| Objective: To explore the feasibility of the evaluation on the pathological subtypes of early lung adenocarcinoma by using pulmonary nodule analysis semi-automatic measurement software,dynamic CT enhancement scanning technology and CT perfusion imaging technology,it could play an important role in guiding and predicting the prognosis of patients with lung adenocarcinoma,helping clinical decision-making treatment plan and efficacy evaluation.Methods: Retrospective collections of the clinical data and CT imaging data of 296 lesions in 290 patients with early lung adenocarcinoma confirmed by surgery and pathology in Tangshan Gongren Hospital from February 2015 to December 2018,which included 160 pure ground-glass nodules(p GGN),95 mixed ground-glass nodules(m GGN),43 solid nodules(SN).According to the 2015 edition WHO lung tumor histological classification criteria,the pathological subtypes of pulmonary adenocarcinoma were mainly divided into five groups: lepidic predominant group,acinar predominant group,papillary predominant group,solid predominant group and micro-papillary predominant group.According to the different prognosis of lung adenocarcinoma,the pathological subtypes of lung adenocarcinoma were divided into three groups: group Ⅰ involved lepidic predominant group;group Ⅱ involved acinar and(or)papillary predominant group and group Ⅲinvolved other subtypes.The clinical data included patient gender,age,CT imaging morphology and functional data: edge structure included lobulation sign and spiculated sign,internal structure included vocuole sign,air bronchogram and abnormal vascular sign(vascular entry or cluster sign),adjacent structural changes such as pleural indentation sign,nodular density classification,the size(effective diameter)and the distribution of the lesion,plain mean CT value,arterial phase mean CT value,venous phase mean CT value and lesion perfusion parameter value were collected.These variables were compared between the three groups by using SPSS 24.0.The measurement data accorded with normal distribution were analyzed by using one-way ANOVA.Otherwise by Kruskal-wallis H test(Non-parametric test).Count data were compared by using chi-square test.If the theoretical frequency was less than 5,Fisher’s exact test was used.These variables with statistical difference were screened out(P<0.05).And then the variables which exhibited statistically significant differences were included in a logistic regression analysis.The prediction probability of the subjects was calculated,then plotted the Receiver operating characteristic(ROC)curve and confirmed the optimal cut-off value.Some indexes were statistically analyzed by the Spearman correlation coefficient(rs).P<0.05 was regarded as statistically significant differences.Results: In this study,290 patients and 296 lesions were included,including 6 patients with 2 lesions.100 Males accounted for 34.5% and 190 females accounted for 65.5%;They aged from 31 to 78,with a median age of 58,The mean age was 56.5±9.5.Group Ⅰ(196)accounted for 66.2 %,Group Ⅱ(87)accounted for 29.4 %,Group Ⅲ(13)accounted for 4.4 %.Patient gender,nodular density classification,spiculated sign,lobulation sign,abnormal vascular sign,air bronchogram,pleural indentation sign were statistically significant difference between the three groups(P<0.05).The effective diameter,mean CT value,blood volume(BV)and peak enhancement intensity(PEI)were statistically significant difference between the three groups(P<0.05).There was no significant difference in age,vacuole sign,perfusion value(PV)and time to peak(TTP)between the three groups(P=0.209、0.577、0.106、0.823).Comprehensive analysis showed that the combined application of some morphology indexes abnormal vascular sign,spiculated sign,pleural indentation sign,nodular density classification,some quantitative measurement indexes such as effective diameter,plain mean CT value and BV value had a higher diagnostic efficiency between lepidic predominant group and acinar and(or)papillary predominant group.Among these indexes,the plain mean CT value was the most effective one.-241.0HU was the cut-off value between the two groups.Therefore,when the effective diameter of m GGN or SN is more than 1.5cm on the thin MSCT,the associated morphological indexes such as lobular sign,spiculated sign,abnormal vascular sign,air bronchogram sign and pleural depression sign,and the average CT value of plain scan is more than-241.0HU,the possibility of acinar growth and / or papillary growth as the main type of adenocarcinoma should be considered.However,in female patients,p GGN and effective diameter were smaller,without the above morphological indicators,and the average CT value of plain scan is less than-241.0HU,it is possible to consider lepidic predominant adenocarcinoma.The number of cases of other subtypes was only 13,and most of them were mucinous adenocarcinoma,so the statistical efficiency was weak.In the future,we need to increase the sample size for further study.Conclusion:1.The combination of morphological indexes and quantitative measurement indexes can distinguish early lung adenocarcinoma of different pathological subtypes,and predict the prognosis of lung adenocarcinoma patients by simple and non-invasive way.2.The average CT value of plain scan is-241.0HU,which is the effective index to distinguish the main type of lepidic adenocarcinoma from the main type of acinar growth and / or papillary growth.3.The effective diameter measured by semi-automatic segmentation software has a positive correlation with the maximum diameter of the specimen in pathology,and the diameter between them is very close. |
PDF Full Text Request