Evaluation Of The Malignant Risk Of Pulmonary Mixed Ground-glass Nodules Based On Quantitative Analysis Of CT Imagines

Background and objective In recent years,with the implementation of lung cancer screening programs around the world and the wide application of low-dose CT,the detection rate of pulmonary nodules has increased significantly.Among them,mixed ground-glass nodule(m GGN)is considered closely related to early-stage lung adenocarcinoma.Therefore,early identification and effective intervention of malignant m GGN is expected to realize the "port forward" of lung cancer diagnosis and improve the outcomes of patients,which is the focus of clinical work today.At present,chest CT examination is still the main method for pulmonary nodule screening,but the imaging signs of m GGN are often atypical,so it is difficult to distinguish benign nodules from malignant nodules only relying on traditional two-dimensional images.With the development and promotion of 3D reconstruction technology,more quantitative information on CT images can be mined,which brings a new dawn for the characteristic quantification of pulmonary nodules.The purpose of this study was to analyze the differences in quantitative parameters of CT images of different pathological types of m GGN,and to explore the predictive value of CT image quantitative analysis methods for m GGN malignancy risk in the lungs,so as to provide a reference for the early identification of benign and malignant m GGN and clinical intervention.Materials and methods We retrospectively collected patients admitted to our hospital from December 2017 to March 2021,and who were diagnosed with mixed ground-glass nodules in the lung on CT examination and underwent surgical treatment to obtain clear histopathological results were enrolled.According to the pathological type,the enrolled m GGN were divided into precursor glandular lesions(atypical adenomatous hyperplasia,carcinoma in situ)group,minimally invasive adenocarcinoma(MIA)group,and invasive adenocarcinoma(IAC)group.The clinical data(age,gender,smoking history,history of lung diseases,family history of tumor),imaging features(nodule location,spicule sign,vacuole sign,pleural indentation sign,air-bronchogram sign,angioassemblage sign,calcification sign)were recorded,and quantitative parameters(such as the long-axis diameter of nodules,the volume of nodules,the average CT value,the microvascular density,etc.)of m GGN were extracted by Digital Lung TM software(Dexin,Shanxi,China).Differences in clinical,imaging features and quantitative parameters were compared among the three groups,and logistic regression analysis was used to determine the predictors associated with IAC.Screening m GGN with measurable solid components in all nodules,and comparing the measured values of solid components(long-axis diameter of solid components,proportion of long-diameter of solid components,volume of solid components,and volume of solid components)among the three groups.Logistic multivariate regression analysis was used to determine the independent risk factors related to IAC,and the receiver operating characteristic(ROC)curve was drawn to measure the area under the curve(AUC),sensitivity,specificity,then compared the performance of different quantitative parameters in predicting IAC.Further explored the differences in CT image quantitative parameters of different pathological subtypes(lepidic,acinar,papillary,and micropapillary)in the IAC group.Results Finally,267 cases of m GGN were enrolled,including 51 cases in the glandular prodromal disease group,45 cases in the MIA group,and 171 cases in the IAC group.There were no significant differences among the three groups in gender,smoking history,family tumor history,nodule location,spicule sign,vacuole sign,pleural indentation sign,air-bronchogram sign,calcification sign(P>0.05).There were significant differences in age and history of lung diseases among the three groups(P<0.001).There was statistical significance in the angio-assemblage sign,the long-axis diameter of nodules,the volume of nodules,the average CT value,and the microvascular density among the three groups(P<0.05).the long-axis diameter of nodules(OR,1.200;95%CI,1.108-1.300,P < 0.001)and the average CT value(OR,1.013;95%CI,1.009-1.016,P < 0.001)were independent factors for identifying IAC.A total of 123 cases of m GGN’s solid components could be measured,including 7 in the precursor glandular lesions group,31 in the MIA group,and 85 in the IAC group.There were significant differences in the long-axis diameter of m GGN,the long-axis diameter of solid components,the proportion of long-axis diameter of solid components,the volume of nodules,the volume of solid components,the proportion of the volume of solid components,the average CT value,and the density of microvessels among the three groups(P<0.001).Among them,the long-axis diameter of the solid component(OR,1.449;95%CI,1.014~2.069;P=0.042),the proportion of the long-axis diameter of the solid component(OR,1.134;95%CI,1.013~1.269;P=0.028),and the proportion of the volume of the solid component(OR,1.536;95%CI,1.204-1.961;P=0.001)were independent predictors of IAC,with AUCs of 0.927,0.940,and 0.969,respectively.Taking the long-axis diameter of solid components greater than 5.83 mm as the critical value,the sensitivity and specificity to predict IAC were 89.4% and 86.8%,respectively;taking the ratio of the long-axis diameter of the solid component to the long-axis diameter of the nodule greater than 39.965% as the critical value,the predicting sensitivity and specificity were 91.8% and 86.8%,respectively;taking the ratio of the volume of the solid component to the volume of the nodule greater than 19.815% as the critical value,the predicting sensitivity and specificity were 83.5% and 100% respectively.The combination of the three quantitative parameters had the best performance in predicting IAC,with an AUC of 0.988,sensitivity and specificity of 96.5% and 94.7%,respectively.There was no difference in quantitative parameters among the subtypes in the IAC group(P>0.05).Conclusion 1.Compared with imaging morphological signs such as the spicule sign,vacuole sign,pleural indentation sign,air-bronchogram sign,angio-assemblage sign,calcification sign,CT quantitative parameters such as the long-axis diameter of nodules,the volume of nodules,the average CT value,the microvascular density and the quantitative parameters of solid components could provide more objective and effective information for the identification of IAC;2.When the long-axis diameter of the solid component is greater than 5.83 mm,the ratio of the longaxis diameter of the solid component to the long-axis diameter of the nodule is greater than 39.965%,and the ratio of the volume of the solid component to the volume of the nodule greater than 19.815%,it is highly suggested that the nodule has progressed to infiltrative lesions,for which biopsy or/and surgery should be performed as soon as possible.