CT analysis of the posterior anatomical landmarks of the scoliotic spine.

AIM: To use computed tomography (CT) to assess the validity and reliability of the posterior landmarks, spinous processes (SP), transverse processes (TP), and centre of lamina (COL), as compared to the Cobb angle to assess the curve severity and progression of adolescent idiopathic scoliosis (AIS). MATERIALS AND METHODS: A consecutive series of CT examinations of severe AIS patients were included retrospectively. SP, TP, and COL angles were measured for all curves and compared to the Cobb angle. RESULTS: One hundred and five patients were included. The mean Cobb versus SP, TP, and COL angles were, 54° versus 37°, 49°, and 51° in the thoracic curves and 34° versus 26°, 31°, and 34° in the (thoraco)lumbar curves. Intraclass correlation coefficient values for intra-rater measurements of the SP, TP, and COL angles were 0.93, 0.97, and 0.95 and 0.70, 0.90, and 0.88 for inter-rater measurements. The correlations between the Cobb angle and SP, TP, and COL angles in thoracic and (thoraco)lumbar curves were 0.79 and 0.66, 0.87 and 0.84, and 0.80 and 0.70. CONCLUSIONS: The posterior spinal landmarks can be used for assessment of scoliosis severity in AIS; however, they show a systematic underestimation, but a strong correlation with the coronal Cobb angle. TP and COL angles had the highest validity.

Anterior lengthening in scoliosis occurs only in the disc and is similar in different types of scoliosis.

Relative anterior spinal overgrowth (RASO) was proposed as a generalized growth disturbance and a potential initiator of adolescent idiopathic scoliosis (AIS). However, anterior lengthening was also observed in neuromuscular (NM) scoliosis, was shown to be restricted to the apical areas and to be located in the intervertebral discs, not in the bone. In this study the goal was to determine if other scoliotic curves of known origin exhibit the similar mechanism of anterior lengthening without changes in the vertebral body. Therefore CT-scans of 18 patients in whom a short segment congenital malformation had led to a long thoracic compensatory curve without bony abnormality were included. Of each vertebral body and intervertebral disc in the compensatory curve, the anterior and posterior length was measured on CT-scans in the exact mid-sagittal plane, corrected for deformity in all three planes. The total AP% of the compensatory curve in congenital scoliosis showed a lordosis (+1.8%) that differed from the kyphosis in non-scoliotic controls (-3.0%; p<0.001), and was comparable to AIS (+1.2%) and NM scoliosis (+0.5%). This anterior lengthening was not located in the bone; the vertebral body AP% showed a kyphosis (-3.2%), similar to non-scoliotic controls (-3.4%), as well as AIS (-2.5%) and NM scoliosis (-4.5%; p=1.000). However, the disc AP% showed a lordosis (+24.3%), which sharply contrasts to the kyphotic discs of controls (-1.5%; p<0.001), but was similar to AIS (+17.5%) and NM scoliosis (+20.5%). The results demonstrate that anterior lengthening is part of the three-dimensional deformity in different types of scoliosis and is exclusively located in the intervertebral discs. The bony vertebral bodies maintain their kyphotic shape, which indicates that there is no active bony overgrowth. Anterior lengthening appears to be a passive result of any scoliotic deformity, rather than being related to the specific cause of AIS.