Between 1993 and 2003, 67 consecutive revision total hip arthroplasties were performed in 65 patients, including 52 women and 13 men, using hydroxyapatite (HA) granules supported by a Kerboull-type reinforcement acetabular device. The average age at the time of index surgery was 68.6 years. The Acetabular bone loss according to the American Academy of Orthopaedic Surgeons (AAOS) system was type II for 7 hips, type III for 58 hips, and type IV for one hip. The Kerboull-type acetabular reinforcement device used was Kerboull Cross Plate in 18 hips and KT Plate in 49 hips. HA granules of sizes 0.9 to1.2 mm (G4) and 3.0–5.0 mm (G6) were mixed in a ratio of 1:1. Autografts were used to reconstruct the major segmental defects in 7 hips. At the time of this study 30 hips were lost of follow-up. Among 30 hips 22 hips were lost of follow-up because of the death of the patients. The remaining 37 hips were examined clinically and radiologically. The mean follow-up period of the series was 12.8 years. Complications were examined and clinical evaluation was done using Japanese Orthopaedic Association (JOA) hip score. The criterion for loosening of the acetabular component was cup migration exceeding 3 mm or angular rotation exceeding 3 degrees or breakage of the device. Among the entire series of 67 hips postoperative complications included dislocation in 3 hips, infection in 2 hips and revision in 4 hips. Two hips were revised for loosening and the other two hips were revised for infection. The JOA hip score increased from a mean value of 48.0 preoperatively to 76.8 at the last follow-up. Radiologically 5 hips were loose. Two hips among them were revised. Survival rate of the acetabular component at 10 years was 97.1% using acetabular revision for loosening as the end point and 90.6% using radiological loosening as the end point. Acetabular reconstruction with HA granules and a Kerboull-type acetabular device provided satisfactory clinical and radiographic results at 12.8 post-operative years.

Periprosthetic femoral fractures are becoming increasingly common and are a major complication of total hip arthroplasty (THA) and bipolar hemiarthroplasty (BHA). We report a retrospective review of the outcomes of treatment of 11 periprosthetic fractures after femoral revision using a long stem. Eleven female patients with a mean age of 79.2 years (70 to 91 years) were treated for a Vancouver type B1 fracture between 1998 and 2013. The status of the initial arthroplasty was THA in 5 patients and BHA in 6 patients. The original diagnosis was femoral neck fracture in 5 patients, osteoarthritis in 5 patients, and avascular necrosis of femoral head in 1 patient. Seven patients had had a cemented femoral component and 4 had had a cementless femoral component. The mean numbers of previous surgeries were 3.2 times (2 to 5 times). A previous history of fracture in the same femur was found in 7 hips including 5 femoral neck fractures, 3 periprosthetic fractures. The cause of the latest revision surgery was aseptic loosening in 6 hips, periprosthetic fracture in 3 hips, and infection in 2 hips. The average time to fracture after femoral revision using a long stem was 106.5 months (12 to 240 months). The average follow-up was 58.9 months (8 to 180 months). The fracture pattern was a transverse fracture in 6 hips and an oblique fracture in 5 hips. The type B1 fractures were treated with open reduction and internal fixation in 9 hips, 6 of which were reinforced with bone grafts. Seven patients were treated with a locking compression plate and cerclage wiring, and 2 patients were treated with a Dall-Miles system. Two other periprosthetic fractures were treated with femoral revision. One was revised because of stem breakage, and the other was a transverse fracture associated with very poor bone quality, which received a femoral revision with a long stem and a locking compression plate. All fractures except one achieved primary union. This failed case had a bone defect at the fracture site, and revision surgery using a cementless long stem and allografts was successful. These finding suggest that a type B1 fracture after revision using a long stem associated with very poor bone quality or bone loss might be considered as a type B3 fracture, and femoral revision might be the treatment of choice.