TI-RADS分类系统联合超声造影对甲状腺结节良恶性鉴别诊断价值

Comparative studies of thyriod imaging reporting and data system combined with contrast-enhanced ultrasound in differential diagnosis of thyroid nodules

目的评价甲状腺影像报告和数据系统(thyriod imaging reporting and data system,TI-RADS)、超声造影(contrast-enhanced ultrasound,CEUS)及两者联合应用对甲状腺结节良恶性鉴别诊断的价值。方法回顾性分析196例274个甲状腺结节(恶性结节141个、良性结节133个)的常规超声TI-RADS分类、超声造影及两者联合应用结果,将病变分类为良性、可能良性、不能确定、可能恶性及恶性,并分别与手术病理结果对照分析,应用受试者工作特征曲线(ROC曲线)评价三种方法的诊断效果。结果 TI-RADS分类系统和超声造影联合检查诊断甲状腺癌ROC曲线下面积(0.917±0.019)大于单独TI-RADS分类(0.826±0.026)及超声造影(0.851±0.025),且有统计学意义(P=0.000、0.000、0.000)。结论 TI-RADS分类系统和超声造影对甲状腺良恶性结节鉴别诊断具有重要价值,两者联合应用能提高超声对甲状腺癌的诊断准确性。

Objective To assess the value of thyriod imaging reporting and data system（ TI-RADS）,contrast-enhanced ultrasound（ CEUS）,combination of TI-RADS and CEUS in diagnosis of thyroid nodules. Methods The TI-RADS and contrast-enhanced ultrasound signs of 274 thyroid nodules in 196 patients（ 141 malignant,133 benign） were retrospectively analyzed. The lesions were defined as five categories of the benign,probably benign,equivocal,probably malignant and malignant. The results of thyriod imaging reporting and data system,contrast-enhanced ultrasound,combination of TI-RADS and CEUS were compared with pathology. By using the definitive diagnosis and the five level of suspicion categories,receiver operating characteristic（ ROC） curve was drawn for the results of thyriod imaging reporting and data system,contrast-enhanced ultrasound,combination of TI-RADS and contrast-enhanced ultrasound. Results The area under the ROC curve of combination of TI-RADS and CEUS（ 0. 917 ± 0. 019） was larger than that of thyriod imaging reporting and data system（ 0. 826 ± 0. 026） and contrast-enhanced ultrasound（ 0. 851 ± 0. 025）. The increases in the areas were significant（ P = 0. 000,0. 000,0. 000）. Conclusion TI-RADS and contrast-enhanced ultrasound are important in diagnosis of thyroid nodules. Combination of TI-RADS and contrast-enhanced ultrasound results in an increase in diagnostic accuracy of thyroid cancer.