Aim: To evaluate digital chest radiography (CR) performance compared to computed tomography (CT) for characterising small low-risk pulmonary nodules detected incidentally in non-oncological patients. A second aim was to assess the prevalence of calcification and possible false-positive findings mimicking nodules.
Materials and methods: Two hundred and seven patients who presented with a pulmonary nodule on CR and underwent CT were included prospectively. Nine radiologists blinded to the CT images reviewed the CRs assessing for the presence of nodules. Afterwards, the same radiologists evaluated the corresponding CT for the presence of nodules, dimensions, and calcification. If the nodule was not present on CT, it was considered a false-positive finding, and possible confounding factors on CR were investigated.
Results: Among all 213 nodules seen on CR, 32.4% were revealed to be false-positive findings on CT, mostly due to images formed by vessels (53.6%), osseous aetiologies (30.4%), and skin lesions (13%). Most nodules <6 mm detected on radiographies had benign calcification on CT (n=90; 67.7%). Comparatively, only 41.2% of nodules ≥6 mm on the CR had benign calcification. Among all nodules <6 mm detected on CR, 95.5% were calcified or not present at CT against 81.2% for those ≥6 mm (p<0.001).
Conclusion: The present study demonstrated that 95.5% of pulmonary nodules smaller than 6 mm on CRs are either calcified (benign) or represent a false-positive finding on CT. These results suggest that nodule measures on CR smaller than 6 mm most likely represent a benign finding.
Copyright © 2018 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.