Intraoperative Arthrogram Predicts Residual Dysplasia after Successful Closed Reduction of DDH.

OBJECTIVE: To determine the incidence of residual dysplasia after closed reduction (CR) of developmental dysplasia of the hip (DDH) and assess correlations between quality of arthrogram-guided CR and residual dysplasia using a new intraoperative radiographic criterion.

METHODS: Data of a consecutive series of 126 patients with DDH in 139 hips treated at our institution by arthrogram-guided CR from March 2006 to June 2013 were reviewed in this retrospective study. There were 23 boys and 103 girls with 88 affected left hips and 51 right hips. The average age at closed reduction was 14 months (range, 7-19 months) and average duration of follow-up 36 months (range, 24-100 months). Femoral head coverage (FHC) and arthrography type (A/B/C) on best reduced arthrographic images, acetabular index (AI) and Wiberg Center-Edge (CE) angle on anteroposterior (AP) pelvis radiograph at latest follow-up were measured. Residual hip dysplasia was determined according to the Harcke acetabular dysplasia radiographic standard. Patients were divided into non-late acetabular dysplasia (non-LACD) and late acetabular dysplasia (LACD) groups according to final results and age at reduction, sex and side compared between these two groups. Correlations between FHC and arthrography type and residual hip dysplasia were analyzed. Multiple logistic regression analysis was used to analyze sex, AI at CR, arthrography type and FHC with LACD. Receiver operating characteristic (ROC) curve analysis was used to determine the cutoff value for FHC.

RESULTS: Forty-five of 139 hips (32.4%) had residual hip dysplasia. Avascular necrosis of the femoral head occurred in 11 hips (7.9%), nine of which had acetabular dysplasia. There were no significant differences between the two groups in age at reduction, sex or side. FHC differed significantly between the two groups (51.2% ± 15.3% vs . 28.5% ± 15.9%, t = 4.718, P = 0.000). A significantly greater percentage of the arthrography Type C group than Type A and B groups had LACD (χ(2) = 17.170, P = 0.017). According to multiple logistic regression analysis, FHC was the only prognostic factor for LACD. There was a clear cutoff value for FHC (30%), under which 81.8% hips were determined as having LACD according to ROC curve analysis.

CONCLUSIONS: Intraoperative arthrogram-determined FHC is an alternative predictor of residual hip dysplasia after CR of DDH and FHC ≤30% can be considered the criterion for unacceptable reduction.