摘要
目的探讨MRI特征在术前预测甲状腺乳头状癌BRAFV600E突变中的价值。资料与方法回顾性分析2017年8月—2021年3月复旦大学附属闵行医院经手术病理证实为甲状腺乳头状癌并接受MRI检查及BRAFV600E突变检测的67例患者,分析患者年龄、性别、肿瘤部位、大小、数量、淋巴结转移、T2WI低信号、扩散加权成像信号均匀度、表观扩散系数(ADC)信号、ADC平均值和最小值、早期强化程度、延迟强化形态、强化曲线类型和轮廓征。对MRI特征进行Logistic回归分析BRAFV600E突变的预测因素,计算单因素分析中P<0.05的MRI征象及不同组合预测甲状腺乳头状癌BRAFV600E突变的诊断效能,并采用受试者工作特征曲线分析诊断效能。结果67例患者中,51例(76.1%)BRAFV600E突变阳性。ADC信号(χ^(2)=8.881,P=0.005)、延迟强化形态(χ^(2)=42.846,P<0.001)、强化曲线类型(χ^(2)=7.825,P=0.047)在BRAFV600E突变组和野生组间差异有统计学意义,其中延迟强化形态联合强化曲线类型的特异度及阳性预测值最高(分别为93.8%、96.8%),延迟强化形态的敏感度、阴性预测值、准确度最高(分别为94.1%、82.4%、92.5%)。多因素Logistic回归表明,延迟强化形态是甲状腺乳头状癌BRAFV600E突变的独立预测因素(OR=112,P<0.001),其曲线下面积为0.91。结论MRI表现特征ADC信号、延迟强化形态、强化曲线类型有助于术前预测甲状腺乳头状癌BRAFV600E突变。
Purpose To investigate the value of MRI features in predicting BRAFV600E mutation in papillary thyroid carcinoma.Materials and Methods Retrospective analysis was performed on 67 patients with surgically and pathologically confirmed thyroid papillary carcinoma who underwent MRI and BRAFV600E mutation detection from August 2017 to March 2021 in Minhang Hospital Fudan University.Qualitative and quantitative MRI and clinical features were assessed,including age,gender,tumor site,size,number,lymph node metastasis,low signal intensity on T2 weighted imaging,diffusion weighed imaging signal uniformity,apparent diffusion coefficient(ADC)value,the average and minimum value of ADC,degree of early enhancement,delayed enhancement pattern,enhancement curve type,and contour sign.Logistic regression analysis was performed on MRI characteristics to find the risk factors of BRAFV600E mutation,and the diagnostic efficacy of MRI signs with P<0.05 in univariate analysis and different combinations in predicting BRAFV600E mutation in papillary thyroid carcinoma was further calculated.The receiver operating characteristic curve was used to evaluate the diagnostic efficiency and calculate the area under the curve.Results Among 67 patients with papillary thyroid carcinoma,51(76.1%)were positive for BRAFV600E mutation.ADC signal intensity(χ^(2)=8.881,P=0.005),delayed enhancement pattern(χ^(2)=42.846,P<0.001)and enhancement curve type(χ^(2)=7.825,P=0.047)had statistically significant differences between the BRAFV600E mutation group and the wild group.The combination of delayed enhancement pattern and enhancement curve type had the highest specificity and positive predictive value of 93.8%and 96.8%,respectively;while the delayed enhancement pattern has the highest sensitivity,negative predictive value and overall accuracy rate of 94.1%,82.4%and 92.5%,respectively.Multivariate Logistic regression showed that only delayed enhancement pattern(OR=112,P<0.001)was the independent predictor of BRAFV600E mutation in papillary thyroid carcinoma,w
作者
郑婷婷
徐晶晶
谢晓利
宋彬
ZHENG Tingting;XU Jingjing;XIE Xiaoli;SONG Bin(Department of Radiology,Minhang Hospital,Fudan University,Shanghai 201100,China;不详)
出处
《中国医学影像学杂志》
CSCD
北大核心
2023年第3期220-225,共6页
Chinese Journal of Medical Imaging
基金
上海市科学技术委员会自然科学基金(19ZR1446200)
闵行区科学技术委员会自然科学基金(2019MHZ120)。
