100 Complex posterior spinal fusion cases performed with robotic instrumentation.

Robotic navigation has been shown to increase precision, accuracy, and safety during spinal reconstructive procedures. There is a paucity of literature describing the best techniques for robotic-assisted spine surgery for complex, multilevel cases or in cases of significant deformity correction. We present a case series of 100 consecutive multilevel posterior spinal fusion procedures performed for multilevel spinal disease and/or deformity correction. 100 consecutive posterior spinal fusions were performed for multilevel disease and/or deformity correction utilizing robotic-assisted placement of pedicle screws. The primary outcome was surgery-related failure, which was defined as hardware breakage or reoperation with removal of hardware. A total of 100 consecutive patients met inclusion criteria. Among cases included, 31 were revision surgeries with existing hardware in place. The mean number of levels fused was 5.6, the mean operative time was 303 min, and the mean estimated blood loss was 469 mL. 28 cases included robotic-assisted placement of S2 alar-iliac (S2AI) screws. In total, 1043 pedicle screws and 53 S2AI screws were placed with robotic-assistance. The failure rate using survivorship analysis was 18/1043 (1.7%) and the failure rate of S2AI screws using survivorship analysis was 3/53 (5.7%). Four patients developed postoperative wound infections requiring irrigation and debridement procedures. None of the 1043 pedicle screws nor the 53 S2AI screws required reoperation due to malpositioning or suboptimal placement. This case series of 100 multilevel posterior spinal fusion procedures demonstrates promising results with low failure rates. With 1043 pedicle screws and 53 S2AI screws, we report low failure rates of 1.7% and 5.7%, respectively with zero cases of screw malpositioning. Robotic screw placement allows for accurate screw placement with no increased rate of postoperative infection compared to historical controls. Level of evidence: IV, Retrospective review.

Cadaver training module for teaching thoracic pedicle screw placement to residents.

Surgical training using simulators has been shown to be highly effective but is not available for some applications and is too expensive for many programs. The authors piloted a cadaver-based module with the goal of objectively measuring and significantly improving orthopedic residents' surgical skills in placing thoracic pedicle screws, an advanced procedure. An experienced spine surgeon placed thoracic pedicle screws in 7 cadavers (T1-T12) to establish the skilled accuracy rate. For this pilot study, 3 orthopedic residents unfamiliar with the procedure were given didactic training for safe thoracic pedicle screw insertion. Each resident instrumented alternating sides of 5 consecutive cadavers (T1-T12). Screw positions were graded by computed tomography in a blinded fashion, with accuracy defined as no shank breach of the pedicle or vertebral body. Results were reviewed with the residents, instruction was repeated, and alternating sides of 5 cadavers were instrumented by the residents. The experienced surgeon accurately placed 67 (82%) of 82 pedicle screws. Residents accurately placed 80 (44%) of 180 pedicle screws in the initial set of specimens and 105 (58%) of 180 pedicle screws in the second set of specimens (P=.01). Accuracy varied significantly among residents before but not after computed tomography review. The study's results show that a cadaver-based training module that resembles the clinical setting can be used to teach complex surgical skills to orthopedic residents.