OBJECTIVES We aim to compare the diagnostic performance to assess thyroid nodules and reliability for recommending fine needle aspiration biopsy (FNAB) between American College of Radiology thyroid image reporting and data system (ACR TI-RADS) and American Thyroid Association (ATA) guidelines. METHODS In total, this retrospective study included 1001 consecutive thyroid nodules in 918 patients from May 2016 to December 2017. US features of the thyroid nodules, including composition, echogenicity, shape, margins, echogenic foci, and size, were reviewed and were classified according to ACR TI-RADS and ATA guidelines, respectively. The diagnostic performance to assess thyroid nodules and reliability for recommending fine needle aspiration biopsy were compared between ACR TI-RADS and ATA guidelines. RESULTS Of the 1001 thyroid nodules, 609 (60.8%) were benign and 392 (39.2%) were malignant. The sensitivity, specificity, PPV, NPV, and accuracy were 96.7%, 77.3%, 73.3%, 97.3%, and 84.9%, respectively, for ACR TI-RADS and 99.2%, 16.1%, 43.2%, 97.0%, and 48.7%, respectively, for ATA guidelines. AUC of ACR TI-RADS was significantly greater than ATA guidelines (0.935 (0.918, 0.949) vs 0.884 (0.862, 0.903), p < 0.001). Biopsy yield of malignancy, biopsy rate of malignancy, and unnecessary FNAB rate were 59.5%, 91.3%, and 40.5%, respectively, for ACR TI-RDS and 38.5%, 97.4%, and 61.5%, respectively, for ATA guidelines. CONCLUSIONS ACR TI-RADS was more accurate than ATA guidelines for differentiating malignant thyroid nodules from benign nodules and more reliable than ATA guidelines for recommending thyroid nodules for FNAB. KEY POINTS • Malignant risk of thyroid nodules can be stratified by ultrasound. • American College of Radiology guidelines were more accurate for differentiating malignant thyroid nodules from benign nodules. • American College of Radiology guidelines were more reliable for recommending thyroid nodules for biopsy.

中文翻译：

---

甲状腺结节细针穿刺活检的指征：将基于点的风险分层系统与基于模式的风险分层系统进行比较。

目的我们旨在比较美国放射学院甲状腺影像报告和数据系统（ACR TI-RADS）与美国甲状腺协会（ATA）指南之间的诊断性能，以评估甲状腺结节和推荐细针穿刺活检（FNAB）的可靠性。方法这项回顾性研究共纳入了2016年5月至2017年12月期间918例患者的1001个连续甲状腺结节。对甲状腺结节的美国特征（包括成分，回声性，形状，切缘，回声灶和大小）进行了回顾和分类。分别遵守ACR TI-RADS和ATA指南。在ACR TI-RADS和ATA指南之间比较了评估甲状腺结节的诊断性能和推荐细针穿刺活检的可靠性。结果1001个甲状腺结节中有609个（60。良性为8％），恶性为392（39.2％）。ACR TI-RADS的敏感性，特异性，PPV，NPV和准确度分别为96.7％，77.3％，73.3％，97.3％和84.9％，分别为99.2％，16.1％，43.2％，97.0％和48.7 ％，分别用于ATA准则。ACR TI-RADS的AUC显着高于ATA指南（0.935（0.918，0.949）vs 0.884（0.862，0.903），p <0.001）。对于ACR TI-RDS，恶性肿瘤的活检率，恶性肿瘤的活检率和不必要的FNAB率分别为ATA准则的38.5％，91.3％和40.5％，对于ATA指南则分别为38.5％，97.4％和61.5％。结论ACR TI-RADS在区分恶性甲状腺结节和良性结节方面比ATA指南更准确，并且比ATA指南为FNAB推荐甲状腺结节更可靠。要点•超声检查可将甲状腺结节的恶性风险分层。•美国放射科学院的指南对于区分恶性甲状腺结节和良性结节更为准确。•美国放射学院的指南在推荐甲状腺结节进行活检时更为可靠。