Risk factors for acetabular retroversion in developmental dysplasia of the hip: does the Pemberton osteotomy contribute?

Abstract

The purpose of this study was to investigate residual acetabular retroversion after skeletal maturity in patients with Pemberton osteotomy. We compared 40 hips in 36 patients treated with a Pemberton osteotomy (Pemberton group) and 30 hips in 26 patients treated only with a Pavlik harness (Rb group) for developmental dysplasia of the hip. The average age at operation in the Pemberton group was 94.5 months and the follow-up duration was 151.8 months. Radiographic parameters included the acetabular index (α angle) and the center-edge angle of Wiberg, preoperatively and at skeletal maturity. We examined the crossover sign (COS) at the latest follow-up as a sign of acetabular retroversion (AR). We compared the parameters between the two groups and examined the risk factors for acetabular retroversion using a multivariate Cox model. A COS (+) was significantly more frequent in the Pemberton group compared to the Rb group [15 hips (37.5%) vs 3 hips (10%); p = 0.0077]. In the Pemberton group, the average age at operation in COS (+) hips was significantly older than that in COS (-) hips (126.9 vs 72.8 months; p = 0.0005). The preoperative α angle did not vary between hips with and without COS; however, the postoperative α angle was significantly smaller in COS (+) hips. A multiple logistic regression analysis for prediction of COS (+) showed that the age at operation and the amount change of α angle were significant predictors for COS (+) hips. The cut-off of the age at operation was 7 years and 9 months old. AR was present in 37.5% of the hips in the Pemberton group after skeletal maturity. Remodeling of acetabular version was observed in younger patients; however, hips in older patients (>8 years) at the time of operation and greater degrees of correction tended to result in AR.