Font Size: a A A

The Study Of Preoperative Diagnosis And Partly Prognosis Evaluation Factors In Thyriod Carcinoma

Posted on:2020-06-04Degree:DoctorType:Dissertation
Country:ChinaCandidate:F LiFull Text:PDF
GTID:1364330575962981Subject:Surgery
Abstract/Summary:PDF Full Text Request
Purpose1.To explore the value of preoperative diagnosis with quantified CEUS combined TI-RADS in thyroid carcinoma.2.To explore the value of preoperative diagnosis with multiparameters of DECT in thyroid carcinoma.3.To identify the value of preoperative diagnosis with ADC value in thyroid carcinoma and explore the differential diagnosis of ADC value in benign and malignant thyroid nodules based on TI-RADS.4.To explore the value of FNAB guided by quantified CEUS combined TI-RADS for preoperative diagnosis in thyroid carcinoma and relationship between US features,CEUS score and Bethesda classification.5.To explore the predictive value of different imaging features and parameters for lymph node metastasis and extrathyroidal extension in thyroid carcinoma.6.To construct ceRNA network of PTC and explore the value of preoperative diagnosis and prognostic evaluation with miRNAs in PTC.Methods1.To analyze the value of preoperative diagnosis with quantified CEUS combined TI-RADS in thyroid carcinoma by coparing US features and CEUS score in 209 benign and malignant thyroid nodules.2.To analyze the value of preoperative diagnosis with DECT,by comparing multiparameters of DECT in 157 benign and malignant thyroid nodules.3.To analyze the value of preoperative diagnosis with ADC value and the ability of differential diagnosis for benign and malignant thyroid nodules with ADC value based on TI-RADS.4.To analyze the distribution in Bethesda classification and the value of cytopathologic diagnosis in thyroid noldules with US features and CEUS score,by comparing the cytopathologic accuracy of FNAB guided by quantified TI-RADS±CEUS and quantified CEUS+TI-RADS in 188 thyroid nodules.5.To construct logstic regression models based on features and parameters of imaging for lymph node metastasis and extrathyroidal extension in thyroid carcinoma,by analysis of relationship between US features,DECT parameters,DWI ADC value and lymph node metastasis,extrathyroidal extension in thyroid carcinoma of part 1 to 3.6.The network of ceRNA in PTC was constructed based on TCGA,miRcode and miRTarBase.The relative expression of miRNAs were analyzed in PTC and benign thyroid diseases,different prognostic factors in PTC by RT-QPCR.Results1.There was significant difference between benign and malignant thyroid nodules in solid,hypoechoic,irregular,taller than wide,punctate echogenic foci and CEUS score(P<0.001).The areas under the curve(AUC)of TI-RADS and quantified CEUS+TI-RADS for differentiating benign from malignant thyroid nodules were 0.820,0.904 respectively,quantified CEUS+TI-RADS was superior to TI-RADS in preoperative diagnosis of thyroid carcinoma(P<0.05).2.IC,NIC1 and NIC2 of thyroid carcinoma were significantly lower than benign thyroid nodules at plain,arterial and venous scan phases(P<0.05),the best threshold of NIC1aa was 0.690I(AUC 0.949);different CT value,λHU1U1 andλHU2with 40keV,70keV,100keV in thyroid carcinoma were significantly lower than in benign thyroid nodules at the three phases(P<0.05),the best threshold of CT value in 70keVa was 125.05 HU.AUC of TI-RADS was 0.911,which was inferior to the optimal parameters of DECT.3.Mean ADC value of thyroid carcinoma was significantly lower than benign thyroid nodules(P<0.001),the best threshold was 1057.1×10-6mm2/s(AUC0.731);The best cut-off of TI-RADS was 4.5(AUC 0.892).ADC value could further differentiate benign from malignant thyroid nodules in TI-RADS4a and TI-RADS4b,the values of AUC were 0.716 and 0.798;the best cutoff of ADC values were 1025.2×10-6mm2/s and 1095.2×10-6mm2/s;The sensitivity and specificity were 56.2%,84.6%and 73.8%,87.5%,respectively(P<0.05).4.The cytopathologic results of FNAB in BethesdaⅡ,BethesdaⅢ/Ⅳand BethesdaⅤ/Ⅵwere significantly different(P<0.001);FNAB