The Study Of CT Quantitative Parameters In Predicting The Invasiveness Of Pulmonary Ground-glass Nodules

Objective To investigate the predictive ability of CT quantitative parameters on the invasiveness of pulmonary ground-glass nodules,and to establish the best predictive model based on its clinical value.Materials and methods Our study included data of318 pulmonary adenocarcinoma patients(327 nodules in total)presenting as pulmonary ground-glass nodules between January 2013 and December 2016,confirmed by surgical pathology.According to imaging findings,the cases were divided into two groups for analysis.Group A was a pure ground-glass nodule group,and group B was a partial solid nodule group.The pathological types of pure ground-glass nodules group include atypical adenomatous hyperplasia,adenocarcinoma in situ,minimally invasive adenocarcinoma and invasive adenocarcinoma.The pathological types of part solid nodules group include minimally invasive adenocarcinoma and invasive adenocarcinoma.The CT images of selected cases were confirmed and then copied to Osirix for measurement and calculation.The viewers sketched the outline of each nodules layer by layer,and analyzed the maximum diameter of the pulmonary window,the maximum diameter of the pulmonary window,longest diameter perpendicular to maximum diameter of the pulmonary window,the maximum cross-sectional area of the lesion,the maximum diameter of the mediastinum window,longest diameter perpendicular to maximum diameter of the mediastinum window,the mean CT attenuation,volume,mass,and the corresponding TDR(tumor shadow disappearance rate)for each lesion.The relationship between the parameters and different pathological types of lung adenocarcinoma was evaluated by ROC and Logistic regression analysis.Prediction models and corresponding nomograms were established to predict the invasiveness of the lesions.Result Logistic regression analysis of group A showed that the maximum cross-sectional area(OR=2.307,95%CI:1.689-3.150)of the lesions independently predicted the invasiveness of pGGNs.The results of pair-mannered ROC curve comparison showed that the area under the curve(AUC)of the mean CT attenuation was the smallest(P<0.01).Logistic regression analysis of group B showed that the maximum diameter on pulmonary window(OR=3.080,95%CI:1.135-8.355)and the maximum diameter of mediastinum window(OR=5.881,95%CI:1.634-21.166)independently predicted the invasiveness of PSNs.The area under the curve corresponding to the combination of two parameters was 0.855,with sensitivity of77.61%and specificity of 86.67%.The largest diameter of the pulmonary window and the largest diameter of the mediastinum window corresponding to the cut-off point were 1.902 cm and 1.273 cm,respectively.The nomogram of group A consisting of pure ground-glass nodules showed that the total score of all the corresponding independent factors are between 2 and 54 points,the risk range from0.10 to 0.90,the higher the total score is,the greater the risk of pGGNs being invasive.Similarly,the nomogram of group B consisting of part solid nodules showed that the total score of all the corresponding independent factors are between 13 and107,the risk range from 0.20 to 0.99,the higher the total score is,the greater the risk PSNs being invasive.Conclusion CT quantitative parameters are of great value in predicting the invasiveness of pulmonary ground-glass nodules.It can also provide a references for the clinical management and strategy.The largest cross-sectional area of the pulmonary window of pGGNs(≥2.223 cm~2)is an independent predictor for its invasiveness.The maximum diameter of the pulmonary window(≥1.902cm)and the maximum diameter of the mediastinum window(≥1.273cm)of PSNs are independent predictors of its’invasiveness.The application of nomograms makes it more intuitive to assess the invasiveness of GGNs.