Biomechanical modeling of brace design.

OBJECTIVE: To study the biomechanical effectiveness of brace design parameters in right thoracic idiopathic scoliosis. METHODS: A finite element model (FEM) of the spine, rib cage, pelvis and abdomen was adapted to the geometry of 8 patients with right-thoracic idiopathic scoliosis using a multi-view radiographic reconstruction technique. A detailed parametric FEM of a thoraco-lumbo-sacral orthosis and a Box, Hunter & Hunter experimental design method were used to analyze the contribution of brace design parameters (brace size, number of straps, strap tension, position of the thoracic pad, lordosis reduction design) and of patient's spine stiffness. RESULTS: The mean Cobb angle correction of the thoracic curve was 5.1 degrees (0 degrees to 16 degrees). The most influential parameters were, in descending order, the strap tension, lordosis reduction design and spine stiffness. Their effects are independent and remain weak (-3 degrees when strap tension increases from 20 N to 60 N). Changing the position of the thoracic pad (slightly above or below the apex) doesn't have a significant effect. No significant correction of the axial rotation and rib hump was obtained. DISCUSSION & CONCLUSION: Frontal curve correction varied significantly, which justifies the need for an adequate adjustment of the brace. A more efficient design for the correction of transverse deformities remains to be found. The "active" correction component by the muscles was not included, but one can anticipate that its action would be concurrent to the passive brace mechanisms, enabling supplementary correction. A new tool simulating brace treatment has been developed, which allows rational design of braces.