We treated 31 consecutive patients of mean age 43 years (22 to 59) with severe osteoarthritis of the hip secondary to dysplasia by valgus-extension osteotomy. The clinical results were analysed for factors which may have affected the outcome. The procedure was carried out only on one side and was combined with a Chiari pelvic osteotomy in four patients. The angle of valgus was between 30 and 40° with 10 to 20° of extension. The mean follow-up was 15 years (12 to 18).

Clinical evaluation using Charnley’s modification of the system of Merle d’Aubigné and Postel showed that the mean score for pain had improved from 2.3 before operation to 4.2 at 15 years, and function from 2.3 to 4.2. Range of movement had decreased from 4.2 to 3.5. Fifteen patients (48.4%) had a pain score of 5 or 6 and were evaluated as satisfactory. Kaplan-Meier survivorship analysis showed a survival rate of 51% at 15 years after operation. Two preoperative factors had a significant positive correlation with the radiological results, namely the thickness of the capital drop osteophyte and the length of the roof osteophyte.

We suggest that this procedure is effective for advanced osteoarthritis in the dysplastic hip in young and active patients if these factors are satisfactory on the preoperative radiographs.

We performed Charnley total hip arthroplasties on 64 patients (71 hips) between 1976 and 1984 for moderate congenital acetabular dysplasia in which a superolateral cement thickness of less than 20 mm was expected when the cup was placed in the true acetabulum at an angle of 45°. Of these, 59 hips were examined 10 to 17 years after operation; 37 (group A) had been operated on between 1976 and 1982 using Charnley’s original technique of cementing the acetabulum and 22 (group B) between 1983 and 1984 using more modern techniques.

In group A, aseptic loosening of the socket was observed in ten hips (27.0%) and the 17-year survival rate was 81.5%. In group B, loosening was noted in only one socket (4.5%) and the 13-year survival rate was 100%. The improved techniques produced significantly better long-term results in fixation of the cup in dysplastic hips without bone grafting.

We performed total hip replacement on 25 congenitally dislocated hips using corticocancellous bone from the femoral neck as a bone graft to reconstruct the deficient acetabulum. Two patients (two hips) died less than five years after surgery and three hips developed deep infection. We reviewed the remaining 20 hips at an average interval of 8.4 years (5.2 to 12.9). Their functional scores (modified Merle d'Aubigne and Postel) averaged 5.7 for pain, 5.1 for walking ability, and 4.2 for range of motion. Radiographs showed union and remodelling of the grafted bone in every case. There were radiological signs of aseptic loosening in three sockets (15%) without collapse of the grafted bone. None of the femoral stems was loose. No revision operation has been performed. Our results confirm the usefulness in the long term of femoral autografts for severe acetabular deficiency.