Zhao Rui-Na, Zhang Bo, Yang Xiao, Jiang Yu-Xin, Lai Xing-Jian, Zhang Xiao-Yan
Department of Ultrasound, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China.
Department of Ultrasound, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China.
Ultrasound Med Biol. 2015 Dec;41(12):3102-8. doi: 10.1016/j.ultrasmedbio.2015.04.026. Epub 2015 Sep 28.
The purpose of the study described here was to determine specific characteristics of thyroid microcarcinoma (TMC) and explore the value of contrast-enhanced ultrasound (CEUS) combined with conventional ultrasound (US) in the diagnosis of TMC. Characteristics of 63 patients with TMC and 39 with benign sub-centimeter thyroid nodules were retrospectively analyzed. Multivariate logistic regression analysis was performed to determine independent risk factors. Four variables were included in the logistic regression models: age, shape, blood flow distribution and enhancement pattern. The area under the receiver operating characteristic curve was 0.919. With 0.113 selected as the cutoff value, sensitivity, specificity, positive predictive value, negative predictive value and accuracy were 90.5%, 82.1%, 89.1%, 84.2% and 87.3%, respectively. Independent risk factors for TMC determined with the combination of CEUS and conventional US were age, shape, blood flow distribution and enhancement pattern. Age was negatively correlated with malignancy, whereas shape, blood flow distribution and enhancement pattern were positively correlated. The logistic regression model involving CEUS and conventional US was found to be effective in the diagnosis of sub-centimeter thyroid nodules.
本研究的目的是确定甲状腺微小癌(TMC)的具体特征,并探讨超声造影(CEUS)联合传统超声(US)在TMC诊断中的价值。回顾性分析了63例TMC患者和39例甲状腺亚厘米级良性结节患者的特征。进行多因素逻辑回归分析以确定独立危险因素。逻辑回归模型纳入了四个变量:年龄、形态、血流分布和增强模式。受试者操作特征曲线下面积为0.919。以0.113作为截断值时,灵敏度、特异度、阳性预测值、阴性预测值和准确度分别为90.5%、82.1%、89.1%、84.2%和87.3%。CEUS联合传统US确定的TMC独立危险因素为年龄、形态、血流分布和增强模式。年龄与恶性程度呈负相关,而形态、血流分布和增强模式与恶性程度呈正相关。发现涉及CEUS和传统US的逻辑回归模型在甲状腺亚厘米级结节的诊断中有效。