Font Size: a A A

Comparison Of Thin-section CT Findings And Histopathology Of Small Pure Ground-glass Nodules

Posted on:2016-06-14Degree:MasterType:Thesis
Country:ChinaCandidate:W J XiangFull Text:PDF
GTID:2284330464452049Subject:Imaging and nuclear medicine
Abstract/Summary:PDF Full Text Request
【Objective】 To retrospectively compare the thin-section CT features of pure ground-glass nodules(p GGN) of ≤10 mm in diameter with histopathology to figure out the characteristic features between different pathological types. 【Methods】 Two hundred and five pure ground-glass nodules of 10 mm or less were reviewed, all nodules were histologically confirmed and divided into three groups: 40 atypical adenomatous hyperplasia(AAHs), 95 adenocarcinoma in situ(AISs), 70 minimally invasion adenocarcinomas(MIAs). The patient gender, age, maximum diameter, mean attenuation value and morphologic features(included interface, slight lobulation, spiculation, spine-like process, vascular convergence or dilatation, air bronchograms and pleural retraction) were recorded. First using Pearson χ2 test, Fisher’s exact test, analysis of variance to analyze each of the variables, then selected significant(P<0.05) variables as independent variables, pathological type as the dependent variable, using multinomial logistic analysis to assess differences between the three groups. Finally,regrouped the nodules according to the analysis results, drew ROC curves, then the optimal cut-off values were confirmed. 【Results】(1) The presence of maximum diameter, mean attenuation value, interface and air bronchograms were significantly different between the three groups(P<0.05),(2) results of multinomial logistic analysis: maximum diameter was an significant factor that differentiated AAH from AIS[P=0.005,Odds Ratio(OR) =65], AAH from MIA(P=0.001,OR=236), but insignificantly different between AIS and MIA, so AAH was referred as one group, AIS and MIA were combined into another group, ROC curve showed the optimal cut-off for AAH was less than 6.5 mm in diameter, AUC=0.711; mean attenuation value was an significant factor that differentiated AAH from MIA(P=0.01, OR=1.006), AIS from MIA(P=0.006,OR=1.004), but insignificantly different between AAH and AIS, so AAH and AIS were combined into one group, MIA was referred as another group, ROC curve showed the optimal cut-off for AAH or AIS was less than-520 HU, AUC=0.643; interface was an significant factor that differentiated AAH from AIS, MIA respectively(P<0.05), but not significantly different between AIS and MIA; an air bronchogram was an significant factor that differentiated MIA from AAH(P<0.05), but not significantly different between AAH and AIS, AIS and MIA. 【Conclusion】 Among p GGNs of 10 mm or less, maximum diameter and interface are significant factors that differentiate AAH from AIS, MIA respectively, but no differences are found between AIS and MIA; mean attenuation value can be useful to distinguish preinvasive lesions(<-520 HU) from invasive lesions(≥-520 HU), but can’t differentiate AAH from AIS; an air bronchogram can differentiate MIA from AAH(P<0.05), but not significantly different between AAH and AIS, AIS and MIA.
Keywords/Search Tags:Coin lesion, Tomography, X-ray computed, Factor analysis, statistical
PDF Full Text Request
Related items