The cervical pedicle screw (PS) provides strong stabilization but poses a potential risk to the neurovascular system, which may be catastrophic. In particular, vertebrae with degenerative changes complicate the process of screw insertion, and PS misplacement and subsequent complications are more frequent. The purpose of this study was to evaluate the peri- and postoperative complications of PS fixation for nontraumatic lesions and to determine the risk factors of each complication.
Eighty-four patients who underwent cervical PS fixation for nontraumatic lesions were independently reviewed to identify associated complications. The mean age of the patients was 60.1 years, and the mean follow-up period was 4.1 years (range 6–168 months). Pedicle screw malpositioning was classified on postoperative CT scans as Grade I (< 50% of the screw outside the pedicle) or Grade II (≥ 50% of the screw outside the pedicle). Risk factors of each complication were evaluated using a multivariate analysis.
Three hundred ninety cervical PSs and 24 lateral mass screws were inserted. The incidence of PS misplacement was 19.5% (76 screws); in terms of malpositioning, 60 screws (15.4%) were classified as Grade I and 16 (4.1%) as Grade II. In total, 33 complications were observed. These included postoperative neurological complications in 11 patients in whom there was no evidence of screw misplacement (C-5 palsy in 10 and C-7 palsy in 1), implant failure in 11 patients (screw loosening in 5, broken screws in 4, and loss of reduction in 2), complications directly attributable to screw insertion in 5 patients (nerve root injury by PS in 3 and vertebral artery injury in 2), and other complications in 6 patients (pseudarthrosis in 2, infection in 1, transient dyspnea in 1, transient dysphagia in 1, and adjacent-segment degeneration in 1). The multivariate analysis showed that a primary diagnosis of cerebral palsy was a risk factor for postoperative implant failure (HR 10.91, p = 0.03) and that the presence of preoperative cervical spinal instability was a risk factor for both Grade I and Grade II screw misplacement (RR 2.12, p = 0.03), while there were no statistically significant risk factors for postoperative neurological complications in the absence of evidence of screw misplacement or complications directly attributable to screw insertion.
In the present study, misplacement of cervical PSs and associated complications occurred more often than in previous studies. The rates of screw-related neurovascular complications and neurological deterioration unrelated to PSs were high. Insertion of a PS for nontraumatic lesions is surgically more challenging than that for trauma; consequently, experienced surgeons should use PS fixation for nontraumatic cervical lesions only after thorough preoperative evaluation of each patient's cervical anatomy and after considering the risk factors specified in the present study.