Clinical Study Of Chest Thin Slice CT And ¹⁸F-FDG PET/CT In The Diagnosis Of Histopathological Subtypes Of Stage Ⅰ Invasive Adenocarcinoma

Objective:To investigate the value of thin-layer CT morphological parameters and¹⁸F-FDG PET/CT metabolic parameters in the differential diagnosis of LPA and non LPA in the pathological subtypes of stage I invasive lung adenocarcinoma.Methods:（1）We selected 103 patients with stage I invasive lung adenocarcinoma who underwent chest thin-section CT scan before surgery and underwent radical resection of lung cancer.They were divided into LPA group and non-LPA group according to histopathological subtypes.The clinical data of 103 patients were collected,the CT morphological characteristics of lung cancer lesions were analyzed,and the quantitative parameters such as size and CT value were measured.Medcalc20 0.3 and SPSS 23.0 software were used for statistical analysis.（2）A total of 26 cases of stageⅠinvasive lung adenocarcinoma with ¹⁸F-FDG PET/CT scan before operation and radical resection of lung cancer were selected and divided into LPA group and non-LPA group according to histopathological subtypes.Clinical data of these patients were collected and PET/CT metabolic parameters and breath-holding chest CT morphological parameters of lung cancer lesions were measured.Statistical analysis was performed using Med Calc20.0.3 and SPSS 23.0software.Assessing CT morphological characteristics and standardized uptake value of PET standardized uptake value（SUV max）,peak SUV（SUV peak）,mean SUV（SUV mean）,tumor metabolic volume（MTV）,total glycolysis（TLG）are correlated with LPA and non-LPA in stageⅠlung adenocarcinoma.The receiver operating characteristic（ROC）curve was used to determine the best cut-off point to distinguish LPA and non LPA,and its diagnostic efficiency was evaluated.Results:（1）The study of chest thin-slice CT group found that gender,age,lesion location,smoking history,hair prick sign and vascular cluster had no statistical significance in distinguishing LPA and non LPA pathological subtypes of stage I invasive lung adenocarcinoma;Vacuole sign,lobulation sign,pleural depression sign,CT value,long diameter and short diameter have statistical significance in distinguishing the above two pathological subtypes.The probability of vacuole sign was higher in LPA group（P<0.05）,while the probability of lobulation sign and pleural depression sign was higher in non LPA group（P<0.05）.The CT value,long diameter and short diameter of LPA group were significantly lower than those of non LPA group（P<0.001）.The ROC curve showed that the best cut-off points of CT value,long diameter and short diameter in the diagnosis of LPA were-235.50hu,15.50mm and 10.50mm respectively.Multivariate logistic regression analysis showed that CT value,lobulation sign and vacuole sign were independent predictors of LPA in stage I lung adenocarcinoma.（2）The study of PET/CT group found that there was no significant difference in gender,age,smoking history,lesion location,hair prick sign,vacuole sign,lobulation sign,pleural depression sign,vascular bundle sign,long diameter and short diameter of lesions,suvpeak,MTV and TLG between LPA group and non LPA group（P>0.05）;The differences of SUVmax,suvmean and CT values between LPA group and non LPA group were statistically significant（P<0.05）.ROC curve shows that SUVmax 4.50（AUC 0.810,sensitivity 100%,specificity 65%）,suvmean 1.92（AUC 0.854,sensitivity 67%,specificity 94%）,CT value 10.00hu（AUC 0.830,sensitivity 100%,specificity 65%）is the best cut-off point to distinguish LPA from non LPA.Conclusion:SUVmax,suvmean,CT value in PET/CT and vacuole sign,lobulation sign,pleural depression sign,long diameter,short diameter and CT value in thin-layer CT morphological characteristics of chest have significant correlation with different pathological subtypes of stage I lung adenocarcinoma.Combined judgment of CT value,lobulation sign and vacuole sign is of certain value in distinguishing LPA from non LPA.