Abstract
PURPOSE: Single center evaluation of the placement accuracy of thoracolumbar pedicle screws implanted either with fluoroscopy or under CT-navigation using 3D-reconstruction and intraoperative computed tomography control of the screw position. There is in fact a huge variation in the reported placement accuracy of pedicle screws, especially concerning the screw placement under conventional fluoroscopy most notably due to the lack of the definition of screw misplacement, combined with a potpourri of postinstrumentation evaluation methods. METHODS: The operation data of 1,006 patients operated on in our clinic between 1995 and 2005 is analyzed retrospectively. There were 2,422 screws placed with the help of CT-navigation compared to 2,002 screws placed under fluoroscopy. The postoperative computed tomography images were reviewed by a radiologist and an independent spine surgeon. RESULTS: In the lumbar spine, the placement accuracy was 96.4 % for CT-navigated screws and 93.9 % for pedicle screws placed under fluoroscopy, respectively. This difference in accuracy was statistically significant (Fishers Exact Test, p = 0.001). The difference in accuracy became more impressing in the thoracic spine, with a placement accuracy of 95.5 % in the CT-navigation group, compared to 79.0 % accuracy in the fluoroscopy group (p < 0.001). CONCLUSION: This study underlines the relevance of CT-navigation-guided pedicle screw placement, especially when instrumentation of the middle and upper thoracic spine is carried out.