Intraoperative O-arm navigation systems improve the accuracy of spinal instrumentation placement. However, deviation of the pedicle screw from the guide line might occur. The aim of the present study was to explore the causes of and countermeasures for the drift phenomenon during pedicle screw implantation with the aid of an O-arm three-dimensional navigation system in spinal deformity surgery. This was a retrospective analysis of 341 patients with spinal deformity who underwent O-arm navigation system-assisted pedicle screw placement from July 2015 to June 2019. The patient's general condition, Cobb angle, apical vertebra position, softness index, spinal release status, fixed reference frame position, and distance between the navigation vertebral body and the reference frame were collected and compared by independent-samples t test or Pearson's chi-square analysis. The potential risk factors for the drift phenomenon were identified using binary logistic regression analysis. The drift phenomenon occurred in 57 patients during the first navigation-assisted pedicle screw placement, for an incidence of 16.7% (57/341). There were significant differences in factors such as the apical vertebra position, softness index, spinal release status, and distance between the vertebral body and the reference frame when the drift phenomenon occurred (P < 0.05). Binary logistic regression analysis showed that the softness index, spinal release status, and distance between the vertebral body and the reference frame when drifting occurred were independent risk factors for the drift phenomenon during O-arm navigation-assisted pedicle screw placement. During the use of an O-arm navigation system to assist with pedicle screw placement, pedicle screws should not be placed away from the reference frame, and spinal osteotomy and release should be performed after pedicle screw placement. In addition, the accuracy of O-arm navigation-assisted pedicle screw placement will be affected more in those with larger softness indices.
Read full abstract