Ji Hyun Koo MD, Abstract Co-Author: Nothing to Disclose

Myung Jin Chung MD, Abstract Co-Author: Advisory Board, Samsung Advanced Institute of Technology Research Consultant, Samsung Mobile Display Company

Boram Choi, Presenter: Nothing to Disclose

Hyun Jung Yoon MD, Abstract Co-Author: Nothing to Disclose

John M. Sabol PhD, Abstract Co-Author: Employee, General Electric Company

Kyung Soo Lee MD, PhD, Abstract Co-Author: Nothing to Disclose

Hye Na Jung MD, Abstract Co-Author: Nothing to Disclose

The importance of small pulmonary nodules detected on chest radiography (XR) and CT varies substantially according to their size and characterization. Thus RECIST 1.1 recommended that plain radiography should be used for lung nodules larger than 2 cm in diameter. Recent advances in volumetric X-ray digital tomosynthesis (DT) may enable the detection of small lung nodules. With this potential it’s imperative to understand nodule measurement accuracy and its implications on nodule follow-up and management. This accuracy was proved in phantom study but not yet in clinical condition. Thus the purpose of this study was to evaluate the accuracy of measuring the size of each nodule with both XR and DT compared to CT in real clinical conditions.

61 lung nodules (diameter 4 – 20 mm on CT) were selected from 51 patients who had chest CT, XR, and DT within a week. We assessed characteristics (location, marginal sharpness) and coronal diameters (longest diameter on the plane with maximum nodule size) of each nodule on coronal CT as a reference. Two readers independently measured each nodule on XR and DT. We analyzed overall and individual performance and variability.

The mean diameters of the nodules were 12.3 ± 5.0 mm, 12.5 ± 5.1 mm, and 13.3 ± 5.5 mm, on coronal CT, DT, and XR, respectively. Inter-observer correlation coefficients were .95, .92, and .79 on CT, DT, and XR, respectively. DT (R = .97, P < .01) was correlated better than XR (R = .79, P < .01) with CT . Mean difference of diameter (ΔD) between DT and CT was 0.1 ± 1.2 mm (P = .20, paired T-test)). ΔD of XR was 0.7 ± 3.0 mm (P < .01). For anterior nodules, ΔD were not significant in both DT (0.2 ± 1.4 mm, P = .40) and XR (0.3 ± 2.3 mm, P = .46). For posterior nodules, ΔD was not significant in DT (0.8 ± 1.1 mm, P = .92) but significant in XR (2.0 ± 1.5 mm, P < .01). Variations of difference were not different between well (0.0 ± 1.1 mm) and poorly marginated nodules (0.3 ± 1.4 mm) on DT (P = .18, Levene test), or between left (0.3 ± 1.2 mm ) and right (0.0 ± 1.2 mm) lungs on DT (P = .39).

DT was well correlated and not significantly different than CT for lung nodule size measurement in clinical conditions. DT is comparable to CT for accuracy of lung noulde size.

With digital tomosynthesis, the size of small lung nodules can be accurately determined, which may improve patient management and follow-up.

Koo, J, Chung, M, Choi, B, Yoon, H, Sabol, J, Lee, K, Jung, H, Small Lung Nodule Size Measurement with Digital Tomosynthesis: Accuracy in Clinical Conditions.  Radiological Society of North America 2010 Scientific Assembly and Annual Meeting, November 28 - December 3, 2010 ,Chicago IL. http://archive.rsna.org/2010/9001628.html