Long-term follow-up of Perthes' disease treated nonoperatively.

Forty-six patients with Perthes disease treated nonoperatively were reviewed 10 to 43 years later. The median follow-up was 17 years. The patients were analyzed according to sex, age of onset, quantity of head involvement, type of treatment, duration of treatment, functional end result based on the Harris hip scale, and anatomic end result based on Mose templete evaluation. Most patients were treated with an ischial weight-bearing brace. The anatomic results of this treatment tended to be poor, probably worse than no treatment at all. In general, poor results were correlated with late onset, greater quantity of head involvement, and lateral subluxation of the femoral head. As reported in other series, the anatomic results did not correlate well with the functional results, only two patients having Harris hip scales below 70.