Misplacement pattern of pedicle screws in pediatric patients with spinal deformity: a computed tomography study

Abstract

Study design: A retrospective study.

Objective: To assess the accuracy and safety of the placement of pedicle screws (PSs) on pediatric patients with congenital scoliosis younger than 10 years through postoperative computed tomography scanning.

Summary of background data: The accuracy of the placement of PSs in adolescent and adult population with various spinal problems has been thoroughly studied. As to pediatric patients with congenital scoliosis, the related studies were inadequate.

Methods: A retrospective review was carried out on 96 pediatric patients with congenital scoliosis younger than 10 years, who received posterior instrumentation from February 2006 to June 2012. Postoperative computed tomography scan was utilized to assess the accuracy of PS placement. Pedicle perforations were classified as medial, lateral, or anterior and categorized into 1 of the 4 grades: grade 1, ≤2 mm; grade 2, 2.1-4.0 mm; grade 3, 4.1-6.0 mm; and grade 4, ≥6.1 mm. The positions of screws were also classified as acceptable (fully contained screws or with screws either ≤2 mm of medial wall perforation or ≤6 mm of lateral wall perforation and without impingement of visceral organs) or unacceptable. The variables of PS position as defined by fully contained were statistically examined using the Fisher (2-tailed) exact test with regard to: (1) side of the curvature (convex or concave); (2) region (thoracic or lumbar); (3) magnitude of the curvature (<80 or ≥80 degrees); (4) apical region and overall; (5) deformed vertebrae versus normal vertebrae.

Results: A total of 625 screws were inserted; 512 screws (81.9%) were fully contained within the cortical boundaries of the pedicle. Five hundred and sixty-nine (91.0%) screws were regarded as acceptably positioned, whereas 56 (9.0%) as unacceptably positioned. Among the 625 screws inserted, 317 were inserted in the thoracic spine, of which 67 (21.1%) led to perforations, and 308 screws were inserted in the lumbar spine, of which 46 (14.9%) led to perforations. The breach rate was found to be higher in thoracic spine than in the lumbar spine (P<0.05). Of the 625 screws inserted, 306 screws were inserted on the concave side of the curve, of which 50 (16.3%) led to perforations, and 319 were inserted on the convex side, of which 63 (19.7%) led to perforations. There is no difference with regard to the breach rates between the concave side and the convex side of the curve (P>0.05). Of the 625 screws, 328 were inserted in patients with a Cobb angle <80 degrees, of which 41 (12.5%) screws perforated the pedicles, and 297 screws were inserted in the patients with a Cobb angle >80 degrees, of which 72 (24.2%) screws perforated the pedicles. The breach rate was higher in the patients with the Cobb angle >80 degrees than in the patients with the Cobb angle <80 degrees (P<0.05). One hundred and fifty-four screws were inserted in the apical vertebrae, of which 57 (37.1%) screws perforated the pedicle walls. The breach rate in apical vertebrae was higher than the overall (55/154 vs. 113/625) (P<0.05). Among the 625 screws, 252 screws were inserted in the deformed vertebrae, of which 58 (23.0%) perforated the pedicle walls, and 373 screws were inserted in the normal vertebrae, of which 55 (14.7%) perforated the pedicle walls. The breach rate in the deformed vertebrae was higher than the normal vertebrae (P<0.05).

Conclusions: PSs insertion in pediatric patients with congenital scoliosis was less accurate than that in adult patients. However, the safety of placement was quite satisfactory. The possibility of pedicle breaching increased when placing PSs at deformed vertebra, on the apex of the curve, and in the thoracic spine, which reminds surgeons to be more cautious when placing PSs in these regions.