gms | German Medical Science

Objective: For spinal surgery, images for navigational systems are usually acquired before surgery with patients in a supine positione. Registration of pre-operative images for navigation is comparably time consuming and has to be carried out point-to-point for each vertebral body leaving a substantial risk of inaccurate screw placement. The aim of this study was to evaluate prospectively navigated procedures in spinal surgery with data acquisition by intraoperative computed tomography (iCT) before or during the operative procedure.

Methods: High-resolution CT data of 34 patients [thoracolumbar instability (n=21), C1/2 instability (n=6), cervicothoracic ventral stabilization (n=6), lateral disk herniation (n=1)] were acquired after positioning the patient in the final prone position on a radiolucent operating table (Jupiter, Trumpf) with optional use of a radiolucent head clamp. A sliding gantry 24 detector row CT (Somatom Open, Siemens) was used for image acquisition. In all cases, the data were imported to the frameless infrared-based neuronavigation station (Vector Vision Compact, BrainLAB). A postprocedural intraoperative CT was obtained to assess the extent of decompression and the accuracy of instrumentation.

Results: Intraoperative region-matched registration revealed computed accuracy <1mm (0.9±0.1 mm) in all cases, while anatomical and fluoroscopic landmark control showed sufficient accuracy even in cases of up to five vertebral bodies below the vertebral body, utilized for initial registration. Time needed for stabilization procedure (navigation-CT, screw-placement, including intraoperative control-CT) was within usual limits for C1/2 (127±21 min), lumbar (4 screws; 100±24 min) and thoracic (8 screws; 170±38 min) transpedicular screw placement. Follow-up-iCT revealed an incorrect screw position >2 mm in 9/132 screws (6.8%), allowing immediate correction in 5 screws without any damage to vessels or nerves. There were 3 transient complications with constant clinical improvement in all patients (Odom 2.8 to 1.8).

Conclusions: Intraoperative CT in combination with neuronavigation provides high accuracy of screw placement and thus safety for patients undergoing spinal stabilization in all areas of the spine. The system can be installed into a pre-existing operating environment without the need for special surgical instruments. The procedure is rapid and easy to perform without restricting access to the patient and – by replacing pre- and postoperative imaging – is not associated with additional exposure to radiation.