Summary: Vertebral rotation was systematically analyzed in the normal, non‐scoliotic infantile, juvenile and adolescent thoracic spine. The well‐known predominance of right‐sided thoracic curves in adolescent idiopathic scoliosis (IS) and left‐sided curves in infantile IS are explained by the observed patterns of vertebral rotation that preexist at the corresponding age.
Introduction: Distribution of gender as well as distribution of left and right‐sided curves is different in infantile, juvenile and adolescent IS. In infantile IS, boys are affected more often than girls and the curve is typically left‐sided. Whereas in adolescent IS, predominantly girls are affected and the thoracic curve is typically right‐sided. It has been hypothesized that the direction of the curve is determined by an already built‐in rotational tendency of the spine, which is present at the time a spine starts to develop a scoliosis. Recently such a preexisting rotational pattern was identified in the normal, non‐scoliotic adult spine. A systematic analysis of possible preexisting vertebral rotation in the normal, non‐scoliotic infantile, juvenile and adolescent spine was performed to test this hypothesis.
Methods: Vertebral rotation of T2‐T12 was measured with semi‐automated software in 142 children (0‐16 years old) that had undergone CT examination of the thorax for reasons such as pulmonary disease or malignancies. A repeated measures ANOVA was performed to analyze differences in vertebral rotation patterns between the infantile (0‐3 years old), juvenile (4‐9 years old) and adolescent (10‐16 years old) boys and girls.
Results: Mean vertebral rotation for each age group and gender is shown in Fig. 1. In the adolescent spine, T3‐T4 were significantly rotated to the left and T6‐T12 to the right. In the infantile spine, most thoracic vertebrae were significantly more rotated to the left. In the juvenile spine, direction of vertebral rotation was equally distributed. Rotation to the left was more pronounced in infantile boys, however no statistical differences were found between the genders.
Conclusion: Although considerably less in magnitude, the preexistent rotational patterns at the infantile and adolescent age groups, are comparable to the most prevalent types of IS occurring at that age group. These data support the hypothesis that once the spine starts to decompensate, for still unknown reasons, this built‐in rotational pattern determines the direction of the spinal curve.
Significance: This is the first systematic analysis of preexistent vertebral rotation in the growing spine. Results explain the predominant direction of curvature in IS.