Clinical contact with the trauma patient enhances accuracy of chest roentgenogram interpretation in predicting traumatic rupture of the aorta.

In a previous retrospective clinical study, we reported that clinicians having contact with the trauma patient interpreted widened mediastinum (WMED) in association with TRA with greater accuracy than did staff radiologists reviewing the same chest films (0.96 vs 0.83), while staff radiologists interpreted WMED with far greater frequency than did their clinical counterparts (0.40 vs. 0.27, p = 0.01). These findings may be related to differences in chest film interpretation or to the influence of clinical contact with the patient. To answer this question, data from a study in which a group of four surgeons and two radiologists was asked individually and in a blinded fashion to interpret, without the clinical history, the chest x-rays of 149 trauma patients who had undergone aortography to rule out TRA were examined by specialty affiliation. Sixteen of the patients had TRA. Surgeons interpreted WMED with the same frequency as radiologists (0.47 vs 0.54) and with far greater frequency (0.47 vs 0.27, p = 0.06) and poorer accuracy (0.61 vs 0.96) than they did in the previous clinical review. No significant differences were detected between surgeons and radiologists in the overall rate of interpretation of WMED, in the prediction of aortic rupture, or in the ordering of aortography based purely upon chest film interpretation. We conclude that trauma surgeons and radiologists, isolated from the patient interpret the chest film signs associated with TRA similarly. Clinical contact with the trauma patient may modify what is seen on chest film in a way that contributes to greater accuracy in the diagnosis of TRA. Radiographic criteria for ordering aortography are thus best used in conjunction with careful clinical evaluation of the patient.