Introduction: To determine if left thoracic adolescent idiopathic scoliosis (AIS) is merely a mirror image of the common right thoracic curve pattern.
Methods: A prospective multi‐center AIS database was queried to identify 44 left (L) thoracic curves (Lenke 1, 2 or 3). These were compared to 895 right (R) thoracic curves collected over the same time period. The coronal and sagittal measures were compared with ANOVA (p<0.05).
Results: The distribution of Lenke types differed slightly between groups (Lenke 1: 79% L, 71% R, Lenke 2: 9% L, 24% R, and Lenke 3: 11% L, 5% R, p=0.03), with fewer double thoracic L curves. Age at surgery was similar between the groups (14.8 L, 14.6 R, p=0.5). The coronal main thoracic Cobb angle at surgery was also similar between the groups (L: 55 ± 15, R: 54 ±11, p=0.5), suggesting a similar age of onset given the similarity in age at surgery. Sagittal plane differences were noted with thoracic kyphosis (T5‐T12) significantly less in the R curves compared to L curves (p<0.001). Although the two groups had similar lordosis (p>0.05), the R curves also had increased pelvic incidence and sacral slope compared to L curves (p<0.05, p<0.02 respectively), see Table.
Conclusion: Although left and right curves appear nearly as mirror images in the coronal plane, the sagittal profiles were significantly different between these curve patterns. The decreased thoracic kyphosis found in the right thoracic curve patients (especially in the setting of increased pelvic incidence) is consistent with the theory of relative anterior thoracic spinal overgrowth as a cause for thoracic scoliosis.
Significance: The normal thoracic kyphosis and pelvic incidence found in the left thoracic curve patients is not consistent with the overgrowth theory and may signify another pathomechanism for this curve pattern development. This difference in sagittal alignment should also be kept in mind when planning surgical correction for left thoracic curves.