Comparison of Superior‐Level Facet Joint Violations Between Robot‐Assisted Percutaneous Pedicle Screw Placement and Conventional Open Fluoroscopic‐Guided Pedicle Screw Placement

OBJECTIVE: To compare the superior‐level facet joint violations (FJV) between robot‐assisted (RA) percutaneous pedicle screw placement and conventional open fluoroscopic‐guided (FG) pedicle screw placement in a prospective cohort study. METHODS: This was a prospective cohort study without randomization. One‐hundred patients scheduled to undergo RA (n = 50) or FG (n = 50) transforaminal lumbar interbody fusion were included from February 2016 to May 2018. The grade of FJV, the distance between pedicle screws and the corresponding proximal facet joint, and intra‐pedicle accuracy of the top screw were evaluated based on postoperative CT scan. Patient demographics, perioperative outcomes, and radiation exposure were recorded and compared. Perioperative outcomes include surgical time, intraoperative blood loss, postoperative length of stay, conversion, and revision surgeries. RESULTS: Of the 100 screws in the RA group, 4 violated the proximal facet joint, while 26 of 100 in the FG group had FJV (P = 0.000). In the RA group, 3 and 1 screws were classified as grade 1 and 2, respectively. Of the 26 FJV screws in the FG group, 17 screws were scored as grade 1, 6 screws were grade 2, and 3 screws were grade 3. Significantly more severe FJV were noted in the FG group than in the RA group (P = 0.000). There was a statistically significant difference between RA and FG for overall violation grade (0.05 vs 0.38, P = 0.000). The average distance of pedicle screws from facet joints in the RA group (4.16 ± 2.60 mm) was larger than that in the FG group (1.92 ± 1.55 mm; P = 0.000). For intra‐pedicle accuracy, the rate of perfect screw position was greater in the RA group than in the FG group (85% vs 71%; P = 0.017). No statistically significant difference was found between the clinically acceptable screws between groups (P = 0.279). The radiation dose was higher in the FG group (30.3 ± 11.3 vs 65.3 ± 28.3 μSv; P = 0.000). The operative time in the RA group was significantly longer (184.7 ± 54.3 vs 117.8 ± 36.9 min; P = 0.000). CONCLUSIONS: Compared to the open FG technique, minimally invasive RA spine surgery was associated with fewer proximal facet joint violations, larger facet to screw distance, and higher intra‐pedicle accuracy.