A new 3-dimensional method for measuring precision in surgical navigation and methods to optimize navigation accuracy

Description of a novel method for evaluation of pedicle screws in 3 dimensions utilizing O-arm^® and StealthStation^® navigation; identifying sources of error, and pearls for more precise screw placement.

O-arm and StealthStation navigation were utilized to place pedicle screws. Initial and final O-arm scans were performed, and the projected pedicle probe track, projected pedicle screw track, and final screw position were saved for evaluation. They were compared to evaluate the precision of the system as well as overall accuracy of final screw placement.

Thoracolumbar deformity patients were analyzed, with 153 of 158 screws in adequate position. Only 5 screws were malpositioned, requiring replacement or removal. All 5 were breached laterally and no neurologic or other complications were noted in any of these patients. This resulted in 97 % accuracy using the navigation system, and no neurological injuries or deficits. The average distance of the screw tip and angle of separation for the predicted path versus the final pedicle screw position were analyzed for precision. The mean screw tip distance from the projected tip was 6.43 mm, with a standard deviation of 3.49 mm when utilizing a navigated probe alone and 5.92 mm with a standard deviation of 3.50 mm using a navigated probe and navigated screwdriver (p = 0.23). Mean angle differences were 4.02° and 3.09° respectively (p < 0.01), with standard deviations of 2.63° and 2.12°.

This new technique evaluating precision of screw placement in 3 dimensions improves the ability to define screw placement. Pedicle screw position at final imaging showed the use of StealthStation navigation to be accurate and safe. As this is a preliminary evaluation, we have identified several factors affecting the precision of pedicle screw final position relative to that predicted with navigation.