Robotic-assisted pedicle screw placement achieves high accuracy and narrows the experience gap: a preclinical evaluation
摘要
To assess a robotic-assisted system versus open manual techniques to: (1) compare pedicle screw accuracy between fellows and attendings with and without robotic assistance, and (2) evaluate executed versus planned trajectory fidelity relative to prior reports. Robotic-assisted spine systems aim to improve the accuracy, reproducibility, and safety of pedicle screw placement. Limited evidence compares outcomes across differing surgeon experience levels. A total of 255 thoracolumbar pedicle screws were placed in synthetic torsos simulating adult degenerative anatomy. Four surgeons (two fellows, two attendings) each placed half their screws using open fluoroscopy-guided techniques and half using a minimally invasive percutaneous approach with the Mako Spine system and intraoperative CT-based planning. Postoperative CT and the Gertzbein-Robbins classification were used to assess accuracy. Optimal placement (Grade A), breach rates, and positional/angular deviation from planned trajectories were recorded. The robotic cohort achieved a higher rate of clinically acceptable screw placement compared with the manual cohort, although this difference was not statistically significant (124/127, 97.6% vs. 120/128, 93.8%; P = 0.223). Fellows and attendings both achieved high clinically acceptable screw placement rates using robotics (63/64, 98.4% vs. 61/63, 96.8%). Among fellows, robotic assistance increased optimal screw placement compared with the manual technique (53/64, 82.8% vs. 41/64, 64.1%; P < 0.05). Robotic screws demonstrated low mean deviations from planned trajectories (1.4 ± 0.9 mm positional, 1.7 ± 1.1° angular). The Mako Spine system enabled precise screw placement and narrowed performance gaps for less experienced surgeons. It shows promise as both training tool and means to standardize outcomes.