Estimation of the pathological invasive size of pulmonary adenocarcinoma using high-resolution computed tomography of the chest: A consideration based on lung and mediastinal window settings

Objectives: Since the proposal of the new classification of pulmonary adenocarcinoma (PADC), the size of pathological invasion has become more important. We aimed to determine whether high -resolution computed tomography (HRCT) could be used to preoperatively evaluate PADC invasive size. Methods: We investigated 360 complete resected cT1a-1b-2aNO PADCs. We examined the correlation of pathological invasive size with three HRCT parameters [whole tumor dimension in the lung window (LD), consolidation dimension in the lung window (CD), and tumor dimension in the mediastinal window (MD)]. HRCT prediction of an invasive size of <= 5 mm was determined using receiver operating characteristic curve analysis. Results: Pathological invasive size correlated well with both CD (r(2) = 0.710) and MD (r(2) = 0.743) comparably, and moderately with LD (r(2) = 0.514). CD and MD tended to be slightly larger and smaller, respectively, than the actual invasive size. Invasive size roughly approximated to MD +3 mm, and an invasive size of <= 5 mm was best predicted by MD, followed by CD. MD of <= 2 mm and 0 mm predicted an invasive size of <= 5 mm with 64.1% and 47.4% sensitivity and 96.5% and 98.9% specificity, respectively. Lymphovascular invasion was best predicted by MD followed by CD. Pleural invasion and lymph node metastasis was predicted well by both MD and CD. Conclusion: Preoperative estimation of the invasive size of PADC and evaluation of other parameters of invasiveness were possible using MD. This approach using HRCT may play a complementary role in more thorough clinical staging of PADC. (C) 2016 Elsevier Ireland Ltd. All rights reserved.