Stephen R. Cannon

We treated 35 patients with primary malignant tumours of the periacetabular area by resection and prosthetic reconstruction of the defect. At a mean follow-up of 84 months, 15 patients (43%) were free from disease. The most common complications were deep infection (26%), local recurrence (24%) and recurrent dislocation of the hip (17%). The surviving patients achieved an average of 70% of their premorbid function. This method of reconstruction has a high morbidity and should be performed only at specialist centres, but the functional and oncological outcomes are satisfactory.

The use of extendible distal femoral replacements is a relatively new treatment alternative for malignant bone tumours in growing individuals. Although their appearance was widely appreciated, questions about functional practicality and longevity remain unclear. With longer follow-up, advantages of immediate functional restoration and beneficial psychological aspects seem to be overshadowed by an increase in complications such as aseptic loosening, infection or prosthetic failure. We have reviewed 18 children with such tumours who were treated between 1983 and 1990 by custom-made Stanmore extendible distal femoral replacements. Four died from metastatic disease within 2.5 years of operation and two required amputation for local recurrence or chronic infection. The remaining 12 patients were followed for a mean of 8.7 years (6 to 13.2). A mean total lengthening of 5.2 cm was achieved, requiring, on average, 4.3 operations. Using the Musculoskeletal Tumor Society rating score the functional result at review was, on average, 77% of the expected normal function, with seven patients achieving > or = 80%. Revision of the prosthesis was required in ten patients, in six for aseptic loosening, at a mean of 6.2 years after the initial procedure.

Although rarely required, extendable reconstruction devices for replacing the entire femur offer children with malignant bone tumors the opportunity of a nearly normal development by overcoming an expected leg length discrepancy. Femoral integrity can be restored, allowing most patients to walk without the use of aids. There are no data available to provide evidence regarding long term results and morbidity in such patients. Six patients (range 2-12 years of age), three with osteogenic osteosarcoma and three with Ewing's sarcoma, were treated between 1988 and 1996 with custom made Stanmore extendable prosthetic total femoral replacements. One patient died 12 months after surgery because of complications relating to pulmonary metastasis. The remaining five patients were observed between 2.7 and 8.9 years (average, 5 years). No tumor recurrence has been recorded and no amputation has been performed. All surviving patients underwent an average of 9.4 operative procedures (range, 4-16 procedures) including 6.4 extension procedures (range, 3-10 procedures), and one prosthetic revision (range, 0-3 procedures). Five revisions in two patients were necessary because of infection, loosening of the prosthesis, mismatch between femoral head and acetabulum, or full extension of the extending mechanism. The functional results were measured in accordance with the Musculoskeletal Tumor Society rating score, with an average result of 77.3%. Total femoral replacement in a growing individual achieves good functional results yet has various risks for an uncertain outcome. Careful selection of the patient and realistic appraisal of the long term prospects are essential for successful treatment.