Planning C2 pedicle screw placement with multiplanar reformatted cervical spine computed tomography

Object: Careful preoperative planning with thin-slice computed tomography (CT) scan is useful for hardware placement at C2. Prior studies have shown considerable variability in the proportion of C2 vertebrae considered safe for pedicle screw placement, depending on the imaging technique used. Our work sought to more carefully define that proportion using a refined imaging technique on a large number of submillimeter CT scans. Materials and Methods: We reviewed 150 submillimeter cervical spine studies randomly selected from CT scans performed at a Level 1 trauma center. OsiriX (TM) image analysis software was used to propagate a 5-mm cylinder through the plane of the pedicle on paracoronal reformatted CT scans. Hounsfield unit attenuation was used to determine whether the cylinder violated the pedicle. Binomial data were generated to determine the proportion of pedicles that would allow safe screw placement. Results: We analyzed 300 pedicles in 150 patients. Using a standard C2 pedicle starting point, 32% of pedicles were breached by the 5-mm diameter cylinder. When screw trajectory was adjusted by moving the cylinder to fit the pedicle isthmus, establishing an optimized starting point, only 14% of pedicles were breached. Average pedicle length was 27.3 mm for screws that would have crossed the isthmus versus 13.2 mm for screws that would have stopped short due to potential breach. Conclusions: Findings of the current work suggest that preoperative imaging analysis or navigation can be useful adjuncts when anatomical variants are present.