Morphometric Analysis of Anatomic Scoliotic Specimens

¹Hôpital Ste-Justine, Montreal, QC, Canada
²École Nationale Supérieure d'Arts et Métiers, Paris, France
³Cleveland Museum of Natural History, Cleveland, OH, USA
⁴École de Technologie Supérieure, Montreal, QC, Canada

PURPOSE:
Idiopathic scoliosis is a 3-D deformity affecting the orientation and position of the spine in space. The regional deformity has been studied extensively but most of the knowledge we currently have regarding the local deformity is the result of isolated observations made on rare scoliotic specimens with severe scoliotic deformities.

OBJECTIVE:
The objective of this study was to identify a typical deformation pattern for thoracic and lumbar vertebrae in idiopathic scoliosis.

MATERIAL AND METHOD:
Thirty scoliotic specimens were studied using a 3-D digitizing protocol developed by our research group using a 3-D digitizer (Fastrack, Polhemus, Vermont, USA) creating a precise 3-D reconstruction of the vertebrae. Fifteen scoliotic specimens were identified from the Hamann-Todd Osteological Collection in Cleveland and 15 were identified from the Terry Osteological Collection in Washington. We then matched every scoliotic specimen with a normal specimen based on age, sex, race, weight and height to provide for a representative control group.

RESULTS:
We measured a total of 984 vertebrae (472 scoliotic and 512 normal vertebrae) by recording the 3-D coordinates of 200 points each representing a specific anatomical landmark, creating the largest database of normal and scoliotic vertebral specimens. Parameters (134) were then calculated for each vertebra from these sets of points. One-way and two-way ANOVA calculations were done on scoliotic and normal parameters to determine a statistically significant deformation pattern. A characteristic deformity pattern was identified consisting of progressive vertebral wedging, decreased pedicle width on the concave side of the curve and articular facets surface varying greatly with all findings increasingly more important towards the apex of the curve and as curve severity increased. A relation between vertebral wedging and pedicle width on the concave side of the curve was also identified. Spinous process orientation was variable. All findings were statistically significant with p < 0.05.

CONCLUSION:
These results are of critical importance for the understanding of the local and regional deformity and in understanding curve progression. Our results also advocate caution in the use of pedicle screws in the thoracic spine especially on the concave side of the curve. With this deformation pattern available, it should be possible to predict the 3-D local vertebral deformity of any scoliotic curve based on simple PA and Lat radiographs of adolescents with idiopathic scoliosis.