We evaluated the use of a part-task simulator with 3D and haptic feedback as a training tool for a common neurosurgical procedure – placement of thoracic pedicle screws.

To evaluate the learning retention of thoracic pedicle screw placement on a high-performance augmented reality and haptic technology workstation.

Fifty-one fellows and residents performed thoracic pedicle screw placement on the simulator. The virtual screws were drilled into a virtual patient’s thoracic spine derived from a computed tomography data set of a real patient.

With a 12.5% failure rate, a two-proportion z-test yielded P= 0.08. For performance accuracy, an aggregate Euclidean distance deviation from entry landmark on the pedicle and a similar deviation from the target landmark in the vertebral body yielded P=0.04 from a two-sample t-test in which the rejected null hypothesis assumes no improvement in performance accuracy from the practice to the test sessions, and the alternative hypothesis assumes an improvement.

The performance accuracy on the simulator was comparable to the accuracy reported in literature on recent retrospective evaluation of such placements. The failure rates indicated a minor drop from practice to test sessions, and also indicated a trend (P=0.08) towards learning retention resulting in improvement from practice to test sessions. The performance accuracy showed a 15% mean score improvement and over 50% reduction in standard deviation from practice to test. It showed evidence (P=0.04) of performance accuracy improvement from practice to test session.