目的:探讨分化型甲状腺癌（DTC）患者术后n 131I治疗前甲状腺球蛋白抗体（TgAb）水平及n 131I治疗后TgAb的变化趋势对n 131I治疗效果的预测作用。n 方法:回顾性分析2016年12月至2018年5月于华中科技大学同济医学院附属同济医院接受甲状腺全切除术及n 131I治疗的TgAb阳性DTC患者95例，其中男性16例、女性79例，年龄11~63 (37.74±11.27）岁。所有患者均在手术后2~6个月进行n 131I治疗。根据美国甲状腺协会指南中DTC治疗反应评估体系及影像学表现，按照疗效评价结果将患者分为影像学表现满意组（A组，54例）、影像学表现不满意组（B组，19例）、影像学表现不确切组（C组，22例）。观察3组患者n 131I治疗前TgAb水平、治疗后1.5、3~6、6~12个月的TgAb变化率及变化速率等指标，并对其可能影响疗效的因素进行分析。以治疗前TgAb水平、治疗后1.5个月和3~6个月TgAb变化率绘制受试者工作特征（ROC）曲线，确定其预测疗效的最佳临界值及诊断效能。计量资料的组间比较采用方差分析或Kruskal-Wallis秩和检验；计数资料的组间比较采用卡方检验或Fisher确切概率法。对单因素分析中有统计学意义的变量采用多因素Logistic回归分析。n 结果:单因素分析结果显示，3组DTC患者在年龄（n F=3.48, n P=0.03）、淋巴结转移率（n χ2=10.28，n P<0.01）、治疗前TgAb水平（n χ2= 16.03，n P<0.01）、n 131I剂量（Fisher确切概率法，n P=0.002）和治疗后1.5、3~6、6~12个月复查时的TgAb变化率及变化速率（6~12个月变化速率除外）的差异均有统计学意义（n χ2=8.55~22.79，均n P<0.05）。多因素Logistic回归分析显示，3组DTC患者在治疗前TgAb水平（n β=-0.006，n OR=0.994，n P<0.01）和治疗后1.5个月TgAb变化率（n β=0.327，n OR=1.387，n P=0.046）、3~6个月TgAb变化率（n β=2.521，n OR=12.439，n P=0.041）的差异均有统计学意义。ROC曲线分析结果显示，治疗前TgAb水平预测n 131I疗效的最佳临界值为201.4 U/mL，灵敏度和特异度分别为74.1%、68.3%，曲线下面积为0.740；治疗后1.5个月和3~6个月TgAb变化率预测n 131I治疗效果的最佳临界值分别为9.7%和19.2%，灵敏度和特异度分别为72.2%、61.0%和87.0%、58.5%，曲线下面积分别为0.660和0.752。n 结论:DTC患者n 131I治疗前TgAb水平可以预测n 131I治疗效果，TgAb水平较低者疗效较好；n 131I治疗后TgAb水平下降越显著，提示治疗效果越好。n “,”Objective:To investigate the predictive effect of thyroglobulin antibody (TgAb) level after operation and the trend of TgAb change after n 131I treatment on the curative effect of n 131I treatment in patients with differentiated thyroid carcinoma (DTC).n Methods:A total of 95 patients with DTC who were positive for TgAb and who underwent total thyroidectomy and n 131I treatment in the Tongji Hospital of Tongji Medical College of Huazhong University of Science and Technology from December 2016 to May 2018 were retrospectively analyzed. The patients included 16 males and 79 females, aged 11-63 (37.74±11.27) years old. n 131I treatment was administered 2-6 months after surgery. Based on the response to therapy assessment system of DTC in the American Thyroid Association Management Guidelines and according to the images, the curative effect was divided into three groups, namely, structural excellent response group (A group, 54 cases), structural incomplete response group (B group, 19 cases), and structural indeterminate response group (C group, 22 cases). The following parameters were observed among the three groups: TgAb level before n 131I treatment; percentage change at 1.5 months, 3-6 months, and 6-12 months; and rate of change of TgAb. The factors that might affect the curative effect were analyzed. Receiver operating characteristic (ROC) curves were drawn with the following: the TgAb level before n 131I treatment; percentage change of TgAb at 1.5 months after treatment; and percentage change of TgAb at 3-6 months after treatment. The best critical value for predicting therapeutic effect was determined, and the diagnostic efficiency was calculated. The measurement data were compared through variance analysis or Kruskal-Wallis rank sum test. n Chi-square test or Fisher\'s exact text were used to compare the count data among the groups. The multivariate Logistic regression analysis was used to analyze the factors that may affect the curative effect.n Results:The univariate regression analysis results showed that the mean age (n F=3.48, n P=0.03), lymph node metastasis rate (n χ2=10.28, n P<0.01), TgAb level beforen 131I treatment (n χ2=16.03, n P<0.01),n 131I dose (Fisher\'s exact text,n P=0.002), percentage changeand and rate of change of TgAb (except percentage change at 6-12 months) at 1.5 months, 3-6 months, and 6-12 months were significantly different among the three groups (n χ2=8.55-22.79, all n P<0.05). The multivariate Logistic regression analysis result showed that TgAb levels beforen 131I treatment were significantly different among the three groups (n β=-0.006, n OR=0.994; n P<0.01), percentage change of TgAb at 1.5 months (n β=0.327, n OR=1.387; n P=0.046), and percentage change of TgAb at 3-6 months (n β=2.521, n OR=12.439; n P=0.041) had statistical significance. The optimal cut-off value of TgAb level on ROC curve was 201.4 U/mL with a sensitivity of 74.1% and specificity of 68.3%. The area under ROC curve (AUC) was 0.740. The optimal cut-off values of the percentage change of TgAb in 1.5 months and for 3-6 months after treatment were 9.7% and 19.2%; the sensitivities were 72.2% and 61.0%; the specificities were 87.0% and 58.5%; and the AUCs were 0.660 and 0.752, respectively.n Conclusions:Postoperative TgAb level enabled the prediction of the therapeutic effect of n 131I. Patients with lower TgAb level showed better therapeutic effect. The decrease of TgAb after treatment was more significant than before treatment, suggesting that the n 131I therapeutic effect was satisfactory.n