Abstract

Background: Radiographs are vital diagnostic tools that complement physical examination in trauma patients. A study was undertaken to assess the performance of residents in emergency medicine in the interpretation of trauma radiographs.

Methods: 348 radiographs of 100 trauma patients admitted between 1 March and 1 May 2007 were evaluated prospectively. These consisted of 93 cervical spine (C-spine) radiographs, 98 chest radiographs, 94 radiographs of the pelvis and 63 computed tomographic (CT) scans. All radiological material was evaluated separately by five emergency medicine residents and a radiology resident who had completed the first 3 years of training. The same radiographs were then evaluated by a radiologist whose opinion was considered to be the gold standard. The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were calculated.

Results: The mean (SE) age of the patients was 29 (2) years (range 2–79). There were no statistically significant differences in terms of pathology detection between the emergency medicine residents and the radiologist. The agreement between the emergency medicine residents and the radiology resident was excellent for radiographs of the pelvis and the lung (kappa (κ) = 0.928 and 0.863, respectively; p<0.001) and good for C-spine radiographs and CT scans (κ = 0.789 and 0.773, respectively; p<0.001).

Conclusions: Accurate interpretation of radiographs by emergency medicine residents who perform the initial radiological and therapeutic interventions on trauma patients is of vital importance. The performance of our residents was found to be satisfactory in this regard.