guided by quantified CEUS+TI-RADS comparing with TI-RADS±quantified CEUS,the rate of malignancy descended from 28.6%to 26.2%in BethesdaⅡand ascended from 97.8%to 98.7%in BethesdaⅤ/Ⅵ.The distribution of solid,hypoechoic,irregular,Punctate echogenic foci and CEUS score were significantly different in BethesdaⅡ,BethesdaⅢ/Ⅳ,BethesdaⅤ/Ⅵ(P<0.001);which showed positive correlation with Bethesda classification and might differentiate BethesdaⅡfrom BethesdaⅤ/Ⅵ(P<0.05).The sensitivity,specificity of hypoechoic and CEUS score were 81.7%,64.3%and 84.4%,60.7%,respectively,but taller than wide was only positive correlation with Bethesda classification(P<0.05).5.Margin,punctate echogenic foci of US;CT values of 100keVa,100keVv and NIC2aa in DECT;ADC value of DWI could predict lymph node metastasis of thyroid carcinoma.The logstic regression model based on parameters in DECT was superior to the others.CT Value of 100keVv in DECT may predict extrathyroidal extension of thyroid carcinoma.All the logistic regression models were indetified by ROC cures(P<0.05).6.PTC ceRNA network was constructed,including 22 lncRNAs,8 miRNAs and26 mRNAs.The relative expression of miRNA-221,miRNA-21,miRNA-31 and miRNA-375 was significantly different in beingn and malignant thyroid diseases by RT-QPCR(P<0.05).The higher expression was found in patients less than 45years old and with extrathyroidal extension of PTC,which was statistic difference(P<0.05).Conclusions1.Quantified CEUS+TI-RADS improved the system of malignancy risk score in thyroid nodules and the accuracy of preoperative diagnosis in thyroid carcinoma,which decreased omission diagnosis in thyroid carcinoma and misdiagnosis in benign thyroid nodules.2.DECT provided a series of VMI for clearly displaying the lesions without increasing radiation dose.IC,NIC and CT values,λHUU in 40keV、70keV、100keV were reliable parameters for differentiating benign from malignant thyroid nodules.Multiparamteres of DECT were superior to TI-RADS in preoperative diagnosis of thyroid carcinoma.3.The ADC value and TI-RADS were both reliable methods for preoperative diagnosis in thyroid carcinoma.TI-RADS was superior to the ADC value in preoperative diagnosis of thyroid carcinoma,but the ADC value might further differentiate benign from malignant thyroid nodules in TI-RADS4a and TI-RADS4b,which were difficultly diagnosed by US.4.FNAB guided by quantified CEUS+TI-RADS was an effecitive method to improve the quality of cellular specimen in benign and malignant thyroid nodules.US features including CUES were used to interpretate FNAB result,which was a feasible method,and the method improved the accuracy of preoperative diagnosis in thyroid carcinoma.5.The logstic regression models for predicting lymph node metastasis and extrathyroidal extension in thyroid carcinoma based on different imaging features and parameters were constructed,which could better predict lymph node metastasis and extrathyroidal exstension before operation.The logstic regression models provided objective proof for preoperative TNM stage and assited in selecting reasonable operation.6.Construction of ceRNA network and preliminary exploration of miRNAs in preoperative diagnosis and prognostic evaluation of PTC assited in improving the system of preoperative diagnosis and prognosis evaluation in thyroid carcinoma.In conclusion,the system of preoperative disgnosis was constructed in our study,including US、CEUS、DECT、DWI ADC value and FNAB;logstic regression models based on features and parameters of imaging for predicting lymph node metastasis and extrathyroidal extension before operation were constructed;we constructed PTC ceRNA network based on TCGA and explored the value of miRNAs for preoperative diagnosis and prognosis evaluation in PTC.From clinical practice to fundamental research,we systematically studied the preoperative diagnosis and partly prognostic factors before operation.
Keywords/Search Tags:Thyroid carcinoma, imaging, FNAB, miRNAs, diagnosis, prognosis
PDF Full Text Request
Related items