From 1986 to 1991 we fitted 20 children with endoprostheses after resection of malignant bone tumours of the leg; six have reached skeletal maturity and are the subject of this study. Reconstruction of defects in growing limbs in which the eventual shortening can be predicted requires the use of extendable prostheses. The mean age at operation was 11 years (9.2 to 13.7) and the average follow-up period was 6.3 years (4.3 to 7.6). The diagnosis was osteosarcoma in five patients and Ewing's sarcoma in one. All tumours were Enneking stage-IIB. When seen for follow-up all patients were free from disease. The extendable implants used included the Pafford-Lewis prosthesis and the Kotz Modular Femur Tibia Reconstruction system with a compatible, newly-designed growth module. Telescope-like elongation of the prostheses was performed by insertion of a screwdriver through a small skin incision. Active epiphyseal growth in the adjacent growth plate was preserved by using prosthetic stems with a smooth surface. The mean length gained was 13.15 cm (4.5 to 19.5) requiring 53 planned procedures. Seven revision operations were necessary for complications. Functional evaluation showed excellent and good results in all cases. Stress-shielding at the site of anchorage of the prosthesis was more pronounced than in adults. Implantation of extendable endoprostheses in children provides a reasonable alternative to rotationplasty, but limb salvage requires more operations.

A study is presented of the aetiology and results of treatment in a group of 125 proven osteosarcomas present in children under fifteen years of age. These cases have been collected from the records of one English and six European treatment centres. There is a slight male preponderance, but the striking aetiological feature is the very high proportion of tumours of the long bones of the limbs (96 per cent).

The two and a half and five year disease-free survival rates were respectively 15 and 12 per cent, with a further 9 per cent still living, but under observation for less than two and a half years. Evidence of metastasis after two and a half years is very unusual, but no child with a tumour of an axial or girdle bone lived this length of time.

Although the differences in the results of the different methods of treatment employed are not statistically valid, the largest number of long survivors had been treated by early amputation, which method also provided the lowest rate of local tumour recurrences. Reasons are discussed which indicate that prompt ablation is the treatment of choice, perhaps with certain advantages in the light of recent advances in adjuvant treatment. The past situation in connection with childhood osteosarcoma certainly provides strong support for immediate carefully designed clinical trials of the new adjuvant methods cited.