Establishment of A Clinical Prediction Model of Solid Solitary Pulmonary Nodules

• Main Authors: Wei YU, Bo YE, Liyun XU, Zhaoyu WANG, Hanbo LE, Shanjun WANG, Hanbo CAO, Zhenda CHAI, Zhijun CHEN, Qingquan LUO, Yongkui ZHANG
• Format: Article
• Language: zho
• Published: Chinese Anti-Cancer Association; Chinese Antituberculosis Association 2016-10-01
• Series: Chinese Journal of Lung Cancer
• Subjects: Solitary pulmonary nodules (SPNs); Prediction model; Independent predictors
• Online Access: http://dx.doi.org/10.3779/j.issn.1009-3419.2016.10.12

Background and objective The solitary pulmonary nodule (SPN) is a common and challenging clinical problem, especially solid SPN. The object of this study was to explore the predictive factors of SPN appearing as pure solid with malignance and to establish a clinical prediction model of solid SPNs. Methods We had a retrospective review of 317 solid SPNs (group A) having a final diagnosis in the department of thoracic surgery, Shanghai Chest Hospital from January 2015 to December 2015, and analyzed their clinical data and computed tomography (CT) images, including age, gender, smoking history, family history of cancer, previous cancer history, diameter of nodule, nodule location (upper lobe or non-upper lobe, left or right), clear border, smooth margin, lobulation, spiculation, vascular convergence, pleural retraction sign, air bronchogram sign, vocule sign, cavity and calcification. By using univariate and multivariate analysis, we found the independent predictors of malignancy of solid SPNs and subsequently established a clinical prediction model. Then, another 139 solid SPNs with a final diagnosis were chosen in department of Cardiothoracic Surgery, Affiliated Zhoushan Hospital of Wenzhou Medical University as group B, and used to verify the accuracy of the prediction model. Receiver-operating characteristic (ROC) curves were constructed using the prediction model. Results Multivariate Logistic regression analysis was used to identify eight clinical characteristics (age, family history of cancer, previous cancer history, clear border, lobulation, spiculation, air bronchogram sign, calcification) as independent predictors of malignancy of in solid SPNs. The area under the ROC curve for our model (0.922; 95%CI: 0.865-0.961). In our model, diagnosis accuration rate was 84.89%. Sensitivity was 90.41%, and specificity was 78.79%, and positive predictive value was 80.50%, and negative predictive value was 88.14%. Conclusion Our prediction model could accurately identify malignancy in patients with solid SPNs, thereby it can provide help for diagnosis of solid SPNs.