Purpose: Acetabular revision has become a challenging situation due to the importance of bone stock loss encountered in SOFCOT stage III acetabula. The number of failures due to loosening are explained by the strong mechanical stress on the bone grafts or inadequate restitution of the rotation centre of the hip. The purpose of this study was to evaluate mid-term results of the Kerboull support used to achieve anatomic recentring of the hip and progressive weight bearing on the bone grafts.

Material and methods: This retrospective series included 54 acetabular revisions performed for stage III loosening between 1989 and 1996. A Kerboull support was used in all cases. The patients were assessed with the Postel Merle d’Aubigné score and radiographically on plain pelvis films in order to search for recurrent loosening or arthroplasty failure. The log rank test was used to compare actuarial survival.

Results: The series included 62% women. Mean age was 62.3 years (33–87). This was the first revision for 78% and a second or more revision for 22%. The preoperative Postel Merle d’Aubligné score was 9.18 points. This score was 12.3 postoperatively, 15.6 at one year, 15.5 at five years and 14.8 at last follow-up. Dislocation was the most frequent complication, with 55% occurring on cups more than 46° oblique. Grafts were considered radiographically integrated in 58% of the cases. There were 5.5% failures due to migration, 13.8% due to fracture of the superior screw. The actuarial survival was 97.4% at three years, 94.7% at four years, 89.2% at five years and 73% at seven and ten years.

Conclusion: On the basis of these good short-, mid- and long-term clinical and radiographic results, we recommend Kerboull support for the treatment of stage III acetabular loosening.

Purpose: The purpose of this study was to analyse the results and morbidity of video-assisted minimally invasive thoracoctomy for anterior arthrodesis of thoracolumbar fractures treated with a two-stage procedure and to evaluate mid-term outcome.

Material and methods: This retrospective series included 6 patients with an unstable thoracolumbar fracture who underwent surgery between November 1997 and June 2002. A two-stage procedure was used: posterior reduction osteosynthesis and anterior arthrodesis with a tricortico-cancellous graft via video-assisted minimally invasive thoracotomy. The cohort included six women and eighteen men, mean age 34.5 years. Fractures were located at: L1 (n=4), T12 (n=10), T11 (n=2). At initial assessment the Franckel classification was: A (n=3), B (n=1), C (n=1), D (n=3), 3 (n=18). Time between the posterior procedure and the anterior thoracotomy was 30.2 days (range 6–86). Postoperative results as well as the time to fusion were recorded. Mean follow-up was 21 months (range 6–45) for functional and radiological assessment. No patient was lost to follow-up.

Results: Mean operative time was 188 min (range 80–240). Mean blood loss was 235 ml (range150–1000) with no intraoperative event requiring conversion to open thoracotomy. Mean duration of morphine administered postoperatively was 2.2 days, the same as for thoracic drainage. Residual pleural effusion was observed in one patient and residual pneumothorax in two; all resolved spontaneously. Mean hospital stay was 12 days (range 6–27). Twenty-five patients had achieved fusion at four months. One patient developed a radiological non-union which was asymptomatic at one year. At last follow-up, the Oswestry function score was 22.6% for the entire series, 18% for Franckel D or E patients (n=21) and 42% for Franckel A, B and C patients (n=5). Loss of angular correction of the spinal kyphosis and regional traumatic angulation between the postoperative films and the last follow-up films was 2 (mean).

Discussion: This series of complementary anterior arthrodesis by video-assisted minimally invasive thoracotomy confirmed the minimally invasive nature of this approach in comparison with thoracophrenolaparotomy and its complications. At mid term, this technique has provided satisfactory functional and radiographic results. Applied for thoracolumbar fractures, this combined surgical option can limit intraoperative morbidity and assure good mid-term results.

Purpose: The purpose of this study was to analyse results of the inverted shoulder prosthesis (Grammont Delta III prosthesis) for surgical treatment of malignant tumours of the upper portion of the humerus in twelve patients.

Material and methods: Twelve patients were treated in two centres (six patients each). There were five women and seven men, mean age 51 years (34–69). Seven had a primary tumour and five had a unique secondary tumour. All tumours were implanted after Malawer I resection (IA four case and IB two cases), with an autologous bone graft using resected irradiated bone. All resections were wide as needed for healthy margins. Resection extended to the lower fourth of the humerus in four cases, to between the lower fourth and third in six cases and to the upper third or less in two cases. Ten patients were reviewed at a mean follow-up of 22 months (7–60) to assess clinical function (Constant score) and radiographic results. Two patients with a unique secondary tumour died within the six months following surgery due to progression of the primary tumour.

Results: The mean Constant score was 58.75 (30–81.1). Weighted for age and gender, the result was 70.3% (32.6–82). Five sedentary patients were able to resume their former activities (1/2). Radiographically, there were two resorptions of the autograft, two glenoid lucent lines that did not progress, four notches in the column, and three Sneppen 1 calcifications. Mechanically, there were no prosthesis dislocations in patients whose resection extended to the lower fourth of the humerus, four dislocations in the one-fourth one-third group (2/6) and two dislocation in the upper group (2/2).

Discussion, conclusion: Functional and radiographic outcome after anatomic prosthesis implantation for surgical treatment of malignant tumours of the upper portion of the humerus was directly related to the status of the rotator cuff. The Grammont Delta III inverted prosthesis provided satisfactory functional and radiographic results in this indication after Malawer I resection, irrespective of the rotator cuff sacrifice. The risk of prosthesis instability appeared to be related to the extent of the humeral resection required for cancerological cure. For patients who undergo resection of one third or more of the humerus, muscle plasty should be associated.