Safety and risk factors of TINAVI robot-assisted percutaneous pedicle screw placement in spinal surgery

Objective To determine the rates and risk factors of pedicle screw placement accuracy and the proximal facet joint violation (FJV) using TINAVI robot-assisted technique. Methods Patients with thoracolumbar fractures or degenerative diseases were retrospectively recruited from June 2018 and June 2020. The pedicle penetration and proximal FJV were compared in different instrumental levels to identify the safe and risk segments during insertion. Moreover, the factors were also assessed using univariate and multivariate analyses. Results A total of 72 patients with 332 pedicle screws were included in the current study. The optimal and clinically acceptable screw positions were 85.8% and 93.4%. Of the 332 screws concerning the intra-pedicular accuracy, 285 screws (85.8%) were evaluated as Grade A according to the Gertzbein and Robbins scale, with the remaining 25 (7.6%), 10 (3.0%), 6 (1.8%), and 6 screws (1.8%) as Grades B, C, D, and E. Moreover, in terms of the proximal FJV, 255 screws (76.8%) screws were assessed as Grade 0 according to the Babu scale, with the remaining 34 (10.3%), 22 (6.6%), and 21 screws (6.3%) as Grades 1, 2, and 3. Furthermore, the univariate analysis showed significantly higher rate of penetration for patients with age < 61 years old, sex of female, thoracolumbar insertion, shorter distance from skin to insertion point, and smaller facet angle. Meanwhile, the patients with the sex of female, BMI < 25.9, grade I spondylolisthesis, lumbosacral insertion, longer distance from skin to insertion point, and larger facet angle had a significantly higher rate of proximal FJV. The outcomes of multivariate analyses showed that sex of male (adjusted OR 0.320, 95% CI 0.140-0.732; p = 0.007), facet angle [greater than or equal to] 45[degrees] (adjusted OR 0.266, 95% CI 0.090-0.786; p = 0.017), distance from skin to insertion point [greater than or equal to] 4.5 cm (adjusted OR 0.342, 95% CI 0.134-0.868; p = 0.024), and lumbosacral instrumentation (adjusted OR 0.227, 95% CI 0.091-0.566; p = 0.001) were independently associated with intra-pedicular accuracy; the L5 insertion (adjusted OR 2.020, 95% CI 1.084-3.766; p = 0.027) and facet angle [greater than or equal to] 45[degrees] (adjusted OR 1.839, 95% CI 1.026-3.298; p = 0.041) were independently associated with the proximal FJV. Conclusion TINAVI robot-assisted technique was associated with a high rate of pedicle screw placement and a low rate of proximal FJV. This new technique showed a safe and pr