| Objective:To evaluate the application value of Shear wave elastography(SWE)in the identification of benign and malignant thyroid nodules,selecte the stability of the measurement index and analyze the influencing factors.Methods:1.In this research,139 cases of thyroid nodules were confirmed by pathology,and EmaxandEmeanvaluesweremeasured,analysisofthesensitivity,specificity,accuracy,PPV and NPV.2.The nodules were divided into two groups according to their length,?1.0cm and>1.0cm,base on the surgery or thyroid fine needle aspiration(FNA)pathological results as the gold standard,Emax and Emean value ROC curves of two groups of nodules were constructed respectively,the best benign and malignant optimal cut-off value were obtained respectively.Compare the sensitivity,specificity,accuracy PPV,NPV between the two groups,and compare the diagnostic efficacy of Emax and Emean values for two groups respectively.According to the statistical result,explore the influence of different size nodules on the diagnosis of SWE technology,and analyze the reasons.3.Analysis of the“stiff rim”sign in the shear wave elastic image ofthyroid nodules,to calculate the sensitivity and specificity.4.Compared with all benign nodules and all malignant nodules respectively,the values of Emax and Emean whether have statistically different on the longitudinal and transverse sections.By comparing the two measures,choose the better.5.To analyze the possible factors affecting SWE technical diagnosis by consulting literature and operational experience,and combined with 139 nodules in this research,analyzed the causes of false positive and false negative.Results:1.There were 139 noduls in this research,76 cases were benign and 63 cases were malignancy.Benign nodules included 45 nodular goiter,9 follicular adenoma and eosinophilic adenoma,1 granulomatous thyroiditis,4 hashimoto's thyroiditis nodules and 2 patients with nodular goiter with adenomatous hyperplasia,1 patient with nodular goiter with atypical hyperplasia,14 cases of FNA biopsy were considered benign lesions.The malignant nodules were all papillary carcinoma.2.The general data of 139 benign and malignant nodules were compared:the age,sex and nodules of benign and malignant nodules were not statistically significant(P>0.05).3.Among the 139 nodules,there were statistically significant differences in Emax and Emean values between benign and malignant.The sensitivity,specificity,accuracy,PPV and NPV of Emax and Emean values in longitudinal section were63.5%?82.9%?74.1%?75.5%?73.3%and 71.4%?68.4%?69.8%?65.2%?74.3%respectively;in transverse section were 73.0%?80.3%?77.0%?75.4%?78.2%and87.3%?53.9%?69.8%?61.1%?83.7%respectively.4.The 139 nodules was divided into two groups according to their maximum diameter,namely group I:the maximum diameter was less than 1.0cm,and group II:the largest diameter was greater than 1.0cm.There were 58 cases of nodules in I group,27 cases were benign and 31 cases were malignant according to pathological results.There were 81 cases of nodules in group II,49 cases were benign and 32cases were malignant according to pathological results.There were statistically significant differences between the two groups(P<0.05).5.With pathologic results as the gold standard,build the I group of the Emax values ROC curve,the optimal cut-off value was 33.4 KPa,AUC was 0.840,95%confidence interval was 0.736-0.944,the sensitivity,specificity,accuracy,PPV and NPV were 77.4%,85.2%,81.0%,85.2%and 76.7%.With pathologic results as the gold standard,build the II group of the Emax values ROC curve,the best cut-off value was 46.6 KPa,AUC was 0.767,95%confidence interval was 0.660-0.874,the sensitivity,specificity,accuracy,PPV and NPV were 68.8%,81.6%,76.5%,81.6%and 80.0%.Two groups nodules Emax value between the sensitivity,specificity,accuracy,PPV and NPV comparison,P values were 0.438,0.694,0.526,0.172 and 0.724,the result shows all the P values>0.05,there was no statistically significant difference.6.With pathologic results as the gold standard,build the I group of the Emean values ROC curve,the optimal cut-off value was 24.5 KPa,AUC was 0.849,95%confidence interval was 0.746-0.953,the sensitivity,specificity,accuracy,PPV and NPV were 71.0%,92.6%,81.0%,92.6%and 73.5%.With pathologic results as the gold standard,build the II group of the Emean values ROC curve,the optimal cut-off value was 18.1 KPa,the AUC was 0.667,95%confidence interval was 0.545-0.790,the sensitivity,specificity,accuracy,PPV and NPV were 81.2%,46.9%,60.5%,46.9%and 79.3%.The two groups nodules Emean value between the sensitivity,specificity,accuracy,PPV and NPV comparison,P values were 0.338,0.000,0.010,0.000 and0.591,the results shown the P values of the specificity,accuracy,and PPV was<0.05,the difference was statistically significant;the P values of the sensitivity and NPV was>0.05,and the difference was not statistically significant.7.When the Emax optimal cut-off value was 33.4KPa in I group,there were 4cases false positive,7 cases false negative.Among the group II,when the Emax optimal cut-off value was 46.6KPa,there were 9 cases false positive,and 10 cases false negative.When the Emean optimal cut-off value was 24.5KPa in I group,there were 2 cases false positive,and 9 cases false negative.Among the group II,when the Emean optimal cut-off value was 18.1KPa,there were 26 cases false positive,and 6 cases false negative.8.The AUC under the ROC curve of group I and II was 0.840 and 0.767respectively,and the diagnostic efficacy of the two groups of nodules was compared,The Z value=0.96,P=0.33,and the difference was not statistically significant(P>0.05).The AUC under the ROC curve of group I and II was was 0.849 and 0.667respectively,and the diagnostic efficacy of the two groups of nodules was compared,The Z value=2.23,P=0.026,and the difference was statistically significant(P<0.05).9.Among the 63 malignant nodules,11 cases showed the“stiff rim”sign,accounting for 17.5%;Of the 76 benign nodules,4 of the images showed the“stiff rim”sign,accounting for 5.3%;It accounts for 10.8%of the 139 nodules.Malignant nodules showed that the results were significantly higher than benign nodules,?~2=5.285,P=0.022.Sensitivity,specificity,PPV and NPV were 17.5%,94.7%,73.3%and 58.1%respectively.10.76 cases benign nodules and 63 cases malignant nodules of Emax values in the longitudinal and transverse sectionsin of P value was 0.428 and 0.747respectively,there was no significant difference(P>0.05);76 cases benign nodules and 63 cases malignant nodules of Emean values in the longitudinal and transverse sectionsin of P value was 0.000 and 0.015 respectively,there was significant difference(P<0.05).Conclusions:1.The SWE technique is an effective method for differential diagnostic of the benign and malignant thyroid nodules,there was significant difference.2.For Emax values,the optimal cut-off value according to ROC curve of length?1.0cm and leng>1.0cm nodules was 33.4KPa and 46.6KPa respectively,the size of nodules had no effect on the diagnosis of Emax values.3.For Emean values,the optimal cut-off value according to ROC curve of length?1.0cm and leng>1.0cm nodules was 24.5KPa and 18.1KPa respectively,the size has a great influence on the diagnosis of Emean value,and the diagnostic of the nodules with no more than 1.0 cm are better than>1.0cm.4.In SWE image,the“stiff rim”sign diagnosis of thyroid malignant nodules is highly specific,so the occurrence of the“stiff rim”sign should be cautious of the possibility of malignancy.5.The Emax value is better than Emean value in measuring the thyroid nodules in the longitudinal and transverse sectionsin.Choosing a better index of stability could help improve the accuracy of diagnosis. |
PDF Full Text Request