What is the Difference in Morphologic Features of the Thoracic Pedicle Between Patients With Adolescent Idiopathic Scoliosis and Healthy Subjects? A CT-based Case-control Study.

Background: Describing the morphologic features of the thoracic pedicle in patients with adolescent idiopathic scoliosis is necessary for placement of pedicle screws. Previous studies showed inadequate reliability owing to small sample size and heterogeneity of the patients surveyed.

Questions/purposes: To use CT scans (1) to describe the morphologic features of 2718 thoracic pedicles from 60 female patients with Lenke Type 1 adolescent idiopathic scoliosis and 60 age-, sex-, and height-matched controls; and (2) to classify the pedicles in three types based on pedicle width and analyze the distribution of each type.

Methods: A total of 2718 pedicles from 60 female patients with Lenke Type 1 adolescent idiopathic scoliosis and 60 matched female controls were analyzed via CT. All patients surveyed were diagnosed with adolescent idiopathic scoliosis, Lenke Type 1, at the First Affiliated Hospital of Sun Yat-sen University, and all underwent pedicle screw fixation between January 2008 and December 2013 with preoperative radiographs and CT images on file. We routinely obtained CT scans before these procedures; all patients who underwent surgery during that period had CT scans, and all were available for analysis here. Control subjects had CT scans for other clinical indications and had no abnormal findings of the spine. The control subjects were chosen to match patients in terms of age (15 ± 2.6 years versus 15 ± 2.6 years) and sex. Height of the two groups also was matched (154 ± 9 cm versus 155 ± 10 cm; mean difference, -1.06 cm; 95% CI, -1.24 to -0.81 cm; p < 0.001). Pedicle width and length were measured from T1 to T12. The thoracic spine was classified in four regions: apical vertebra in the structural curve (AV-SC), nonapical vertebra in the structural curve (NAV-SC), apical vertebra in the nonstructural curve (AV-NSC), and nonapical vertebra in the nonstructural curve (NAV-NSC). Pedicles were classified in three types: pedicle width less than 2 mm as Type I, 2 mm to 4 mm as Type II, and greater than 4 mm as Type III. Types I and II were defined as dysplastic pedicles. Paired t test, independent samples t test, one-way ANOVA, followed by Bonferroni's post hoc test and chi-square or Fisher's exact tests were used for statistical comparisons between patients and controls, as appropriate.

Results: No difference was found between pedicle width on the convex side (PWv) and in controls (PWn), but pedicle width on the concave side (PWc) (4.99 ± 1.87 mm) was found to be narrower than PWv (6 ± 1.66 mm) and PWn (6 ± 1.45 mm). The variation degree of pedicle width (VDPW) was greatest in the AV-SC region (34% ± 37%), in comparison to AV-NSC (20% ± 25%) (mean difference, 14%; 95% CI, 1.15%-27%; p = 0.025), NAV-SC (17% ± 30%) (mean difference, 17%; 95% CI, 7%-27%; p < 0.001), and NAV-NSC (11% ± 24%) (mean difference, 24%; 95% CI, 13%-34%; p < 0.001). Dysplastic pedicles appeared more in patients with adolescent idiopathic scoliosis (22%; 293 of 1322) compared with controls (13%; 178 of 1396) (odds ratio [OR] = 0.51; 95% CI, 0.42-0.63; p < 0.001). In patients with adolescent idiopathic scoliosis, they commonly occurred on the concave side 34% (228 of 661) and on the AV-SC region (32%; 43 of 136).

Conclusions: Pedicle width on the concave side was narrower than pedicle width on the convex side and pedicle width in healthy control subjects. The apical vertebra in the structural curve was the most variegated region of the curve with the highest prevalence of dysplastic pedicles.

Clinical Relevance: Our study can help surgeons perform preoperative assessments in females with adolescent idiopathic scoliosis, and with preoperative and intraoperative management for difficult pedicle screw placement. In particular, our results suggest that surgeons should exercise increased vigilance when selecting pedicle screw dimensions, especially in the concave aspect of the mid-thoracic curve, to avoid cortical breeches. Future studies should evaluate other Lenke types of adolescent idiopathic scoliosis, and males with adolescent idiopathic scoliosis.