超声简化型甲状腺影像报告和数据系统在甲状腺结节诊疗中的应用

Simplified thyroid imaging reporting and data system for diagnosis of thyroid nodules

目的 探讨超声简化型甲状腺影像报告和数据系统（simplified thyroid imaging reporting and date system，S-TIRADS）在甲状腺结节诊疗中的应用价值。方法 回顾文献报道的甲状腺结节超声良恶性指标对甲状腺结节进行分级，建立S-TIRADS。由3名不同年资的超声医生分别对490个甲状腺结节分级，计算S-TIRADS不同分级结节的恶性率，对不同年资医生之间进行一致性检验，并利用敏感性、特异性、阳性预测值（positive predictive value，PPV）、阴性预测值（negative predictive value，NPV）、准确性及受试者操作特性（ROC）曲线评价S-TIRADS。结果 S—TIRADS 1～5级的恶性率平均值为0％、4．09％、25．44％、64．86％及100％。不同年资超声医生之间一致性Kappa检验k值分别为0．63、0．69及0．63。诊断敏感性、特异性、PPV、NPV及准确性在高年资可达94．11％、78．35％、53．33％、98．06％及81．63％，不同年资超声医生ROC曲线下面积分别为0．918、0．877及0．905。结论 S-TIRADS是一种简化且重复性较好的影像报告和数据系统，可作为不同年资、不同学科、不同诊疗中心医生之间关于甲状腺结节性质的交流工具。

Objective To explore the value of simplified thyroid imaging reporting and data system（ S-TIRADS） for diagnosis of thyroid nodules. Methods A total of 490 nodules were included in the study. Their literature were reviewed to establish S-TIRADS based on malignant features. Three sonographers with different years of experience evaluated 490 nodules independently. The rate of malignancy was calculated for each grade. Inter-observer agreement was calculated using the kappa statistic. Sensitivity, specificity, positive predictive value （PPV）, negative predictive value（ NPV）, accuracy and receiver-operating characteristic curve（ ROC curve）were used to assess the value of S-TIRADS. Results The average malignancy rate for each grade of S-TIRADS was 0%, 4. 09%, 25.44%, 64. 86%, and 100% respectively. The kappa values between radiologists with different experience were 0. 63, 0. 69, and 0. 63. Sensitivity, specificity, PPV, NPV, and accuracy were 94. 11%, 78.35%, 53. 33%, 98.06%, and 81.63% for experienced radiologists. The area under the ROC curve was 0. 918, 0. 877, and 0. 905 respectively. Conclusion S-TIRADS is a simplified and repeatable imaging reporting and date system, which can be used as a communication tool of thyroid nodules between different physicians.