Treatment of Perthes' disease of the hip by cancellous bone grafting. Preliminary report.

It is generally conceded that the conservative treatment of Perthes’ disease of the hip is unsatisfactory. In a recent study of the late results of conservative treatment in fifty-two cases followed up for ten or more years, Evans (1958) showed that one-third developed good femoral heads, one-third fair heads and one-third poor heads, as judged radiologically. Howorth (1948) pointed out that there is a considerable individual variation in all the features of the disease, especially in the radiographic appearances. Not only does this fact make the evaluation of treatment difficult, it makes prognosis impossible. Confronted with an early case the surgeon cannot tell whether the end-result will be a deformed head or a well shaped head, no matter what treatment is given. The findings of Evans and Lloyd-Roberts (1958) suggested that present practice has little effect on the course of the disease. These authors compared the late results of in-patient and out-patient methods of treatment. The in-patients were treated by rest in bed and immobilisation of the hip, and the out-patients with a Snyder sling and crutches. They found no significant difference in the final radiographic appearances in the two groups. Prolonged conservative treatment would be acceptable if it could be shown to be really effective, but, as these reviews indicate, this is not the case. Any method that would hasten regeneration of the femoral head without detriment to the final result would therefore be a real advance. Perhaps some of the poor results of conservative treatment are due to failure to prevent weight bearing for long enough. Brailsford (1953), with a prolonged experience of 1 15 cases, considered that the period of plasticity of the femoral head lasts from three months to four years from the onset of the disease. Howorth (1948) stated that Perthes’ disease runs a characteristic course of four to seven years. The logical conclusion is that the young patient must not be allowed to bear weight on the hip for at least four years. In the past, patients and surgeons alike have baulked at this bleak prospect. For example, the average period of nonweight bearing in Evans and Lloyd-Roberts’s cases was only two years and one month. Before criticising conservative treatment too severely therefore, we should remember that very few patients are treated for what appears to be the optimum time.