Three-dimensional volumetric assessment with thoracic CT: a reliable approach for noncalcified lung nodules?

We read with great interest the article by Dr Gavrielides and colleagues (1) in the April 2009 issue of Radiology. The authors examined the performance of different algorithms for nodule segmentation and volume estimation. They also provided a review of findings from several published studies regarding the volumetric computed tomographic (CT) analysis of lung nodules focused on the main factors that affect volumetric analysis in terms of precision and accuracy. In their study (1), a number of factors that may affect the volumetric CT analysis of lung nodules, such as the field of view (FOV) and the effect of contrast media on three-dimensional (3D) volumetric measurement of pulmonary nodules, were not analyzed (2,3). In a study by Honda et al (3) and in our experience, when using a 64-channel multidetector CT scanner, the volume of a pulmonary nodule calculated with 3D volumetric software (Lung VCAR; GE Healthcare, Milwaukee, Wis) is substantially larger after administration of contrast media when the same technical parameters are used (mean percentage error, 17.67%). This is probably more related to the adaptative threshold algorithm than to the segmentation and shape-analysis algorithm of the software. We agree with the conclusions of Dr Gavrielides and colleagues about the section thickness (section width), which is one of the most important CT acquisition parameters to control. It is our opinion that this feature is particularly important for small (<10-mm) poorly circumscribed pulmonary nodules; consequently, a section thickness greater than 1.25 mm should not be used. Section thickness can also affect nodule attenuation. By using a section thickness of 5 mm (full FOV), solid nodules smaller than 10 mm are automatically differentiated by the software as nonsolid or partially solid. For these nodules, the volume calculated is greater than the real volume. To promote broader clinical application of 3D volumetric assessment of lung nodules, efforts should focus not only on optimizing CT technique, but also on ensuring that the volumetric analysis is reproducible with a variety of different software packages.