Comparative Study On CT Signs And Pathology Of Positive And Negative Patients With Spread Through Air Spaces Of Peripheral Lung Adenocarcinoma

Objective:To compare and study the clinical,pathological and CT characteristics of the spread through air spaces(STAS)positive group and the negative group on the similarities and differences and their correlations.The purpose of this study was to explore and clarify the feasibility and clinical application value of evaluating or predicting the occurrence of STAS in patients with lung adenocarcinoma based on CT signs.Methods:A total of 158 patients with invasive lung adenocarcinoma who underwent routine chest CT scans and confirmed by surgery and pathology in Affiliated Zhongshan Hospital of Dalian University from May 2014 to December 2020 were retrospectively collected,including 57 males and 101 females.The age range is 30～81 years old,with an average age of(61.0±9.1)years old.According to pathological observation,64 cases were divided into STAS positive group and 94 cases were STAS negative group.Record the clinical,pathological and CT data as follows:gender,age,blood type,smoking history,operation method,histopathological subtype,pathological stage,pathological differentiation,lymph node metastasis,location,lesion type,tumor length,and size The longest diameter of the sex component,consolidation tumor ratio(CTR),morphology,tumor-lung interface,lobular sign,burr sign,vacuole/cavity,air bronchial sign,pleural depression sign,vascular changes.The SPSS 20.0 software package was used for statistical analysis.Continuous variables were expressed as mean±standard deviation;Mann-Whitney U rank sum test was used for measurement data;χ~2test was used for count data.Binary logistic regression analysis was used to determine the independent risk factors of STAS;the receiver operating characteristic curve(ROC)of STAS-positive patients was drawn,and the AUC value,sensitivity and specificity of predicting STAS were calculated;Spearman correlation coefficient(r)was used to assess STAS The degree of correlation with CT signs.P value<0.05 was considered statistically significant.Results:(1)In the clinical and pathological data,the age of patients with lung adenocarcinoma(χ~2=5.435,P=0.020),pathological stage(χ~2=7.930,P=0.047),degree of differentiation(χ~2=21.345,P<0.001),whether Containing micropapillary components(χ~2=10.758,P=0.001)and the difference between two groups of STAS was statistically significant.(2)Among CT features,tumor lesion type(χ~2=25.048,P<0.001),tumor morphology(χ~2=10.266,P=0.001),tumor-lung interface(χ~2=13.005,P<0.001),lobular sign(χ~2=32.073,P<0.001),burr sign(χ~2=37.569,P<0.001),pleural depression(χ~2=12.194,P<0.001),vascular changes(χ~2=24.913,P<0.001),tumor long diameter(Z=-3.214,P=0.001),the maximum diameter of the solid component(Z=-5.863,P<0.001)and the CTR value(Z=-5.558,P<0.001)had statistically significant differences between the STAS groups;Among them,STAS had a moderate correlation with the lobular sign,burr sign,the maximum diameter of the solid component and the CTR value(r=0.451,0.488,0.468,0.443,all P values<0.001),and it had a moderate correlation with the type of tumor lesions(r=-0.394,P<0.001),tumor morphology(r=0.255,P=0.001),tumor-lung interface(r=0.287,P<0.001),pleural depression sign(r=0.278,P<0.001),vascular changes(r=0.397,P<0.001)and tumor long diameter showed a weak correlation(r=0.256,P=0.001).(3)Draw the ROC curve with the leaflet sign,burr sign and tumor-lung interface as the detection variables,and the area under the curve AUC was 0.71,0.74,0.64,and the sensitivity was 93.8%,89.1%,and 76.6%,respectively.The specificity was 48.9%,59.6%,and 52.1%,respectively.The ROC curve was drawn with the tumor long diameter,the maximum diameter of the solid component and the CTR value as the detection variables.The AUC of the tumor long diameter was 0.651,the sensitivity was 76.6%,and the specificity was 52.1%;The AUC of the largest diameter of the solid component was 0.78,the sensitivity was 71.9%,and the specificity was 75.5%;the AUC of the CTR value was 0.75,the sensitivity was84.4%,and the specificity was 67.0%.Conclusions:(1)There were differences in the pathological differentiation,tumor type,lobular sign,burr sign,pleural depression sign,maximum diameter of solid components,and CTR value between STAS positive and negative groups in patients with lung adenocarcinoma;CTR value,lobular sign,burr sign,tumor-lung interface were independent risk factors for predicting STAS of lung adenocarcinoma.(2)The lobular sign,burr sign,and tumor-lung interface CT signs of some solid tumors had better correlation with STAS and better predictive performance;through the analysis of clinical and integrated CT signs,it is expected to treat lung adenocarcinoma STAS before surgery Preliminary predictions are made to provide an important reference for clinically targeted formulation of reasonable treatment plans,and have a good clinical application prospect.