Accuracy of pedicle screw insertion for unilateral open transforaminal lumbar interbody fusion: a side-by-side comparison of percutaneous and conventional open techniques in the same patients.

BACKGROUND

The aim of the study was to compare the accuracy of percutaneous pedicle screw (PPS) insertion (P-side) with that of conventional open screw insertion (O-side) during unilateral open transforaminal lumbar interbody fusion (TLIF) in the same patients. We also sought to determine the incidence of pedicle screw misplacement and to identify relevant risk factors.

METHODS

The study was a retrospective analysis of prospectively collected data for 766 pedicle screws placed in 181 consecutive patients who underwent a unilateral open-TLIF procedure in the lumbosacral spine. Our minimally invasive TLIF was performed by unilateral open freehand insertion of pedicle screws for decompression on one side and PPS on the opposite side. Using this approach, we were able to compare the accuracy of PPS insertion with that of conventional open screw insertion in the same patients. There were 383 PPSs and 383 screws inserted by the open method. The accuracy of screw placement was evaluated on reconstructed computed tomography images obtained postoperatively, and screw misplacement was classified. Potential risk factors for screw misplacement were investigated in three-level mixed-effects logistic regression analysis.

RESULTS

Thirty-four screws (8.9%) were misplaced on the P-side and 37 (9.5%) were misplaced on the O-side; the difference was not statistically significant (P = 0.803). Subclassification analysis revealed minor perforation of 28 screws (7.3%) on the P-side and 32 (8.4%) on the O-side, moderate perforation of 5 screws (1.3%) on the P-side and 4 (1.0%) on the O-side, and severe perforation of 1 screw (0.3%) on each side. Three-level mixed-effects logistic regression analysis identified body mass index as a significant risk factor for screw misplacement on the P-side (odds ratio 1.194, 95% confidence interval 1.066-1.338).

CONCLUSIONS

Accuracy of pedicle screw insertion was not significantly different between PPS insertion and conventional open screw insertion in the same patients. Body mass index had a significant influence on the risk of screw misplacement in PPS insertion.