True accuracy of percutaneous pedicle screw placement in thoracic and lumbar spinal fixation with a CT-based navigation system: Intraoperative and postoperative assessment of 763 percutaneous pedicle screws.

PURPOSE

To investigate intraoperative reinsertion of percutaneous pedicle screw (PPS) with intraoperative CT-based navigation and to evaluate the rate of deviation of PPS at postoperative radiographic examination.

METHODS

Seven hundred sixty-three screws were inserted in 138 patients. We investigated the rate of occurrence of intraoperative PPS reinsertion after the diagnosis of screw deviation by fluoroscopy and the causes of each screw deviation. The subsequent distribution of PPS deviation was evaluated by postoperative CT. We also assess the difference in variance between the group judged to be PPS misplaced intra-/postoperatively (IOD group/POD group) and appropriate PPS placement (ND group).

RESULTS

Among all the screws inserted, 10 (1.3%) were diagnosed as being deviated by fluoroscopy during surgery, and 74 (9.7%) screws were found to be deviated at postoperative CT evaluation. We found more pedicle screw mismatch in the POD group than in the ND group (52.7 vs 11.0%, P < 0.001). The distance between the screw and the reference was greater in the IOD group than that in the ND group (1.4 ± 1.2 vs 2.4 ± 1.1 vertebral levels, P = 0.016). In one patient in the IOD group, a motor function deficit was observed postoperatively.

CONCLUSION

PPS fixation under intraoperative CT-based navigation did not prevent screw deviation completely. It is necessary to consider errors that occur during surgery and to confirm placement with real-time assistance such as fluoroscopy even in a surgery performed under CT navigation assistance.