Abstract
Study objective:To evaluate the diagnostic performance of chest ultralow-dose computed tomography (CT) compared with chestradiograph for minor blunt thoracic trauma.Methods:One hundred sixty patients with minor blunt thoracic trauma were evaluatedfirst by chest radiograph and subsequentlywith a double-acquisition nonenhanced chest CT protocol: reference CT and ultralow-dose CT with iterative reconstruction. Twostudy radiologists independently assessed injuries with a structured report and subjective image quality and calculated certaintyof diagnostic confidence level.Results:Ultralow-dose CT had a sensitivity and specificity of 100% compared with reference CT in the detection of injuries (187lesions) in 104 patients. Chest radiograph detected abnormalities in 82 patients (79% of the population), with lower sensitivityand specificity compared with ultralow-dose CT (P<.05). Despite an only fair interobserver agreement for ultralow-dose CT imagequality (k¼0.26), the diagnostic confidence level was certain for 95.6% of patients (chest radiograph¼79.3%). Ultralow-dose CTeffective dose (0.203 mSv [SD 0.029 mSv]) was similar (P¼.14) to that of chest radiograph (0.175 mSv [SD 0.155 mSv]) andsignificantly less (P<.001) than that of reference CT (1.193 mSv [SD 0.459 mSv]).Conclusion:Ultralow-dose CT with iterative reconstruction conveyed a radiation dose similar to that of chest radiograph and wasmore reliable than a radiographic study for minor blunt thoracic trauma assessment. Radiologists, regardless of experience withultralow-dose CT, were more confident with chest ultralow-dose CT than chest radiograph.