Radiographic Outcome Following Treatment of Residual Hip Dysplasia with Pemberton Versus Salter Osteotomy: Comparison of Results in Patients Followed to Skeletal Maturity

Background:Salter osteotomy (SO) and Pemberton acetabuloplasty (PA) are procedures to treat skeletally immature patients with developmental dysplasia of the hip (DDH). The purpose of this study was to compare the radiographic results and rate of residual dysplasia (RD) after treatment with SO and with PA. Methods:This was a retrospective analysis of pediatric patients treated with either SO or PA for DDH between 1980 and 2013 who were skeletally mature at the time of follow-up. The preoperative and postoperative acetabular index (AI) and postoperative lateral center-edge angle (LCEA) and anterior center-edge angle (ACEA) were collected. RD was defined as an LCEA or ACEA of <20 degrees or an acetabular inclination (AIn) of >15 degrees at skeletal maturity. Continuous variables were compared between treatment groups with 2-sample t tests, and categorical variables were compared using chi-square tests. Multivariable analysis was used to identify risk factors for RD with p < 0.05. Results:The study included 113 hips in 102 patients with a mean follow-up of 10.5 years. The SO group had 73 hips treated at a mean patient age of 5.4 years, and the PA group had 40 hips treated at a mean age of 5.3 years. The preoperative AI did not differ between the treatment groups (32.2 degrees versus 31.3 degrees, p = 0.658), nor did the immediate postoperative AI (20.0 degrees versus 20.2 degrees, p = 0.459). At the time of final follow-up, the SO group had a significantly smaller AIn (8.6 degrees versus 13.2 degrees, p = 0.001), a trend toward a greater LCEA (26.2 degrees versus 21.6 degrees, p = 0.056), and a similar ACEA (24.1 degrees versus 26.1 degrees, p = 0.808). By the abovementioned definition, 36 hips (31.9%) had evidence of RD, with a lower rate in the SO group (26.0% versus 42.5%, p = 0.07). Patients who had a positive family history of hip dysplasia were more likely to have RD (odds ratio = 4.311, 95% confidence interval = 1.125 to 16.528). Conclusions:Patients with RD could be effectively treated with either SO or PA, with overall good radiographic outcomes. However, SO achieved a better AI and may thus yield better long-term health of the affected hip.