Purpose of the study: There are limited data on the behaviour of intervetebral discs below arthrodesis for scoliosis. We have developed a reproducible MRI protocol for measuring the volume of the different components of the intervertebral disc and an original method for measuring disc hydration (ratio between the volume of the nucleus and the global volume). The discs studied were below fusions. The purpose of this study was to search for correlations between the disc volume and hydration and clinical outcome assess on standard x-rays.

Material and method: This was a prospective study conducted from 2005 to 2008. The series included 46 children with idiopathic scoliosis requiring arthrodesis (41 girls, 5 boys, mean age 15 years). The protocol included anteroposterior and lateral x-rays and MRI pre- and postoperatively and at 3 months and 1 year. 3D MRI reconstruction relied on a standard protocol validated in our research laboratory. These reconstructions produced a measurement of disc volume and its state of hydration. Two groups were identified using the plain x-rays: group A with what was considered to be a good result, well balanced spine and a horizontal disc subjacent to the arthrodesis; group B with what was considered a poor result with an unbalanced spine or oblique subjacent disc. Two groups were also defined as a function of postoperative reduction of the COBB angle. Group A’, reduction > 65% and group B’ reduction < 65%. Student’s t test was used for the statistical analysis.

Results: MRI series were obtained in 46 patients at 3 months and 28 at 1 year. At 3 months, there was an increase in nuclear volume (9.3%, p=0.056), global volume (5.2%, p=0.0017) and hydration (4.6%, p=0.056). At 1 year, the significant increases were, respectively, 15.4, 5.3 and 11.6%. At 3 months, there was a significant increase in disc volume in the groups A and A’. In group B, increase in disc volume (4.%, p=0.02) and hydration (13.9%, p=0.07) was only seen at 1 year.

Conclusion: This work enabled us to establish a significant correlation between increased disc volume and hydration as measured on the MRI and clinical outcome as assessed on the plain x-rays.

Purpose: Finite element analysis can be used to assess the behaviour of loaded structures. We used this method to evaluate the influence of glenoid implant design on the behaviour of an osteoarthritic scapula.

Material and methods: A 76-year-old female patient scheduled for a shoulder prosthesis underwent preoperative computed tomography of the osteoarthritic shoulder. Two polyethylene implants were evaluated: one with a triangular stem and the same prosthesis with three studs. 3D reconstruction of the glenoid cavity with the implants was then obtained and processed with the finite elements method. Three loadings were applied to the model: centred loading to reproduce the case of an ideally stable prosthesis with a normal tendinomuscular environment and excentred loading to simulate a deficient rotator cuff or prosthesis instability.

Results: With centred loading, stress remained low, to the order of 7 MPa, at the stem-glenoid cavity interface. Excentered loading produced peak stress on the borders of the glenoid implants, directly under the loading zone and at the tip of the stem, at the bone-cement interface, reaching 20 MPa. The implant tended to bend in the anteroposterior direction producing strong shear forces on the posterior part of the glenoid cavity. These forces caused micromovement at the cement-bone interface. There was no significant difference between the stem and stud implants.

Discussion: Eccentric loading of the glenoid implant appears to have a negative effect on long-term survival, the stress reaching levels greater than the values of cement fatigue fracture. Peak stress was situated on the posterior border of the cement layer due to the small space available between the implant the cortical bone in the posterior part of the osteoarthritic scapula. In this situation, the tip of the stem or the studs tend to come into contact with the posterior cortical of the scapula. When inserting a total shoulder prosthesis, it appears to be more important to keep in mind the geometry and the mechanical properties of the scapula than the implant design.