Reliability of the Ahlback classification of knee osteoarthritis

Objective: The aim of the study was to assess the inter-observer reliability and intra-observer reproducibility of the Ahlback radiographic classification for knee osteoarthritis (OA). Design: Ninety-six knee radiographs of patients with clinical signs of arthritis were evaluated and classified into five grades of Ahlback classification by three observers with different experience. The evaluation was repeated after 1 month. The inter- and intra-rater agreements were assessed by statistical analysis (k and k(w) coefficients). Results: The inter-observer agreement was statistically significant, but with low or medium values of the coefficients. The intra-rater agreement was less significant than the inter-rater. Inter- and intra-rater agreement coefficients decreased when the cases of stage V arthritis were excluded from the analysis. The evaluations were influenced by the age and working experiences of each observer. Conclusions: The authors stress that the Ahlback classification, as routinely applied, should not be used in orthopaedic surgery without the support of clinical and/or arthroscopic examinations, because of its poor reliability. (C) 2003 OsteoArthritis Research Society International. Published by Elsevier Ltd. All rights reserved.