OBJECTIVETo evaluate the effect of axial rotation of lumbar vertebrae on the accuracy of pedicle screw placement using the traditional method, as well as to assess the value of intraoperative three-dimensional (3D) navigation in improving the accuracy.

METHODSSixteen lumbar simulation models at different degrees of axial rotation (0°, 5°, 10° and 20°), with every four assigned with the same degree, were equally divided into two groups (traditional method group and intraoperative 3D navigation group). Random placement of pedicle screws was carried out, followed by CT scan postoperatively. Then the outer pedicle cortex contours were depicted from reconstructed sectional pedicle images using Photoshop. The accuracy of pedicle screw placement was evaluated by determining the interrelationship between screw trajectory and pedicle cortex (qualitative) and measuring the shortest distance from pedicle screw axis to outer cortex of the pedicle (quantitative).

RESULTSEighty pedicle screws were implanted respectively in each group. In traditional method group, statistical difference existed in the accuracy of pedicle screw placement at different axial rotational degrees (P < 0.05). With degrees increasing, the accuracy declined. The accuracy of intraoperative 3D navigation group was higher than traditional method group in vertebrae with axial rotation (P < 0.01). In qualitative evaluation, the accuracy of the two methods had statistical difference when the degree was 20°, and in quantitative evaluation, statistical difference existed in 5°, 10° and 20° of vertebral axial rotation.

CONCLUSIONSScrew malposition can be caused by vertebral axial rotation in lumbar spine using traditional method. Accuracy of pedicle screw placement declines with the increase of axial rotational degrees. However, the accuracy can be improved by using intraoperative 3D navigation.