MR-IMAGING OF THE ROTATOR CUFF TENDON - INTEROBSERVER AGREEMENT AND ANALYSIS OF INTERPRETIVE ERRORS

PURPOSE: To evaluate accuracy in interpretation of shoulder magnetic r esonance (MR) images and interobserver agreement and to characterize t he types of errors. MATERIALS AND METHODS: Five radiologists with vary ing experience independently and retrospectively twice interpreted the MR images of 222 symptomatic patients who underwent bath MR imaging a nd shoulder arthroscopy. The first interpretation was a blind review; the second was with knowledge of the arthroscopic findings. RESULTS: F or full-thickness tears, the sensitivity, specificity, and accuracy we re 84%-96%, 94%-98%, and 92%-97%, respectively, and for partial tears, 35%-44%, 85%-97%, and 77%-87%, respectively. There was no statistical ly significant difference between readers in the detection of partial or full-thickness tears. There was a statistically significant differe nce between readers in the no-tear category. Cohen kappa values genera lly indicated improved interobserver agreement proportional to the rea ders' experience (full-thickness tears, 0.731-0.881; partial tears, 0. 168-0.443). CONCLUSION: Diagnosis of a full-thickness tear can be lear ned to a high degree of accuracy. Despite the radiologist's level of e xperience and knowledge of the arthroscopic findings, the sensitivity for diagnosis of partial tears is poor.