Analysis of Preexistent Vertebral Rotation in the Normal Infantile, Juvenile, and Adolescent Spine

Study Design: Vertebral rotation was systematically analyzed in the normal, nonscoliotic thoracic spine of children aged 0 to 16 years. Subgroups were created to match the infantile, juvenile, and adolescent age groups according to the criteria of the Scoliosis Research Society. Objective. To determine whether a distinct pattern of vertebral rotation in the transverse plane exists in the normal, nonscoliotic infantile, juvenile, and adolescent spine. Summary of Background Data. We assume that, once the spine starts to deteriorate into a scoliotic deformity, it will follow a preexisting rotational pattern. Recently, we identified a rotational pattern in the normal nonscoliotic adult spine that corresponds to the most common curve types in adolescent idiopathic scoliosis. In infantile idiopathic scoliosis, curves are typically left sided and boys are affected more often than girls, whereas in adolescent idiopathic scoliosis, the thoracic curve is typically right sided and predominantly girls are affected. The present study is the first systematic analysis of vertebral rotation in the normal children's spine. Methods. Vertebral rotation in the transverse plane of T2-T12 was measured by using a semiautomatic method on 146 computed tomographic scans of children (0-16 years old) without clinical or radiologic evidence of spinal pathology. Scans were mainly made for reasons such as recurrent respiratory tract infections, malignancies, or immune disorders. Vertebral rotational patterns were analyzed in the infantile (0-3-year-old), juvenile (4-9-year-old), and adolescent (10-16-year-old) boys and girls. Results. In the infantile spine, vertebrae T2 T6 were significantly rotated to the left (P < 0.001). In the juvenile spine, T4 was significantly rotated to the left. In the adolescent spine, T6-T12 were significantly rotated to the right (P <= 0.001). Rotation to the left was more pronounced in infantile boys than in the girls (P = 0.023). In juvenile and adolescent children, no statistical differences in rotation were found between the sexes. Conclusion. These data support the hypothesis that the direction of the spinal curve in idiopathic scoliosis is determined by the built-in rotational pattern that the spine exhibits at the time of onset. The well-known predominance of right-sided thoracic curves in adolescent idiopathic scoliosis and left-sided curves in infantile idiopathic scoliosis can be explained by the observed patterns of vertebral rotation that preexist at the corresponding age.