Nomogram for Predicting the Risk of Invasive Pulmonary Adenocarcinoma for Pure Ground-Glass Nodules

Abstract

Background: The pure ground-glass nodule is a common clinical concern. The purpose of this study was to develop a nomogram to predict the risk of invasive pulmonary adenocarcinoma in patients with pure ground-glass nodules 1 cm or less in diameter based on the computed tomography imaging features.

Methods: We reviewed a total of 250 patients who had undergone resection of pure ground-glass nodules 1 cm or less in diameter in our hospital, including the data of histopathologic examinations after surgical resection. Clinical information and imaging features were analyzed by the use of univariable and multivariable logistic regression analysis. A nomogram to predict the risk of invasive pulmonary adenocarcinoma was developed, and the calibration curves for the probability were drawn.

Results: Two hundred fifty patients were enrolled for analysis: 13 as having benign lesions, 8 as atypical adenomatous hyperplasia, 64 as adenocarcinoma in situ, 137 as minimally invasive adenocarcinoma, and 28 as invasive adenocarcinoma. On the basis of the final regression analysis, the lesion size, spiculation, lobulation, air bronchogram, vascular convergence, pleural tag, and the computed tomography window width were identified and were entered into the nomogram. The nomogram showed an excellent discrimination, with an area under the receiver operating characteristics curve of 0.916 for internal validation and 0.970 for external validation. The calibration curves for the probability of invasive pulmonary adenocarcinoma showed optimal agreement between the predicted probability by the nomogram and the actual probability.

Conclusions: We constructed a novel nomogram that can predict the risk of invasive pulmonary adenocarcinoma for patients with pure ground-glass nodules 1 cm or less in diameter. With highly accurate, excellent calibration and discrimination of the model, clinicians could more precisely make a specific treatment strategy for each patient.