To analyze the risk factors of solitary pulmonary nodules and establish a predictive model for benign and malignant pulmonary nodules.

The clinical data of 112 patients with solitary pulmonary nodules, who had a definite pathological diagnosis and underwent thoracic surgery in Ya′an People′s Hospital from January 2017 to August 2018, were reviewed respectively. Their age, gender, smoking history, family history of cancer, past cancer history, serum cancer biomarkers including carcinoembryonic antigen (CEA), neuron specific enolase (NSE) and cytokeratin 19 fragment (CYFRA21-1), and the radiological charateristics including the nodule density, diameter, location, lobulation, burr, pleural indentation, vascular cluster sign, vacuole sign, air information of bronchial sign and calcification sign, were summarized. The patients were divided into two groups according to the pathological diagnosis of their benign or malignant pulmonary nodules. After a univariate analysis, the clinical information with significant differences was chosen for logistic regression analysis and these independent risk factors for malignant pulmonary nodules were screened. Finally, a predicative model for benign and malignant pulmonary nodules was established.

There were significant differences in age, past cancer history, serum CEA and CYFRA21-1, nodule density, lobulation, burr, pleural indentation sign, vascular cluster sign, vacuole sign and calcification sign between the patients with benign and malignant pulmonary nodules (P<0.05). Logistic regression analysis showed that age, CEA, CYFRA21-1, ground glass density and lobulation were independent risk factors for malignant pulmonary nodules. One model was established as following: P=e^x/(1+ e^x), x=-8.816+ (3.018×density)+ (0.073×age)+ (0.482×CEA)+ (0.426×CRFRA21-1)+ (1.421×lobulation).

Age, CEA, CYFRA21-1, ground glass density and lobulation are independent risk factors for malignant pulmonary nodules. And the predictive model of malignant pulmonary nodules has a better sensitivity and specificity and a high diagnostic accuracy.