Can Triggered EMG Thresholds Accurately Predict Thoracic Pedicle Screw Placement?

PURPOSE:
Triggered EMG stimulation with lower extremity myogenic recordings has proven helpful in identifying medially placed lumbar pedicle screws. Use of intercostal muscle recordings for thoracic pedicle screw placement has proven non-specific in an animal model. This study focuses on the use of the rectus abdominis muscles as a recording site for screws placed in spinal levels T6-12 to determine the efficacy of using triggered electromyographic threshold stimulation (TrgEMG) to assess positioning of thoracic pedicle screws.

METHODS:
501 thoracic pedicle screws were placed in 70 consecutive patients (July 1999 through February 2001). All screws placed between T6 & T12 were evaluated using an ascending method of stimulation to record a compound muscle action potential from rectus abdominis muscles bilaterally. The lowest intensity TrgEMG threshold in each patient resulted in screw removal & a recheck of all pedicle borders by careful palpation at that spinal level.

RESULTS:
There were 5 thoracic screws (Group A) with medial wall breakthrough confirmed by tactile, visual and/or radiographic inspection. TrgEMG thresholds for these screws were all ‹ 6.0 mA. 10 appropriately placed screws also had thresholds ‹ 6.0 mA (Group B) and were rechecked with intraoperative confirmation of intact medial walls. 486 screws had TrgEMG thresholds › 6.0mA (Group C). Due to the lack of absolute threshold value consensus, each threshold ‹ 6.0mA (Groups A & B) was compared to the mean of all other TrgEMG responses obtained within each individual patient. The percentage of decrease from this mean was calculated for each low threshold. Group A screws (n=5) had a mean percentage decrease of 66.7%(range 46.1-76.8). Group B screws had a mean percentage decrease of 50.5 % (range 34.7-62.3). A two-sample t-test demonstrated a statistically significant difference between the two mean values (p = 0.0130). Due to the small sample size in both groups, this should be viewed as a trend rather than an absolute indication.

CONCLUSIONS:
Similar to lumbar pedicle screw stimulation, thoracic pedicle screw (T6-T12) stimulation with rectus abdominis recordings can provide important information on the integrity of the medial pedicle wall. Suspicion for medially placed screws should arise when the TrgEMG is ‹ 6.0mA; however, that threshold is not absolute confirmation of medial wall breakthrough.