A Comparison of Percutaneous Pedicle Screw Accuracy Between Robotic Navigation and Novel Fluoroscopy-Based Instrument Tracking for Patients Undergoing Instrumented Thoracolumbar Surgery.

Background: The accuracy of pedicle screws placed with instrument tracking and robotic navigation are individually comparable or superior to placement using standard fluoroscopy, however head-to-head comparisons between these adjuncts in a similar surgical population have yet to be performed.

Methods: Consecutive patients undergoing percutaneous thoracic and lumbosacral spinal instrumentation were retrospectively enrolled. Instrumentation was performed using either fluoroscopy-based instrument tracking system (TrackX, TrackX Technologies) or robotic-navigation (ExcelsiusGPS, Globus Medical). Postinstrumentation computed tomography scans were graded for breach according to the Gertzbein-Robbins scale, with "acceptable" screws deemed as Grade A or B and "unacceptable" screws deemed as Grades C through E. Accuracy data was compared between both instrumentation modalities.

Results: Fifty-three patients, comprising a total of 250 screws (167 robot, 83 instrument tracking) were included. The overall accuracy between both modalities was similar, with 96.4% and 97.6% of screws with acceptable accuracy between instrument tracking and robotic navigation, respectively (I-squared 0.30, df = 1, P = 0.58). Between instrument tracking and robotic navigation, 92.8% and 95.8% of screws received Grade A, 3.6% and 1.8% a Grade B, 1.2% and 1.2% a Grade C, 1.2% and 0.6% a Grade D, and 1.2% and 0.6% a Grade E, respectively. The robot was abandoned intraoperatively in 2 cases due to unrecoverable registration inaccuracy or software failure, leading to abandonment of 8 potential screws (4.8%).

Conclusions: In a similar patient population, there is a similarly high degree of instrumentation accuracy between fluoroscopy-based instrument tracking and robotic navigation. There is a rare chance for screw breach with either surgical adjunct.