Spinal navigation: standard preoperative versus intraoperative computed tomography data set acquisition for computer-guidance system: radiological and clinical study in 100 consecutive patients.

STUDY DESIGN: A retrospective clinical and radiological study. OBJECTIVE: To compare the safety and accuracy of pedicle screw insertion using two different computed tomography (CT) data set acquisitions (preoperative and intraoperative) for computer-guidance systems in a series of 100 consecutive patients. SUMMARY OF BACKGROUND DATA: Misplacement and pedicle cortical violation occurs in over 20% of screw placements and can result in potential neurovascular complications. Many technological innovations have been described to help reduce this range of error, such as image-guided surgery using fluoroscopy or CT-based image guidance. However, these techniques are not without their drawbacks. The next technological evolution is the use of an intraoperative CT scan, which would allow us to solve some of the critical phases of spinal navigation, such as position-dependent changes, thus granting a higher accuracy of the navigation system. The authors have compared and discussed the results of a preoperative and intraoperative CT data set acquisition mode for spinal navigation. METHODS: One hundred consecutive patients with a diagnosis of lumbar degenerative spondylolisthesis who underwent a surgical approach of lumbar pedicle screw fixation using a CT-based computer-guidance system were evaluated. The population was divided into two groups: in group I, a preoperative CT scan was used for the navigation system; whereas in group II, an intraoperative CT scan acquired during surgery was used. Epidemiological and surgical data of the patients in the two groups were then analyzed and compared. The Pearson χ test was used for comparisons between groups (significance level 0.05). The evaluation and classification of the screw positioning was performed on the basis of a control CT scan according to the classification proposed by Laine. RESULTS: Out of 504 screws, 471 were correctly inserted into the pedicles (93.5%): the accuracy of group I was of 91.8%, whereas in group II it was 95.2% (no statistical significance). The overall rate of perforation was 6.5% (33 screws): 21 in group I and 12 in group II. Twenty-eight screws had a perforation of the pedicle less than 2 mm (Grade I), three comprised from 2 to 4 (Grade II), and only two more than 4 mm and less than 6 mm (Grade III). Out of 33 misplaced screws only one was replaced (graded as III in group II). Surgical time was shorter for group II, with a statistically significant difference. This result is mainly because of the automatic recognition and merging of the intraoperative images with the surgical anatomy that avoided the phase of registration with a paired-point technique. CONCLUSION: The results of this study suggest that the CT-based computer-assisted surgical navigation systems are precise, granting an elevated accuracy in pedicle screw positioning.