Limited intraobserver and interobserver reliability for the common measures of hip joint congruency used in dysplasia

Background: A congruent hip frequently is cited in the literature as a prerequisite for performing a reconstructive pelvic osteotomy for hip dysplasia. The designation as congruent is important as incongruent joints generally are regarded as requiring a salvage-type procedure. Several measures of congruency are described in the literature, but it is unclear whether these measures are reliable. Questions/purposes: We therefore evaluated the intraobserver and interobserver reliabilities of three measures of hip congruency in patients with mild to severe dysplasia. Methods: We selected radiographs of 30 skeletally mature patients with symptomatic hip dysplasia who were potential candidates for a periacetabular osteotomy. All radiographs were rated by six reviewers on two occasions. Reviewers were shown a series of AP and von Rosen views of either the symptomatic hip or the contralateral side and asked to rate congruency using two previously published measures - the classifications of Yasunaga et al. and Okano et al. The reviewers also were asked to determine whether the hip was congruent based on their subjective opinion (yes/no). The radiographs were reviewed and measurements were repeated 1 month later. Kappa analysis was performed to determine intraobserver and interrater reliabilities between and among individual raters, respectively. Results: Using the classifications of Yasunaga et al. and Okano et al., we found intraobserver reliability kappa values of 0.43 and 0.37, respectively and interobserver reliability values of 0.18 and 0.25, respectively. The intraobserver reliability of a subjective opinion of yes/no response was 0.74 while the interobserver reliability was 0.21. Conclusions: When evaluating a spectrum of hip dysplasia, commonly used measures of hip congruency have low intraobserver and interobserver relclassifications of Yasunaga et al. and Okano et al. Interobserver reliability for all three methods was low. New radiographic parameters to consistently identify congruency will be helpful for evaluating the preoperative indications for reconstructive osteotomy. Clinical Relevance: Hip congruency has long been cited as a requirement for a reconstructive pelvic osteotomy, but based on the above findings, there is no clear agreement regarding what comprises a congruent joint. Thus, better criteria are needed to assess preoperative hip congruency to determine the role of congruency in patient outcomes after surgery for hip dysplasia.