Analysis of osteonecrosis following Pemberton acetabuloplasty in developmental dysplasia of the hip: long-term results.

Background: The favorable results of Pemberton acetabuloplasty in children with developmental dysplasia of the hip have been well reported. We reviewed our long-term results related to osteonecrosis of the femoral head after this surgery, especially with regard to the effect of excessive inferior displacement of the femoral head.

Methods: From 1993 to 1997, we performed 167 Pemberton acetabuloplasties in patients with developmental dysplasia of the hip who were eighteen months of age or older. Patients who had had prior treatment or developmental dysplasia of the hip due to neuromuscular disease were excluded. We selected patients who had unilateral developmental dysplasia of the hip, had undergone simultaneous open reduction and Pemberton acetabuloplasty between the ages of eighteen and thirty-six months, and had been followed for a minimum of ten years. Forty-nine patients met these criteria. The patients were divided into osteonecrosis-absent and osteonecrosis-present groups according to the criteria described by Kalamchi and MacEwen. Preoperative, interim follow-up and final radiographs were available for evaluation, as were the results of clinical examination. We used the femoral head inferior displacement percentage, measured on the radiographs, to quantify the amount of excessive correction postoperatively. Outcomes were measured with use of the McKay criteria and the Severin criteria.

Results: The mean age at the time of surgery was 20.8 months, and the mean duration of follow-up was 134.6 months. Twenty-four patients (49%) were classified as not having osteonecrosis (the osteonecrosis-absent group) and twenty-five patients (51%), as having osteonecrosis (the osteonecrosis-present group). There were no significant differences between the two groups in terms of sex, age, laterality, Tönnis grade, or preoperative acetabular index. Seven of the cases of osteonecrosis were type I, thirteen were type II, one was type III, and four were type IV. The inferior displacement percentage revealed significant differences between the two groups (p < 0.0001). In the osteonecrosis-absent group, 96% of the patients had a radiographically satisfactory result (Severin class I or II); however, only 76% of the patients in the osteonecrosis-present group had a radiographically satisfactory result (p < 0.0001). According to the McKay criteria, there were significant clinical differences between the groups (p < 0.0001).

Conclusions: Our results showed significant correlation between excessive reduction of the femoral head and the development of osteonecrosis. In light of the high prevalence of type-II osteonecrosis, we postulated that the lateral epiphyseal branch of the medial circumflex artery was vulnerable to compression with increased inferior displacement of the femoral head. The latest radiographic and functional results corresponded to the severity of the osteonecrosis.