We examined the concordance between chest ultrasonography and chest radiography in patients with dyspnea, using chest CT scanning as the gold standard in case of mismatch between the two modalities.
A prospective, blinded, observational study was conducted in the ED of a university-affiliated teaching hospital. All consecutive patients presenting for dyspnea during a single emergency physician shift were enrolled independently from the underlying disease. Only patients with trauma were excluded.
Both ultrasonography and radiography were performed in 404 patients; CT scanning was performed in 118 patients. Ultrasound interpretation was completed during the scan, whereas the average time between radiograph request and its final interpretation was 1 h and 35 min. Ultrasonography and radiography exhibited high concordance in most pulmonary diseases, especially in pulmonary edema (κ = 95%). For lung abnormalities such as free pleural effusion, loculated pleural effusion, pneumothorax, and lung consolidation, the concordance was similar for both left- and right-side lungs (all P not significant). When ultrasound scans and radiographs gave discordant results, CT scans confirmed the ultrasound findings in 63% of patients (P < .0001). Particularly, ultrasonography exhibited greater sensitivity than radiography in patients with free pleural effusion (P < .0001).
When performed by one highly trained physician, our study demonstrated high concordance between ultrasonography and radiography. When ultrasound scans and radiographs disagreed, ultrasonography proved to be more accurate in distinguishing free pleural effusion. Thus, considering the short time needed to have a final ultrasound report, this technique could become the routine imaging modality for patients with dyspnea presenting to the ED.