Sensitivity and Specificity of Plain Radiographs and CT in Assessing Lumbar Pedicle Screw Placement

· (a – USA Synthes)

INTRODUCTION: This study was designed to evaluate the accuracy of radiographs and/or CT in determining the position of pedicle screws.

METHODS: Eighteen human cadaveric specimens had the lumbar spine exposed by a posterior approach. Pedicle screws (6.5mm) were inserted in the standard fashion. The spines were then dissected out en bloc. AP and lateral plain films and CT scans were obtained. Dissection was performed to examine screw location. A different examiner reviewed the radiographs and CT to determine pedicle screw position. Screws were considered out if there was evidence of cortical perforation or if the screw was extra-pedicular.

RESULTS: The sensitivity and specificity of plain radiographs was 84.3%, and 28.6%, and 89.3% and 61.9% for CT. The combination of CT and radiographs increased the sensitivity and specificity to 99.3% and 100%. A non-parametric chi square analysis failed to demonstrate any statistical difference between plain radiographs and CT (P< .05). However, statistical comparison of either CT or plain radiographs to that of both modalities showed a significant difference (p < .05). In addition, the accuracy of either CT or plain radiographs were similar, however, when imaging modalities were combined, the accuracy increased to 99%.

DISCUSSION: Our findings indicate that there was no statistical difference between CT and plain radiographs. However, there was a statistically significant difference when both modalities were utilized together.

· If noted, the author indicates something of value received. The codes are identified as: a-research or institutional support; b-miscellaneous funding; c-royalties; d-stock options.