The Latarjet procedure utilises the coracoid as a vascularised bone autograft to augment the glenoid in patients with shoulder dislocation, especially where there is a bony lesion affecting the glenoid. A modification of the Latarjet procedure, pioneered in Cape Town, South Africa, rotates the coracoid so that its curved under-surface matches that of the glenoid. The aim of this study was to measure the radii of curvature of the glenoid and the coracoid to see how well the curved under-surface of the coracoid matches the glenoid’s surface curvature.

An initial study of 210 cadaveric scapulae was performed in which the radii of curvature of the surface of the glenoid and the curved under-surface of the coracoid were measured. We found that the curves are very similar. The glenoid’s surface had a median curvature of 30mm (inter-quartile range from 25mm to 30mm) and the coracoid had a median curvature of 22.5mm (inter-quartile range from 20mm to 25mm). The curvature of the glenoid in these dry specimens was slightly larger than the corresponding coracoid curvature. In life this difference would be minimised by articular cartilage, labrum and the attachment of capsule (another Cape Town modification).

A further parallel CT based study was set up at Derbyshire Royal Infirmary in England. The same radii of curvature where measured and compared using 3D CT reconstruction on a further 20 scapulae from living patients. These measurements also support the cadaveric similarities with a mean glenoid curvature of 23.9mm and coracoid of 25.4mm respectively. Using a paired t-test no statiscally significant difference was found between the corresponding data (p=0.2488)

This study confirms the native anatomy of the coracoid is perfectly suited for this modification of the Latar-jet procedure.

Purpose: The causes of glenoid loosening are multifactorial (implant design, surgical technique, bone properties, soft tissue properties). This biomechanical study was conducted to evaluate the consequences of two clinical problems often encountered in shoulder arthroplasty: subscapular tension and glenoid retroversion.

Material and methods: We developed a 3D model of the shoulder including the rotator cuff. A total prosthesis was implanted by digital modellisation. The humeral prosthesis imitated the adaptable third-generation implants, with a stem and a portion of a metal sphere, were used to achieve anatomic reconstruction of the proximal humerus. The polyethylene glenoid, cemented to bone, had a central stem and a flat base. Two subscapular tension (normal and twice normal) and two glenoid positions (0° and 20° retroversion) were tested. External rotation (0–40°) and internal rotation (0–60°) were simulated. We calculated displacement of the glenohumeral contact point, joint forces and contact pressures, interosseous glenoid stress, and micromovement of the bone-cement-implant interfaces.

Results: Subscapular tension produced increased forces and joint pressures, associated with moderate posterior translation of the glenohumeral contact point. Retroversion induced more marked posterior displacement of the contact point, leading to significantly higher intraosseous glenoid stress and micromovements at the interfaces. The association of subscapular tension and glenoid retroversion produced important concentration of stress forces in the posterior part of the glenoid and increased all the micromovements.

Discussion: Subscapular tension and retroversion of the glenoid implant have significant biomechanical effects which can favour glenoid loosening. Correction of these two parameters must be carefully controlled during shoulder arthroplasty.

Purpose: Dislocation is a short-term complication frequently encountered after implantation of a total hip arthroplasty (THA). Different strategies can be used to limit the influence of technical, particularly surgical, factors. The position of the acetabular element is a key factor, particularly the anteversion angle and the abduction angle. The purpose of this work was to determine the precision, the reproducibility, and the ease of use of a new mechanical guide for insertion of the acetabular cup.

Material and methods: After calculating the sample size necessary to achieve 90% statistical power for a 5% type I error, we had five surgeons who regularly implanted THA implant 310 press-fit hip cups on a plastic anatomic model of the pelvis. A new mechanical guide was developed using the constant direction of gravity as the reference frame. We determined the precision of acetabular cup implantation, its reproducibility, and ease of use compared with that of the Müller mechanical guide during in vitro implantation of 310 cups via a posterolateral approach that allowed the usual vision of the operative field.

Results: The error of cup anteversion relative to the reference set at 15 was 10.4±5.0 (range 3–21) for the Müller guide and 0.4±0.7 (range 1–3) for the new guide. Cup abduction, relative to the reference set at 45, was −4.7±2.3 (range 7–11) for the Müller guide and 0.3±0.5 (range 0–3) for the new guide. Mean time for positioning the cup was comparable with the two guides (mean 6s for the Müller guide and 5s for the new guide).

Discussion: The precision and reproducibility of cup positioning obtained with the new guide are better than those obtained with mechanical guides currently available on the market (p< 0.00001 with the Müller guide). They are more comparable with values found in in vitro studies using computer-assisted surgery techniques. Use of the new guide was also found to be rapid and simple.

Conclusion: The excellent results obtained with this new mechanical guide, as assessed in terms of cup position for THA, should be confirmed with in vivo trials.

In order to determine the incidence of avascular necrosis after osteotomy of the talar neck, we re-evaluated 11 patients (16 feet) with idiopathic club foot who had undergone this procedure at a mean age of eight years (5 to 13) to correct a residual adduction deformity. All had been initially treated conservatively and operatively. The mean follow-up was 39 years (36 to 41). Surgery consisted of a closing-wedge osteotomy of the talar neck combined, in 14 feet, with lengthening of the first cuneiform and a Steindler procedure.

At follow-up eight feet were free from pain, three had occasional mild pain and five were regularly painful after routine activities. Two patients were unlimited in their activity, six occasionally limited after strenuous and three regularly limited after strenuous activity. Using the Ponseti score, the feet were rated as good in four, fair in three and poor in nine. In seven feet avascular necrosis with collapse and flattening of the talar dome had occurred. In all of these feet the children were younger than ten years of age at the time of surgery. In three feet, avascular necrosis of the talar head was also observed.

We conclude that osteotomy of the talar neck in children under the age of ten years can cause avascular necrosis and should be abandoned.

Two calcium phosphate cements, brushite and hydroxyapatite, have been recently developed as bone substitution materials. The brushite cement is biocompatible, resorbable, osteoconductive and injectable since it hardens in physiological conditions. In contrast, hydroxyapatite is less resorbable and is not injectable. However, hydroxyapatite presents a higher strength, which may open the perspective of use in weight-bearing regions of the skeleton subjected to multi-axial stresses. The purpose of this work is a full characterization of the multiaxial elastic and failure behaviour of these two cements in a moist environment.

The brushite cement was prepared by mixing three phosphate powders in presence of water. A mixture of monetite and calcite powders in presence of water was used to obtain hydroxyapatite self-setting cement. Cylindrical, hollow specimens (Øext=18mm, Øint=14mm, L=40mm) were manufactured to apply uniaxial and torsional deformations. The specimens were cast with a custom mould, avoiding any machining, and thus, residual stresses. Scanning electron microscopy and x-ray diffraction were used to examine the cement microstructures and to determine their final material phases. An MTS axial-torsional machine was used for all mechanical tests. Compression, tension and torsion tests were performed each on five brushite and five hydroxyapatite specimens under moist conditions. Uniaxial and biaxial extensometers were used to measure the elastic moduli and the Poisson ratio.

The brushite cement exhibited failure properties comparable or below those of average human cancellous bone and confirmed its indication as a bone filling material (Brushite failure strength : 1.3±0.3 MPa in tension, 2.9±0.4 MPa in shear and 10.7±2.0 MPa in compression). The hydroxyapatite cement had an order of magnitude larger compressive strength (75±4.2 MPa), comparable tensile (3.5±0.9 MPa) and shear (4.8±0.3 MPa) strengths as average human cancellous bone. As expected, the latter cement seems to be more compatible with a multiaxial weight-bearing function in bone substitution.