The Oswestry-Bristol Classification of Trochlear Dysplasia: Displays Reliability Only for Normal or Severe Dysplasia in the Skeletally Immature

Abstract

Abstract Purpose: Trochlear dysplasia (TD) is one of the several factors that predispose to recurrent instability and long-term morbidity. Subclassification can aid in risk stratification with surgery and comparing case-cohort outcomes. The inter- and intra-observer agreement of the Oswestry-Bristol Classification (OBC) for TD has previously been demonstrated in adults but not in children. We aim to assess the inter- and intra-observer reliability of the OBC in skeletally immature patients. Methodology: This was a retrospective review of magnetic resonance imaging scans performed in children presenting with patellofemoral instability or recurrent dislocation. A total of 34 scans were graded according to the OBC by seven orthopedic surgeons in two rounds 6 weeks apart. All reviewers were blinded and scans were randomized. The observations from both rounds were compared for inter- and intra-rater reliability. Results: First-round observations showed low–moderate agreement between raters (mean kappa = 0.39). Second-round observations showed moderate agreement (mean kappa = 0.42). However, subanalysis using S statistics found good reliability. There was no statistically significant difference between the two agreement values. Category-wise agreement was excellent for normal trochlea (OBC 1) and moderate to good for severe dysplasia (OBC 4). Reliability was low moderate to poor for mild (OBC 2) or moderate (OBC 3) dysplasia. Intra-observer reliability was good to excellent (mean kappa = 0.73). Conclusion: The OBC is reliable in categorizing a normal or severely dysplastic trochlear in skeletally immature children although it fails to adequately differentiate between mild and moderate dysplasia.