A proposed prognostic formula for Perthes' disease

We retrospectively reviewed 145 patients with unilateral Perthes' disease and compared a quantitative analysis of early radiographic signs with a predicted prognosis at long-term followup. The average age of the patients at followup was 18.7 years (range, 16.2-27.5 years). We used the age at onset and three radiographic factors as independent variables for multiple regression analysis. The final radiographic results were based on a modified Stulberg's classification as the dependent variable. Fifty patients (35%) had a good outcome, 33 patients (23%) had a fair outcome, and 62 patients (42%) had a poor outcome. The most reliable formula in the stepwise multiple regression analysis was calculated as: y = -0.697 + 0.418 (age score) + 0.860 (involvement score) + 0.248 (subluxation score). The radiographic stage at first visit had no influence on the final results. Multifactorial assessment by combination of age at onset and two radiographic factors (epiphyseal involvement and subluxation) was the most reliable for predicting the prognosis. A score of 1.5 points or less predicted a good prognosis and a score of 2.6 or more indicated a poor prognosis.

Prognostic study, Level IV (case series). See the Guidelines for Authors for a complete description of levels of evidence.