801 with a root mean square error of 2.47, indicating a highly significant fit (p < 0.001). Anatomical dissection at completion of the study revealed that 16 of the 17 K-wires had actually contacted the nerve root. Four had penetrated the nerve root and 12 were compressing the S1 root but had not penetrated its substance. One wire was 1 mm from the root but had not penetrated the S1 canal. This was the only final S-E EMG current threshold measured at > 6 mA (6.3 mA). Fourteen of the remaining 16 current thresholds were < 4 mA; an observed proportion not different >from the expected 16/16 (Fisher exact test, p =0.48). A spontaneous burst of EMG activity was not obtained in a single case (significantly different from the expected, Fisher exact test, p< 0.001). SSEP's could be obtained in only 12 hemipelves due to technical problems, and abnormal SSEP's were obtained in only one of the 12 (significantly different from the expected, Fisher exact test, p< 0.001).

Discussion and Conclusion: This study raises serious question regarding the validity of SSEP and spontaneous EMG monitoring for the purpose of minimizing nerve root injury during the insertion of iliosacral implants. In addition, it further confirms the validity of the S-E EMG current threshold/distance relationship and its potential applicability for nerve monitoring.