Evaluating the true clinical utility of the red dot system in radiograph interpretation

Introduction: The 'red dot', or 'asterisk', system is used in many hospitals as a method for radiographers to identify potential abnormalities on plain radiographs prior to reporting by radiologists. While published studies into the accuracy of the 'red dot' system exist, analysis of its reliability in identifying subtle pathology is lacking. This is relevant because the prevalence and apparent success of the 'red dot' system has been cited by some authors as justification for non-radiologist reporting of medical imaging. It is important that all systems within medical imaging add value, and this audit evaluates the accuracy and clinical value of the 'red dot' system, particularly in the critical area of detecting undisplaced fractures. Methods: All appendicular musculoskeletal trauma radiographs performed in the Department of Emergency Medicine at a major Australian metropolitan hospital over a continuous four-month period were retrospectively assessed to evaluate the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of the 'red dot' system. The presence or absence of an asterisk was correlated with validated radiologist reports. A sub-analysis of fractures displaced <1mm was also performed. Results: A total of 3638 radiographs were assessed and 938 (25.8%) fractures identified; 338 (9.3%) contained undisplaced fractures. Overall sensitivity and specificity of the 'red dot' in detecting appendicular fractures are 80.4% and 98.0%, respectively. PPV is 93.6% and NPV is 93.5%. However, the accuracy of the 'red dot' in detecting undisplaced fractures is significantly reduced, with a sensitivity of 45.9% and a PPV of 74.8%. Conclusions: Detection of subtle abnormalities is fundamental to the service provided by radiologists. The 'red dot's' inability to reliably detect undisplaced fractures following trauma limits its value within a tertiary radiology department and suggests that role extension of plain film reporting to non-radiologists has potential to yield less accurate assessments. To maintain the highest quality of medical imaging services and standards of patient care, it is optimal that clinical decisions are based upon radiograph reports issued by medical specialists who have completed appropriate radiology training. Strategies to maintain this are suggested.