MRI多参数联合评分对腮腺肿瘤的定性诊断价值  

The qualitative diagnostic value of MRI multiparameter combined scoring for the diagnosis of parotid gland tumors

作　　者：李治群[1] 孙笑芬 徐钐 万江花[1] 施玉森[1] 李蕊蕊 刘旭东[1] 张业雨 LI Zhi-qun;SUN Xiao-fen;XU Shan(Department of Radiology,First Affiliated Hospital of Hainan Medical University,Haikou 570102,China)

机构地区：[1]海南医学院第一附属医院放射科,海口570102

出　　处：《放射学实践》2023年第9期1114-1120,共7页Radiologic Practice

基　　金：海南省自然科学基金(编号:822RC824)。

摘　　要：目的:探讨MRI多参数联合评分对腮腺肿瘤的定性诊断价值。方法:搜集2016年1月至2022年12月间我院收治的经手术病理确诊的腮腺肿瘤患者,所有患者术前均接受过多模态MRI检查,包括MRI平扫、扩散加权成像(DWI)和动态增强扫描(DCE),采用ROC曲线分析MRI多参数联合评分(MPCS)鉴别腮腺良、恶性肿瘤的效能。结果:共纳入腮腺肿瘤患者116例,其中良性肿瘤75例(64.7%),恶性肿瘤41例(35.3%)。MRI平扫图像中良、恶性肿瘤的边界和形态差异有统计学意义(P<0.05),但病灶位置和信号均匀性差异无统计学意义(P>0.05)。良性肿瘤的平均ADC值为(1.15±0.22)×10^(-3)mm^(2)/s,恶性肿瘤的平均ADC值为(0.95±0.17)×10^(-3)mm^(2)/s,两者之间差异有统计学意义(P<0.05)。良性肿瘤的TIC曲线以A型最多(43/75,57.3%),恶性肿瘤则以C型曲线最多(29/41,70.7%)。良性肿瘤的MR多参数联合评分平均分值为(2.23±1.23)分,恶性肿瘤为(5.00±1.05)分,两者之间差异有统计学意义(P<0.05)。ROC曲线分析结果显示,MPCS鉴别诊断腮腺良、恶性肿瘤的曲线下面积(AUC)为0.94,明显高于ADC值(0.75)和TIC曲线分型(0.84),差异有统计学意义(P<0.01)。MPCS为3分时鉴别良恶性肿瘤的效能最高,AUC为0.88,敏感度和特异度分别为87.8%和88.0%,阳性预测值和阴性预测值分别为80.0%和93.0%,准确率为87.9%,约登指数为0.76。结论:MRI多参数联合评分可有效鉴别腮腺良、恶性肿瘤,是腮腺肿瘤术前定性评估的重要检查手段。Objective:The purpose of this study was to explore the qualitative diagnostic value of MRI multi-parameter combined scoring for the diagnosis of parotid gland tumors.Methods:We collected patients with tumors in the parotid gland confirmed by surgery and pathology admitted to our hospital from January 2016 to December 2022.All patients underwent multi-modal MR examination before surgery,including plain MRI,diffusion weighted imaging(DWI),and dynamic contrast-enhanced imaging(DCE).The diagnostic efficacy of MR multi-parameter combined scoring(MPCS)in distinguishing benign and malignant parotid gland tumors was analyzed by using the ROC curve model.Results:A total of 116 patients with parotid gland tumors were collected,including 75 benign tumors(64.7%)and 41 malignant tumors(35.3%).There were statistical differences in the boundaries and morphology between benign and malignant tumors in MR plain scan images,but there was no statistical difference in the lesion location and signal uniformity between them.The average ADC value of benign tumors is(1.15±0.22)×10^(-3)mm^(2)/s,with an average ADC value of(0.95±0.17)×10^(-3)mm^(2)/s for malignant tumors,there was a statistical difference between the two.The TIC curve of benign tumors was most common in type A(43/75,57.3%),while that of malignant tumors is most common in type C(29/41,70.7%).The average score of the MR multi-parameter joint score for benign tumors was 2.23±1.23,while the average score for malignant tumors was 5.00±1.05.There was a statistical difference between the two(P<0.05).ROC curve analysis showed that the AUC of MPCS was 0.94,significantly higher than the ADC value(0.75)and TIC curve typing(0.84)(P<0.01).The diagnostic efficacy of distinguishing between benign and malignant tumors was the highest when the cutoff value was greater than 3 minutes,with an area under the ROC curve of 0.88,sensitivity and specificity of 87.8%and 88.0%,positive predictive value and negative predictive value of 80.0%and 93.0%,accuracy of 87.9%,and Jordan index of 0.76.

关 键 词：腮腺肿瘤 磁共振成像 扩散加权成像 动态增强 诊断 鉴别 

分 类 号：R739.8[医药卫生—肿瘤] R445.2[医药卫生—临床医学]