Avascular necrosis (AVN) of the femoral head is a potentially devastating complication of treatment for developmental dysplasia of the hip (DDH). AVN most commonly occurs following operative management by closed (CR) or open reduction (OR). This occurrence has frequently been examined in single centre, retrospective studies, however, little high-level evidence exists to provide insight on potential risk factors. The purpose of this observational, prospective multi-centre study was to identify predictors of AVN following operatively-managed DDH. A multi-centre, prospective database of infants diagnosed with DDH from 0–18 months was analyzed for patients treated by CR and/or OR. At minimum one year follow-up, the incidence of AVN (Salter criteria) was determined from AP pelvis radiographs via blinded assessment and consensus discussion between three senior paediatric orthopaedic surgeons. Patient demographics, clinical exam findings and radiographic data were assessed for potential predictors of AVN. A total of 139 hips in 125 patients (102 female, 23 male) underwent CR/OR at a median age of 10.4 months (range 0.7–27.9). AVN was identified in 37 cases (26.6% incidence) at a median 23 months post-surgery. Univariate logistic regression analysis comparing AVN and no AVN groups identified sex, age at diagnosis, age at surgery, pre-surgery IHDI grade and time between diagnosis and surgery as potential predictive factors. Specifically, male sex (OR 2.21 [0.87,5.72]), IHDI grade IV, and older age at diagnosis (7.4 vs. 9.5 months) and surgery (10.2 vs. 13.6 months) were associated with development of AVN. Likewise, increased time between diagnosis and surgery (2.9 vs. 5.5 months) was also associated with a higher incidence. No association was found with surgery type (CR vs. OR), pre-surgery acetabular index or surgical hip. Development of AVN occurred in 26.6% of hips undergoing CR or OR at a median 23 months post-surgery. Male sex, older age at diagnosis and surgery, dislocation severity and increased time between diagnosis and surgery were associated with AVN. Longer-term follow-up and larger numbers will be required to confirm these findings. Early outcomes from this prospective patient cohort suggest that AVN is an important complication of operative management for DDH, and appears to occur at a comparable rate whether the reduction is performed open or closed. Male patients may be more susceptible to developing AVN and merits further exploration. Potential predictive factors of older age and length of time between diagnosis and surgery emphasize the importance of early detection and treatment to minimize complications and optimize outcomes

Purpose. Following closed or open reduction for developmental dysplasia of the hip (DDH), assessment of reduction is essential. With potentially poor accuracy in confirming reduction, the risk of abnormal hip development and ultimately poor outcome exists if reduction is not achieved. Computed tomography (CT) has been used in recent years to assess reduction. The aim of this study was to compare the accuracy in confirming hip reduction following closed or open reduction in children with DDH, using CT and plain radiographs and to decide whether CT scans improved the assessment of reduction. Methods. We retrospectively reviewed 6 patients treated for DDH at an Academic Hospital. The patients were treated with either closed or open reduction. Post operatively radiographs and CT scans were obtained to assess reduction. Reduction was assessed using Shenton's line, medial joint space, a femoral mid-cervical line through the tri-radiate cartilage and a tri-radiate intersectional line on axial CT. The CT scans were analyzed using Osirix on an Apple Macintosh computer. Results. We were able to obtain measurements in all parameters in only one radiograph, whereas in the CT images all parameters were determined. One patient had radiographs in which no values could be measured. The CT scan however confirmed reduction of that hip. In one patient we assessed a hip as reduced using all the parameters, whereas the tri-radiate intersectional line on axial CT and the Shenton's line on the coronal slice showed that in fact the head was posteriorly subluxed. Conclusion. We conclude that although CT scans did not change our management in this small patient series, CT scans did provide a better means of confirming hip reduction than plain radiographs for patients with DDH following closed or open reduction in a hip spica. NO DISCLOSURES