Persistent pure ground-glass nodules in the lung: interscan variability of semiautomated volume and attenuation measurements.

OBJECTIVE: The purpose of this article is to assess the interscan variability of semiautomated volume and attenuation measurements for pulmonary persistent pure ground-glass nodules. SUBJECTS AND METHODS: Thirty pure ground-glass nodules in 25 patients referred for known pure ground-glass nodules were evaluated with two consecutive unenhanced chest CT examinations performed within 10 minutes. For the second limited scan, patients were asked to first get off and then get back on the table to simulate a follow-up examination. CT images were reconstructed using two different algorithms (sharp, B60F; medium sharp, B50F). Two chest radiologists independently measured the volumes and attenuations of pure ground-glass nodules, twice on the first CT scan and once on the second CT scan, using commercial software. Interscan variability was measured using the Bland-Altman method.
RESULTS: Nodule segmentation was successful in 98.3% (177/180) and 97.8% (176/180) of measurements with B60F and B50F, respectively. With respect to volume measurements, interscan variability for reviewer 1 was -17.7% to 18.3% with B60F and -18.8% to 15.7% with B50F; interscan variability for reviewer 2 was -18.6% to 18.9% with B60F and -15.6% to 17.2% with B50F. With respect to attenuation measurements, interscan variability for reviewer 1 was -7.2% to 7.9% with B60F and -7.9% to 9.4% with B50F; interscan variability for reviewer 2 was -7.7% to 8.1% with B60F and -8.5% to 9.9% for B50F.
CONCLUSION: Variations in volume and attenuation measurements of persistent pure ground-glass nodules using commercial software were reasonably small, allowing the detection of clinically relevant growth.