Rates and risk factors of intrapedicular accuracy and cranial facet joint violation among robot-assisted, fluoroscopy-guided percutaneous, and freehand techniques in pedicle screw fixation of thoracolumbar fractures: a comparative cohort study

Robot-assisted (RA) technique has been increasingly applied in clinical practice, providing promising outcomes of inserting accuracy and cranial facet joint protection. However, studies comparing this novel method with other assisted methods are rare, and the controversy of the superiority between the insertion techniques remains. Thus, we compare the rates and risk factors of intrapedicular accuracy and cranial facet joint violation (FJV) of RA, fluoroscopy-guided percutaneous (FP), and freehand (FH) techniques in the treatment of thoracolumbar fractures.

A total of 74 patients with thoracolumbar fractures requiring pedicle screw instruments were retrospectively included and divided into RA, FP, and FH groups from June 2016 to May 2020. The primary outcomes were the intrapedicular accuracy and cranial FJV. The factors that affected the intrapedicular accuracy and cranial FJV were assessed using multivariate analyses.

The optimal intrapedicular accuracy of pedicle screw placement (Grade A) in the RA, FP, and FH groups was 94.3%, 78.2%, and 88.7%, respectively. This finding indicates no significant differences of RA over FH technique (P = 0.062) and FP technique (P = 0.025), but significantly higher accuracies of RA over FP (P < 0.001). In addition, the rates of proximal FJV in RA, FP, and FH groups were 13.9%, 30.8%, and 22.7%, respectively. RA had a significantly greater proportion of intact facet joints than the FP (P = 0.002). However, FP and FH (P = 0.157), as well as RA and FH (P = 0.035) showed significantly similar outcomes with respect to the proximal FJV. The logistic regression analysis showed that FP technique (OR = 3.056) was independently associated with insertion accuracy. Meanwhile, the age (OR = 0.974), pedicle angle (OR = 0.921), moderate facet joint osteoarthritis (OR = 5.584), and severe facet joint osteoarthritis (OR = 11.956) were independently associated with cranial FJV.

RA technique showed a higher rate of intrapedicular accuracy and a lower rate of cranial FJV than FP technique, and similar outcomes to FH technique in terms of intrapedicular accuracy and cranial FJV. RA technique might be a safe method for pedicle screw placement in thoracolumbar surgery.

3