Heck Robert K, O'Malley Aran M, Kellum Ethan L, Donovan Timothy B, Ellzey Andrew, Witte Dexter A
University of Tennessee Campbell Clinic, Department of Orthopaedic Surgery, Memphis, TN 38104, USA.
Clin Orthop Relat Res. 2007 Jun;459:28-33. doi: 10.1097/BLO.0b013e3180485681.
Interpretation of an MRI of a suspected musculoskeletal neoplasm can be extremely difficult. Fifty-six MRIs originally evaluated by outside radiologists were independently evaluated by an expert panel consisting of three specialized musculoskeletal radiologists. The outside reports were then graded based upon accuracy and completeness of the differential diagnosis. We compared the expert opinions with those of the outside radiologists. According to the expert panel, only 30 of the 56 (54%) outside reports listed the most likely diagnosis as such and only 35 (63%) listed it at all. A complete appropriate differential diagnosis was listed in only 22 (39%) of the outside reports. Furthermore, 18 (32%) of the outside reports listed diagnoses judged to be extremely unlikely by the experts. In a subset of 15 patients with images that the expert panel had judged diagnostic of specific entities, only nine of the outside reports listed the correct diagnosis as such and only 10 listed it at all. Furthermore, 11 (73%) of the outside reports listed extremely unlikely possibilities for these diagnostic images. We found a substantial difference between the expert and the outside opinions.
