OBJECTIVE
Pedicle and lateral mass screw placement is technically demanding due to complex 3D spinal anatomy that is not easily visualized. Neurosurgical and orthopedic surgery residents must be properly trained in such procedures, which can be associated with significant complications and associated morbidity. Current training in pedicle and lateral mass screw placement involves didactic teaching and supervised placement in the operating room. The objective of this study was to assess whether teaching residents to place pedicle and lateral mass screws using navigation software, combined with practice using cadaveric specimens and Sawbones models, would improve screw placement accuracy.
METHODS
This was a single-blinded, prospective, randomized pilot study with 8 junior neurosurgical residents and 2 senior medical students with prior neurosurgery exposure. Both the study group and the level of training-matched control group (each group with 4 level of training-matched residents and 1 senior medical student) were exposed to a standardized didactic education regarding spinal anatomy and screw placement techniques. The study group was exposed to an additional pilot program that included a training session using navigation software combined with cadaveric specimens and accessibility to Sawbones models.
RESULTS
A statistically significant reduction in overall surgical error was observed in the study group compared with the control group (p = 0.04). Analysis by spinal region demonstrated a significant reduction in surgical error in the thoracic and lumbar regions in the study group compared with controls (p = 0.02 and p = 0.04, respectively). The study group also was observed to place screws more optimally in the cervical, thoracic, and lumbar regions (p = 0.02, p = 0.04, and p = 0.04, respectively).
CONCLUSIONS
Surgical resident education in pedicle and lateral mass screw placement is a priority for training programs. This study demonstrated that compared with a didactic-only training model, using navigation simulation with cadavers and Sawbones models significantly reduced the number of screw placement errors in a laboratory setting.