| Objective:To evaluate the shear wave elastography(SWE)value of young's modulus in the differential diagnosis of benign and malignant nodules(diameter ?10mm).The logistic regression model was constructed with the regression parameters of high frequency ultrasound haracteri-stics and Young's modulus.To compare the predictive probability of the model(PRE)with Thyroid imaging reporting and data system(TI-RADS)in the diagnosis of thyroid papillary microcarcinoma(PTMC).Methods:A total of 105 patients(78 patients with PTMC and 27 patients with benign nodules)with thyroid diseases was treated in our hospital from January 1,2017 to August 30,2019 with ultrasonic examination and surgical resection were selected,including 133 nodules(98 PTMC nodules and 35 benign nodules).All patients were in our department before operation:(1)High frequency ultrasonography was used to analyze the boundary,echo,internal structure,presence or absence of microcalcification,Longitudinal and transverse ratio(L/T),blood supply of nodules.(2)Several Young's modulus values were measured by SWE before operation.Maximum(SWEmax),mean(SWEmean)and minimum(SWEmin)were recorded as the parameters of nodule hardness.(3)The nodules were divided into benign group and malignant group.Statistically significant clinical parameters and ultrasonographic features between benign and malignant groups were screened by chi-square test and t-test,Logistic regression analysis was fitted and PRE was generated.Drew receiver operating characteristic curve(ROC)and determined the optimal threshold value of indicators by Yoden index.According to the critical value and the diagnostic "gold standard",a four-grid table was constructed to calculate the sensitivity,specificity,positive predictive value,negative predictive value and accuracy.Results:(1)There were 105 patients and 133 nodules.According to the pathological results,there were 35 benign nodules and 98 malignant nodules.The pathological changes of benign nodules were 15 nodular goiters,18 adenomas,1 chronic lymphocytic thyroiditis and 1 subacute thyroiditis.Pathology of malignant nodules was PTMC.(2)High frequency ultrasound in the diagnosis of thyroid nodules,boundar,microcalcification,L/T>1,blood flow in both benign and malignant groups(P<0.05).(3)The difference of Young's modulus SWEmax,SWEmin and SWmean in SWE was significant in the diagnosis of benign and malignant thyroid nodules(P=0.000).(4)Logistic regression analysis showed that the final indexes included in the model were boundaries(OR:3.47,95%CI:1.12-10.81),microcalcification(OR:5.86,95%CI:1.02-33.63),SWEmax(OR:1.17,95%CI:1.08-1.26)was an independent risk factor.Logistic regression model was established.(5)The proportion of malignant nodules in TI-RADS classifications 4a,4b,4c and 5 were 14.3%,48.1%,83.9%,100%,respectively.(6)The area under the ROC curve of each indicator was PRE:0.909(0.852-0.967),TI-RADS:0.735(0.631-0.838),the sensitivity,specificity,positive predictive value,negative predictive value and accuracy of each indicator were as follows:PRE(94.90%,80.00%,93.00%84.85%,90.98%),TI-RADS(85.71%,57.14%,84.85%,58.82%,78.20%).Conclusion:(1)High-frequency ultrasonography was an important basis for differential diagnosis of benign and malignant thyroid nodules.The diagnosis value of microcalcification in small nodules with blurred boundary was the highest.(2)Traditional TI-RADS classification of thyroid nodules was more standardized.The higher the classification of nodules,the greater the probability of malignant was diagnosised.(3)SWE was an effective method for the diagnosis of benign and malignant thyroid nodules,and SWEmax had the highest diagnostic value.(4)The combination of SWEmax,boundary and microcalcified PRE could improve the sensitivity,specificity,positive predictive value,negative predictive value and accuracy of ultrasound diagnosis of PTMC,It was superior to the traditional TI-RADS classification. |
PDF Full Text Request