[Precision of navigation-assisted surgery of the thoracic and lumbar spine]

The goal of these studies was to evaluate the accuracy of in vivo and in vitro application of CT- and C-arm-based navigation at the thoracic and lumbar spine. With CT based navigation, 82 pedicle screws were consecutively inserted, 53 into the thoracic and 29 into the lumbar spine. Seven (13%) perforations were detected at the thoracic spine and two (7%) at the lumbar spine. Additionally, minor perforations below the thread depth were seen in six (11%) thoracic and in two (7%) lumbar instrumentation. With C-arm-based navigation, 74 screws were consecutively placed into 38 thoracic and 36 lumbar pedicles. Perforations were noted in ten (26%) thoracic and four (11%) lumbar implants. Minor perforations were observed in another nine (24%) thoracic and ten (28%) lumbar pedicles. The observer-independent and standardized in vitro study based on a transpedicular 3.2-mm drill hole aiming a 4-mm steel ball in a plastic bone model showed pedicle perforations of the drill canal only in thoracic vertebrae, 1 of 15 in CT-based and 3 of 15 in C-arm navigation. The quantitative calculation of the smallest distance between the central line through the drill canal and the center of the steel ball resulted in 1.4 mm (0.5-4.8 mm) for the CT-based navigation at the thoracic spine and in 1.8 mm (0.5-3 mm) at the lumbar spine. For the C-arm based navigation the distance was 2.6 mm (0.9-4.8 mm) for the thoracic spine and 2 mm (1.2-3 mm) for the lumbar spine. In our opinion, the clinical results of the comparative accuracy of CT- and C-arm-based navigation in the present study showed moderate advantages of the CT-based technique in the thoracic spine, whereas CT- and C-arm based navigation had comparable perforation rates at the lumbar pedicle. The results of the experimental study correlated with the clinical data.