A computed tomography-based morphometric study of thoracic pedicle anatomy in a random United States trauma population.

Object: The objective of this study was to establish normative data for thoracic pedicle anatomy in the US adult population. To this end, CT scans chosen at random from an adult database were evaluated to determine the ideal pedicle screw (PS) length, diameter, trajectory, and starting point in the thoracic spine. The role of patient sex and side of screw placement were also assessed. The authors postulated that this information would be of value in guiding safe implant size and placement for surgeons in training.

Methods: One hundred patients (50 males and 50 females) were selected via retrospective review of a hospital trauma registry database over a 6-month period. Patients included in the study were older than 18 years of age, had axial bone-window CT images of the thoracic spine, and had no evidence of spinal trauma. For each pedicle, the pedicle width, pedicle-rib width, estimated screw length, trajectory, and ideal entry point were measured using eFilm Lite software. Statistical analysis was performed using the Student t-test.

Results: The shortest mean estimated PS length was at T-1 (33.9 ± 3.3 mm), and the longest was at T-9 (44.9 ± 4.4 mm). Pedicle screw length was significantly affected by patient sex; men could accommodate a PS from T1-12 a mean of 4.0 ± 1.0 mm longer than in women (p < 0.001). Pedicle width showed marked variation by spinal level, with T-4 (4.4 ± 1.1 mm) having the narrowest width and T-12 (8.3 ± 1.7 mm) having the widest. Pedicle width had an obvious affect on potential screw diameter; 65% of patients had a least 1 pedicle at T-4 that was < 5 mm in diameter and therefore would not accept a 4.0-mm screw with 1.0 mm of clearance, as compared with only 2% of patients with a similar status at T-12. Sex variation was also apparent, as thoracic pedicles from T-1 to T-12 were a mean of 1.4 ± 0.2 mm wider in men than in women (p < 0.001). The PS trajectory in the axial plane was measured, showing a marked decrease from T-1 to T-4, stabilization from T-5 to T-10, followed by a decrease at T11-12. When screw trajectory was stratified by side of placement, a mean of 1.7° ± 0.5° of increased medialization was required for ideal pedicle cannulation from T-3 to T-12 on the left as compared with the right side, presumably because of developmental changes in the vertebral body caused by the aorta (p < 0.05 for T3-12, except for T-5, where p = 0.051). The junction of the superior articular process, lamina, and the superior ridge of the transverse process was shown to be a conserved surface landmark for PS placement.

Conclusions: Preoperative CT evaluation is important in choosing PS length, diameter, trajectory, and entry point due to variation based on spinal level, patient sex, and side of placement. These data are valuable for resident and fellow training to guide the safe use of thoracic PSs.