Objective The aim of this study was to validate an imaging technique for evaluation of spinal surgery accuracy and to establish accuracy and safety of freehand technique in the thoracolumbar spine of large breed dogs. Study Design After thoracolumbar spine computed tomography (CT), 26 drilling corridors were planned then drilled to receive 3.2 mm positive profile pins using a freehand technique. After pin removal, CT was repeated. All entry points, exit points and angles of the preoperative planned trajectories were compared with postoperative ones using an image registration and fusion technique by three observers. Corridor coordinates for entry and exit points were evaluated in three dimensions and angles were measured in one plane. Intraclass correlation coefficient (ICC) was used to establish the imaging technique reliability and descriptive statistics were used to report on the freehand technique accuracy. Safety was evaluated using a vertebral cortical breach grading scheme. Results Intraclass correlation coefficient for the entry points, exit points and angle were 0.79, 0.96 and 0.92 respectively. Mean deviations for the entry points, exit points and angle were 3.1 mm, 6.3 mm and 7.6 degrees respectively. Maximum deviations were 6.3 mm, 11.0 mm and 16.4 degrees. Most deviations were lateral and caudal. All corridors were judged as safe. Conclusion The imaging technique reliability was good to excellent to study spinal surgery accuracy. Implant deviations should be anticipated when planning stabilization surgery in large breed dogs using the freehand-guided technique.

Objectives The main aim of this study was to evaluate the feasibility of minimally invasive stabilization with polyaxial screws-rod using neuronavigation and to assess accuracy and safety of percutaneous drilling of screw corridors using neuronavigation in thoracolumbar spine and compare it between an experienced and a novice surgeon. Study Design Feasibility of minimally invasive polyaxial screws-rod fixation using neuronavigation was first performed in the thoracolumbar spine of two dogs. Accuracy and safety of drilling screw corridors percutaneously by two surgeons from T8 to L7 in a large breed dog using neuronavigation were established by comparing entry and exit points coordinates deviations on multiplanar reconstructions between preoperative and postoperative datasets and using a vertebral cortical breach grading scheme. Results Feasibility of minimally invasive stabilization was demonstrated. For the experienced surgeon, safety was 100% and mean (standard deviation) entry point deviations were 0.3 mm (0.8 mm) lateral, 1.3 mm (0.8 mm) ventral and 0.7 mm (1.8 mm) caudal. The exit points deviations were 0.8 mm (1.9 mm) lateral, 0.02 mm (0.9 mm) dorsal and 0.7 mm (2.0 mm) caudal. Significant difference in accuracy between surgeons was found in the thoracic region but not in the lumbar region. Accuracy and safety improvement are noted for the thoracic region when procedures were repeated by the novice. Conclusion This proof of concept demonstrates that using neuronavigation, minimally invasive stabilization with polyaxial screws-rod is feasible and safe in a large breed dog model.