Font Size: a A A

Application Of The Binary Regression Mode In Central Lymph Node Metastasis Of Papillary Thyroid Microcarcinoma

Posted on:2020-01-27Degree:MasterType:Thesis
Country:ChinaCandidate:Q Y SunFull Text:PDF
GTID:2404330590484956Subject:Surgery
Abstract/Summary:PDF Full Text Request
Objectives Analyze the influence factors of central lymph node metastasis(CLNM)in Papillary thyroid microcarcinoma(PTMC),and to explore the correlation between the related factors and CLNM,and to establish a pre-operative risk assessment equation for CLNM.Methods The clinical data of PTMC patients admitted to the Department of Thyroid & Breast of Hebei General Hospital from January 2015 to December 2017 were retrospectively analyzed.Inclusion criteria and exclusion criteria were strictly formulated.Preoperative clinical and ultrasonographic data were collected and analyzed,which include:(1)General information of patients: Including the patient’s number,age,sex,combined with thyroid benign diseases(Hashimoto’s thyroiditis,nodular goiter).(2)Thyroid function test results: The TSH value.(3)Surgical data: Operative methods.(4)Preoperative color Doppler ultrasound features of thyroid and cervical lymph nodes: Tumor diameter,capsular invasion,microcalcification and multiple carcinomas.CLNM was grouped according to whether CLNM occurred or not.SPSS 23.0 statistical software was used for data analysis.Two-Sample t-test or Mann-Whitney U-test were used to compare the mean of measurement data.Chi-Square test Or Fisher’s exact probability method was used to compare the incidence of categorical data.Receiver operating characteristic curve(ROC)was used to determine cut-off for predicting CLNM in PTMC patients.Univariate analysis P the index<0.05 is taken as independent variable and CLNM is taken as dependent variable.The risk probability logistic regression model of CLNM is established by using Binary Logistic regression analysis method.The ROC curve is drawn and the area under the curve is calculated to check the accuracy of the equation.The ultimate model is the one with the highest level of accuracy.Results From the result of this study,210 PTMC patients met the inclusion criteria,53 of them had CLNM,and the incidence of CLNM was 25.20%.Logistic regression analysis showed that five variables,including tumor diameter,age,sex,capsular invasion and multiple cancers,entered the regression model.Male(P=0.02,B=1.02,OR=2.76,95%CI:1.16-6.58);Envelope invasion(P<0.01,B=1.60,OR=4.96,95%CI:1.64-14.97);Multiple carcinomas(P<0.01,B=1.62,OR=5.06,95%CI:2.31-11.09);Tumor diameter ≥ 0.65cm(P<0.01,B=1.35,OR=3.86,95%CI:1.79-8.32);Age≥45(P=0.04,B=-0.85,OR=0.43,95% CI: 0.20-0.94).Tumor diameter ≥ 0.65 cm,male,capsular invasion,multiple cancers and age<45 years old were independent risk factors for CLNM.The area under ROC curve of the logistic regression model was 0.85,and the model had high level of accuracy.Conclusions(1)Independent risk factors for CLNM in PTMC patients: Age<45 years old,male,maximum diameter ≥0.65 cm,preoperative ultrasonography showed invasion of the capsule and multiple carcinomas.(2)The equation for predicting CLNM risk before operation is as follows: logit(P)=1.02(male)+1.60(capsular invasion)+1.35(tumor diameter ≥0.65cm)+1.62(multiple cancers)-0.85(age ≥ 45 years)-2.42.Figure2;Table7;Reference 173...
Keywords/Search Tags:Papillary thyroid microcarcinoma, Central lymph node metastasis, Influencing factors
PDF Full Text Request
Related items