Differentiation of localized pneumonic-type lung adenocarcinoma from localized pulmonary inflammatory lesion based on clinical data and multi-slice spiral computed tomography imaging features.

Abstract

Localized pneumonic-type lung adenocarcinoma (L-PLADC) is a special type of lung adenocarcinoma, which mimicking localized pulmonary inflammatory lesion (L-PIL), and many delayed diagnoses of L-PLADC have been identified due to insufficient clinical understanding or the lack of knowledge regarding the radiological findings. Multi-slice spiral computed tomography (MSCT) not only observes the fine structure of the lesion clearly, but also can evaluate the lesion and its surrounding tissues more intuitively, stereoscopically, and accurately using a variety of reconstruction techniques. The present study aimed to investigate the diagnostic value of clinical data and MSCT imaging features in differentiating L-PLADC from L-PIL. The clinical data and chest MSCT imaging features of 71 patients with L-PLADC and 70 patients with L-PIL were retrospectively analyzed. Seventy-one patients with L-PLADC underwent surgical resection or puncture and were confirmed as having invasive adenocarcinoma by pathology. Seventy patients with L-PIL were confirmed by clinical anti-inflammatory treatment or by puncture and surgery. The Chi-square and Mann-Whitney U tests were used to analyze the clinical data and MSCT imaging features of the included patients. Variables with P<0.05 in the univariate analysis were included in the multivariate logistic regression analysis to determine the independent risk factors for the diagnosis of L-PLADC. The clinical data analysis showed that multivariate logistic regression analysis showed that irregular air bronchogram [odds ratio (OR) =15.946; P<0.001], ground-glass opacity (GGO) component (OR =12.369; P<0.001), pleural traction (OR =10.982; P<0.001), necrosis (OR =0.078; P<0.001), adjacent bronchial wall thickening (OR =0.017; P<0.001), pleural thickening (OR =0.074; P<0.001), and respiratory symptoms were independent risk factors for the diagnosis of L-PLADC [OR =0.117; the area under the curve (AUC), sensitivity, specificity, and accuracy values were 0.989, 97.2%, 94.3%, and 95.7%, respectively]. L-PLADC and L-PIL exhibit different clinical and MSCT imaging features. Determining these characteristics is conducive to the early diagnosis and clinical treatment of L-PLADC.