2:26 Safety of thoracic pedicle screw insertion with and without computer assistance: a cadaveric study

Abstract

Purpose of study: The placement of pedicle screws is technically demanding and poses potential hazards related mainly to penetration of the anterolateral vertebral body or medial pedicle. To investigate safety, we undertook an in vitro analysis of thoracic pedicle screw placement by two senior spine surgeons using standard fluoroscopy, virtual fluoroscopy and computed tomography image guidance.Methods used: In a simulated OR environment, cadaveric thoracic spines with the posterolateral rib cage intact were mounted in foam. Pedicle screws were placed from T1 up to T11 in 20 cadaveric thoracic spines using standard anatomic landmarks and fluoroscopic guidance (Group 1), FluoroNav with one reference (Group 2) or multiple reference points (Group 3), and Stealth (Group 4). Radiation exposure to the surgeon's torso, hands and specimen were measured. The specimens were then dissected, sagittally split and examined for evidence of pedicle or vertebral body cortical screw perforations. Any cortical violation except the costovertebral junction was considered out. The cortical perforations were then graded by 2-mm increments. Statistical analysis was performed.of findings: There were 94, 99, 70 and 74 screws placed in the four groups. There was significantly more average radiation exposure to the specimen in Group 3 (27 mREM) and less in Group 4 (0 mREM) than in Group 1 (7.8 mREM). Radiation exposure to the surgeon was minimal, measuring less than 1 mREM to the surgeon's torso per specimen in all groups except 4 (0 mREM). Average time per screw was significantly longer in Group 4 (6.8 minutes) compared with Group 1 (2.4 minutes). Cortical violations were significantly less in Group 3 (11.4%) and Group 4 (5.4%) compared with Group 1 (20.2%). The only medial cortical violations were in Group 1, and none were greater than 2 mm.Relationship between findings and existing knowledge: Anatomic landmarks and standard fluoroscopy have been shown to have approximately 20% thoracic screw malpositioning. The current study confirms this and demonstrates greater accuracy with image guidance.Overall significance of findings: Although the time for screw placement was longer with Stealth and FluoroNav, the percentage of screws with cortical violations was significantly less. Computer-assisted thoracic pedicle screw placement may provide more accurate placement and less risk to critical structures.Disclosures: Device or drug: pedicle screws. Status: approved. Device or drug: FluoroNav. Status: approved. Device or drug: Stealth. Status: approved.Conflict of interest: Christopher Shaffrey, grant research support; Jens Chapman, grant research support Syntles USAAO/ASIF International Sofamor Danek; Jens Chapman, CHAIR, Surgical Dynamics.