RESUMO
STUDY DESIGN: This was a prospective study of 35 consecutive patients in whom pedicle screw position was assessed after surgery, using lateral radiographs and computed tomography. OBJECTIVE: To evaluate the accuracy of plain radiographs and computed tomography in assessment of pedicle screw position. SUMMARY OF BACKGROUND DATA: Imaging techniques, such as postoperative anteroposterior and lateral plain radiographs and computed tomography, are currently the primary means of assessing pedicle screw placement. METHODS: Postoperative radiographs and computed tomographic scans were used to evaluate the position of 220 pedicle screws inserted in the spines of 35 consecutive patients who underwent thoracic and lumbar spine fusion and instrumentation. No recognized neurologic complication resulted from pedicle screw placement. Screw position was graded as in, out, or questionable. All observations were performed independently by three observers. The authors also analyzed the position of the screws according to the underlying spinal disease. RESULTS: More misplaced screws were clearly seen on computed tomographic scans than on plain radiographs; however, this difference was not statistically significant. Interobserver differences were not statistically significant. Intraobserver differences approached statistical significance when the results of the two test were compared. CONCLUSIONS: Although the accuracy of computed tomographic imaging is better than that of plain radiographs, the difference does not reach statistical significance. Postoperative use of plain radiographs remains a reliable method for evaluation of pedicle screw insertion in the absence of neurologic deficit.
