Background. Several risk factors for the development of osteonecrosis following treatment of developmental dislocated hip have been reported. The need for further research with a large-enough sample size including statistical adjustment of confounders was demanded. The purpose of the present study was to find reliable predictors of osteonecrosis in patients managed for developmental dislocation of the hip. Methods. A retrospective cohort study of children, who have been hospitalized at our department between January 1998 and February 2007 with a developmental dislocation of the hip, was completed. Sixty-four patients satisfied the criteria for inclusion. Three groups according to age and treatment were identified. Group A and B included patients treated with closed or open reductions aged less than twelve months. Patients of group C were past walking age at the time of reduction and were treated by open reduction combined with concomitant pelvic and femoral osteotomies. The average duration of follow-up for all patients was 6.8 years. Logistic regression analysis was conducted to identify predictors for the development of osteonecrosis. Results. The overall rate of osteonecrosis in group A and B was 27.4% compared to 88.2% in patients of group C. After pooling of all data, no protective effect of the ossific nucleus of the femoral head on the development of osteonecrosis was found (p = 0.14). Additionally, an increase of surgical procedures in children of group C could not be demonstrated (p = 0.17). By using logistic regression analysis the type of reduction and secondary reconstructive procedure due to
