Endocr Pract. 2019 May;25(5):413-422. doi: 10.4158/EP-2018-0369. Epub 2019 Jan 18.
This study compares the American Thyroid Association (ATA) classification system with the 2017 American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI-RADS) for predicting cancer risk in thyroid nodules. This is a retrospective review of ultrasound imaging of all adult patients with thyroid nodules >5 mm who underwent thyroidectomy at a tertiary care hospital in 2016. We assessed the ability of either system to predict malignancy based on surgical histopathology. Sensitivity, specificity, negative predictive values (NPV) and positive predictive values (PPV), and area-under-the-curve (AUC) were calculated and compared using McNemar's, Fisher exact, or DeLong's tests. Three hundred and twenty-three nodules from 213 adults were included. Median patient age was 55 years; 75.6% were female. 27.2% nodules were malignant. Both ATA and ACR TI-RADS provide effective diagnostic performance, a sensitivity of 77.3% versus 78.4%, respectively, a specificity of 76.6% versus 73.2%, respectively, a PPV of 55.3% versus 52.3%, respectively, and a NPV of 90% for both. The level of agreement between the two classification systems was almost perfect (weighted Kappa statistic = 0.93, AUC 0.77 ATA versus 0.76 TI-RADS [ = .18]). However, of the 40 (TI-RADS level 3) TR3 nodules (<2.5 cm), 10% were malignant, and of the 31 (TI-RADS level 4) TR4 nodules (<1.5 cm), 38% were malignant. The ATA and TI-RADS classification systems appear to have similar diagnostic value for predicting thyroid cancer. However, subanalysis of TR3 and TR4 nodules with consideration of size criteria showed that there is a higher risk of missing a malignancy if the ACR TI-RADS recommendation is followed. These results should be validated in a different patient cohort with a lower incidence of cancer. = American College of Radiology; = American Thyroid Association; = Fine Needle Aspiration; κ = weighted Kappa statistic; = negative predictive values; = positive predictive values; = Thyroid Imaging Reporting and Data System; = TI-RADS level 1; = TI-RADS level 2; = TI-RADS level 3; = TI-RADS level 4; = TI-RADS level 5.
这项研究比较了美国甲状腺协会（ATA）分类系统与 2017 年美国放射学院（ACR）甲状腺影像报告和数据系统（TI-RADS）在预测甲状腺结节癌症风险方面的表现。这是对 2016 年在一家三级保健医院接受甲状腺切除术的所有>5mm 成人甲状腺结节患者的超声影像学进行的回顾性研究。我们评估了这两种系统基于手术组织病理学预测恶性肿瘤的能力。使用 McNemar 检验、Fisher 确切检验或 DeLong 检验计算并比较了灵敏度、特异性、阴性预测值（NPV）、阳性预测值（PPV）和曲线下面积（AUC）。纳入了 213 例成人的 323 个结节。中位患者年龄为 55 岁，75.6%为女性。27.2%的结节为恶性。ATA 和 ACR TI-RADS 均提供了有效的诊断性能，敏感性分别为 77.3%和 78.4%，特异性分别为 76.6%和 73.2%，PPV 分别为 55.3%和 52.3%，NPV 均为 90%。两种分类系统之间的一致性水平近乎完美（加权 Kappa 统计量=0.93，ATA 的 AUC 为 0.77，TI-RADS 为 0.76[=0.18]）。然而，在 40 个（TI-RADS 级别 3）TR3 结节（<2.5cm）中，有 10%为恶性，在 31 个（TI-RADS 级别 4）TR4 结节（<1.5cm）中，有 38%为恶性。ATA 和 TI-RADS 分类系统在预测甲状腺癌方面似乎具有相似的诊断价值。然而，对考虑大小标准的 TR3 和 TR4 结节进行亚分析表明，如果遵循 ACR TI-RADS 建议，可能会错过恶性肿瘤的风险。这些结果应在癌症发病率较低的不同患者队列中进行验证。 = 美国放射学院； = 美国甲状腺协会； = 细针穿刺抽吸术；κ=加权 Kappa 统计量； = 阴性预测值； = 阳性预测值； = 甲状腺影像报告和数据系统； = TI-RADS 级别 1； = TI-RADS 级别 2； = TI-RADS 级别 3； = TI-RADS 级别 4； = TI-RADS 级别 5。
