Accuracy of 837 pedicle screw positions in degenerative lumbar spine with conventional open surgery evaluated by computed tomography.

BACKGROUND: The spatial and directional accuracy of the positioning of pedicle screws in the lumbosacral spine with conventional open surgery assessed by computed tomography (CT) has been published in several studies, systematic reviews and meta-analyses with a short-term follow-up. Inaccurate pedicle screw insertion may cause neurologic symptoms and weakens the construct.

METHODS: The data of 147 patients operated on with transpedicular screw fixation based on anatomical landmarks, supported by fluoroscopy, by a senior neurosurgeon in our clinic between 2000 and 2010 were analyzed retrospectively. The accuracy of the pedicle screw position was assessed by using postoperative CT images and graded in 2-mm increments up to 6 mm by two independent surgeons and partly by an independent radiologist.

RESULTS: A total of 837 lumbosacral pedicle screws were inserted in 147 randomly selected patients by a senior neurosurgeon. A mean accuracy of 85.7% of the screws being inside the pedicles was identified by the surgeon observers, with 3.3% being perforated 4 mm or more outside the pedicles. Postoperative neurologic symptoms were observed on the side corresponding to the breach in an average of 25.9% of patients with pedicle perforations, and 89.2% of the misplaced screws were either medially or inferiorly inserted.

CONCLUSIONS: Screw application reached a mean accuracy of 85.7% based on anatomical landmarks supported by fluoroscopy, warranting computer-assisted navigation for increased accuracy. Our results of 24 patients (16.3%) with the breached screws indicate that the direction of the breach may be more important than the absolute deviation in causing new neurologic symptoms.