[Acetabular reconstruction using bone allograft in the revision of total hip prosthesis]

Aseptic loosening of the acetabular component is the most worrying problem after hip arthroplasty. During revision surgery we prefer to rebuild a solid bony acetabulum close to the anatomy in which the implant will be cemented. On the basis of the first 48 acetabular reconstructions using deep-frozen bony allografts, we carried out a review of our results in a pathology which will surely increase in the future.

48 hips were operated according to this technique. It has been possible to review 38 of them, with an average follow-up of 7.3 years (extremes 5 years, and 9.6 years). The average age of the population at the time of surgery was 63 years. Two etiologies predominated: congenital hip dislocation sequelae and primitive hip arthritis. In 10 cases of massive deterioration, a Muller's ring was used to stabilize the allograft.

The results were analyzed at 6 months, 2 years, 4 years, and at maximum follow-up, clinically, according to Merle d'Aubigné grading system. Radiologically, Ranawat's criteria were used to assess the re-centering of the reconstructed hips. The development of radiolucent lines and implants migration were also assessed.

Clinically, the patients' comfort was always improved by pain relief. Radiologically, average acetabular upward migration of 5 mm and medialisation of 3.5 mm were observed. 24 hips presented radiolucent lines. 19 radiolucent lines were below 2 mm. 5 were greater than 2 mm and leaded to loosening. In 4 of these 5 cases of radiolucent lines, there were acetabular migrations with failure. The radiological image remained stable afterwards. In these cases there was a real loosening, necessitating further surgery. In all cases, partial resorption of the graft was observed.

Study of our first 38 cases shows that bony allograft and cemented acetabulum, sometimes including an armature, is one possible solution to the problem of difficult acetabular reconstructions. However, with an average follow-up of 7.3 years, we already have 5 (13 per cent) aseptic acetabular loosening, of which one has been operated on. Radiological analysis of these does not question the allograft, but rather imperfect re-centering. Analysis of the good results, 33 (87 per cent) stable acetabulum indicates re-fixing in quasi-anatomical position, in conditions close to those of a first time arthroplasty, with the aid of perfectly stabilized bony transplants, and where contact with the receiver acetabulum is maximal.

Our follow-up is one of the longest in literature. But with a migration rate already of 13 per cent, it is not yet sufficient for us to be permanently assured about the future of our patients, even if their age is greater and their activity less than those of patients having a first hip arthroplasty.