摘要
目的探讨血清促甲状腺素受体抗体(thyrotropin receptor antibody,TRAb)对格雷夫斯病患者^(131)I治疗后永久性甲状腺功能减退(hypothyroidism,HT)的影响及其预测价值。方法选取2013年10月至2021年10月在华北理工大学附属医院行^(131)I治疗的格雷夫斯病患者479例,其中随访发生永久性HT 350例(永久性HT组),未发生永久性HT 129例(非永久性HT组)。收集两组患者的性别、年龄、甲状腺肿大程度等一般临床资料以及治疗前24 h摄碘率、甲状腺功能和治疗剂量等指标。比较两组患者的一般临床资料和治疗前各检测指标水平,分析格雷夫斯病患者^(131)I治疗后发生永久性HT的危险因素及治疗前血清TRAb水平的预测价值,评价预测结果与临床诊断的一致性。呈正态分布的计量资料以x±s表示,两组间比较采用两独立样本t检验;不符合正态分布的计量资料以M(Q_(1),Q_(3))表示,两组比较采用Wilcoxon秩和检验;计数资料以例(%)表示,两组间比较采用四格表χ^(2)检验,等级资料比较采用Wilcoxon秩和检验;多因素分析采用Logistic回归模型。绘制受试者工作特征曲线评估TRAb的预测效能,并根据查约登指数筛选预测临界值;计算敏感度和特异度,计算Kappa值评价预测结果与临床诊断的一致性。结果永久性HT组与非永久性HT组患者的性别、年龄、甲状腺肿大程度、24 h摄碘率和血清促甲状腺激素、游离三碘甲状腺原氨酸、游离甲状腺素水平以及^(131)I治疗剂量比较差异均无统计学意义(均P>0.05);永久性HT组患者治疗前TRAb水平高于非永久性HT组[14.51(4.95,33.58)U/L比3.40(1.67,16.50)U/L],差异有统计学意义(Z=5.87、P<0.001)。多因素Logistic回归分析结果表明,治疗前TRAb水平是^(131)I治疗后格雷夫斯病患者发生永久性HT的危险因素(比值比=1.042,95%CI:1.025~1.059,P<0.001)。治疗前TRAb水平预测格雷夫斯病患者^(131)I治疗后发生永久性HT的受试者工作�
Objective The effect of serum thyrotropin receptor antibody(TRAb)on permanent hypothyroidism(HT)after^(131)I treatment in Graves disease patients and its predictive value Methods 479 Graves'disease patients who underwent^(131)I treatment at the Nuclear Medicine Department of North China University of Technology Affiliated Hospital from October 2013 to October 2021 were selected.Among them,350 cases(permanent HT group)had permanent HT during follow-up,while 129 cases(non permanent HT group)did not.Collect general clinical data such as gender,age,degree of thyroid enlargement,as well as indicators such as iodine intake rate,thyroid function,and treatment dose 24 hours before treatment,from two groups of patients.Compare the general clinical data and pre treatment levels of various detection indicators between two groups of patients,analyze the risk factors for permanent HT in Graves disease patients after^(131)I treatment,and evaluate the predictive value of pre treatment serum TRAb levels,and evaluate the consistency between the predicted results and clinical diagnosis.The econometric data with a normal distribution is represented by x±s,and two independent sample t-tests are used for comparison between the two groups;The econometric data that do not conform to the normal distribution are represented by M(Q_(1),Q_(3)),and the Wilcoxon rank sum test is used for comparison between the two groups;The counting data is represented as an example(%),and the comparison between the two groups is conducted using the four gridχ^(2)test,while the comparison of grade data is conducted using the Willcoxon rank sum test;The logistic regression model was used for multivariate analysis.Draw a subject work characteristic curve to evaluate the predictive efficacy of TRAb,and screen the predictive threshold based on the Chayoden index;Calculate sensitivity and specificity,and evaluate the consistency between the predicted results and clinical diagnosis by calculating Kappa values.Results There were no statistically significant diff
作者
陈思
张雅淇
宫凤玲
翟宁
于鹏
Chen Si;Zhang Yaqi;Gong Fengling;Zhai Ning;Yu Peng(Graduate School,North China University of Science and Technology,Tangshan 063000,China;CT Room,Affiliated Hospital of North China University of Science and Technology,Tangshan 063000,China;Department of Nuclear Medicine,Affiliated Hospital of North China University of Science and Technology,Tangshan 063000,China)
出处
《中国综合临床》
2023年第6期453-458,共6页
Clinical Medicine of China
基金
河北省医学科学研究课题计划(20221535)。
