| 胡蓉,许诺,曹淑娟,孙洪军,李菲,王艺璇,张佩佩,王蓓.超声联合甲状腺影像报告和数据系统评估甲状腺结节恶性风险[J].中国医学影像技术,2020,36(12):1790~1794 |
| 超声联合甲状腺影像报告和数据系统评估甲状腺结节恶性风险 |
| Ultrasound combined with thyroid imaging reporting and data system for assessment of malignant risk of thyroid nodules |
| 投稿时间：2020-01-07 修订日期：2020-12-04 |
| DOI：10.13929/j.issn.1003-3289.2020.12.007 |
| 中文关键词: 甲状腺结节 超声检查 甲状腺影像报告和数据系统 |
| 英文关键词:thyroid nodule ultrasonography thyroid imaging reporting and data system |
| 基金项目:山东省医药卫生科技发展计划项目（2015WS0232）。 |
|
| 摘要点击次数: 1267 |
| 全文下载次数: 657 |
| 中文摘要: |
| 目的 对比观察超声联合我国国家卫生健康委员会（NHC）甲状腺影像报告和数据系统（TI-RADS）与美国放射学会（ACR）TI-RADS判断甲状腺良恶性结节的价值。方法 回顾性分析132例甲状腺结节患者共173个结节，评价结节超声特征及其NHC TI-RADS、ACR TI-RADS分类；以病理结果为金标准，采用受试者工作特征（ROC）曲线比较2种方法的诊断效能。根据结节最大径将其分为≤1 cm组、>1 cm且≤2 cm组和>2 cm组，以多元Logistic回归分析甲状腺良恶性结节的预测因素。结果 NHC与ACR TI-RADS分类ROC曲线下面积（AUC）分别为0.91、0.92，最佳截断值分别为≥4c类、≥5类，敏感度、特异度、阳性预测值（PPV）、阴性预测值（NPV）及准确率分别为87.36%、88.37%、88.37%、87.36%、87.86%和83.91%、89.53%、89.02%、84.62%、86.71%（P均>0.05），一致性较好（Kappa=0.86）。≤1 cm组纵横比>1时恶性风险高，OR为12.07[95% CI（2.49，58.63）]；>1 cm且≤2 cm组纵横比>1、存在微钙化和不规则或分叶时恶性风险高，OR分别为21.00[95% CI（2.31，191.17）]、14.88[95% CI（3.14，70.50）]及48.75[95% CI（5.99，396.50）]；>2 cm组存在微钙化、不规则或分叶和甲状腺外侵犯时恶性风险高，OR分别为51.33[95% CI（7.24，364.03）]、18.50[95% CI（1.36，252.27）]及27.75[95% CI（2.31，333.76）]。结论 NHC与ACR TI-RADS分类诊断甲状腺结节均有较高效能；不同超声征象对不同大小甲状腺结节的诊断效能不同。 |
| 英文摘要: |
| Objective To compare the diagnostic value of ultrasound combined with thyroid imaging reporting and data system (TI-RADS) issued by National Health Commission (NHC TI-RADS) of People's Republic of China and American College of Radiology (ACR TI-RADS) in differentiating benign and malignant thyroid nodules. Methods Totally 173 thyroid nodules in 132 patients were retrospectively analyzed. Ultrasonic characteristics of the nodule were evaluated, and then the nodules were classified with NHC TI-RADS and ACR TI-RADS, respectively. Taken pathology as the gold standard, the receiver operating characteristic (ROC) curve was used to compare the diagnostic efficacy of NHC TI-RADS and ACR TI-RADS. Then the nodules were divided into ≤1 cm group, >1 cm and ≤2 cm group and >2 cm group according to the maximum diameters. Multivariate Logistic regression was used to analyze the predictive factors of benign and malignant thyroid nodules in 3 groups. Results The area under the curve (AUC) of ROC of NHC TI-RADS and ACR TI-RADS was 0.91 and 0.92, respectively, and the best cutoff value of NHC TI-RADS and ACR TI-RADS was ≥4c and ≥5, respectively. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy of NHC TI-RADS and ACR TI-RADS were 87.36%, 88.37%, 88.37%, 87.36%, 87.86% and 83.91%, 89.53%, 89.02%, 84.62%, 86.71%, respectively (all P>0.05),with good consistency (Kappa=0.86). In ≤1 cm group, taller-than-wide indicated high risk of malignancy, OR was 12.07 (95% CI[2.49,58.63]). In>1 cm and ≤2 cm group, taller-than-wide, microcalcification and irregular or lobulated indicted high risk of malignancy, OR were 21.00 (95% CI[2.31,191.17]), 14.88 (95% CI[3.14,70.50]) and 48.75 (95% CI[5.99,396.50]), respectively. For >2 cm nodules, microcalcification, irregularity or lobulated and extra-thyroidal extension predicted high risk of malignancy, OR were 51.33 (95% CI[7.24,364.03]), 18.50 (95% CI[1.36,252.27]) and 27.75 (95% CI[2.31,333.76]), respectively. Conclusion Both NHC and ACR TI-RADS present high diagnostic efficiency for differentiating benign and malignant thyroid nodules. Different ultrasonic signs had different diagnostic efficacies different size thyroid nodules. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
