Abstract
Introduction
The prevalence of adverse reactions to metal debris (ARMD) associated with metal on metal (MoM) hip arthroplasty has been reported to be as high as 69%. Such findings promoted the development of metal-artefact reducing sequence (MARS)-magnetic resonance imaging (MRI) classifications, with the aim of stratifying soft lesions by severity of disease. The Modified Oxford Classification is a straightforward system that has been shown to correlate with disease progression.
Methods
The aim of this study was to test the reliability of this classification between observers. Seven observers were recruited, all with a musculoskeletal background. Using the PACS image analysis system, 20 MARS-MRI scans were provided for interpretation. Observers reviewed these scans in random order at two separate intervals over the course of five weeks. They classified them according to the Modified Oxford Classification as: ‘normal’, ‘trochanteric fluid, ‘effusion’, ‘ARMD type 1’, ‘ARMD type 2’ and ‘ARMD type 3’.
Results
Statistical tests using kappa co-efficients were performed to get intra and inter observer reliabilities. Our results were compared against the grading system proposed by Landis and Koch. Our inter-rater agreement was 0.61. This is considered substantial agreement. Our intra observer scores ranged for 0.57 to 1.00 indicating consistency between the observers as individuals and as a group.
Conclusion
The modified oxford classification is simple and easy to remember and has substantial reliability. It is proving to be effective in communicating, classifying and providing an understanding of disease progression.