DCE-MRI时间-信号曲线及相关定量参数对诊断鼻腔鼻窦肿瘤的价值
作者: |
1邓娟,
1刘进康
1 中南大学湘雅医院放射科,长沙 410008 |
通讯: |
刘进康
Email: 13908482001@139.com |
DOI: | 10.3978/j.issn.2095-6959.2018.04.009 |
摘要
目的:探讨动态增强磁共振(dynamic contrast enhanced magnetic resonance imaging,DCE-MRI)时间-信号曲线(time signal curve,TIC)以及相关定量参数在鼻腔鼻窦肿瘤中的诊断价值。方法:纳入中南大学湘雅医院2015年2月至2017年2月间收治的鼻腔鼻窦肿瘤患者180例,所有患者经病理诊断确诊。于术前接受DCE-MRI扫描,获得TIC曲线及相关定量参数,以手术病理诊断为金标准,分析DCE-MRI的TIC与相关定量参数诊断效果,并绘制受试者工作特征(receiver operating characteristic,ROC)曲线明确各参数预测鼻腔-鼻窦肿瘤的敏感度、特异度,采用Kappa检验分析这种诊断方式与病理诊断的一致性。结果:利用DCE-MRI检出良性96例(53.33%),恶性84例(46.67%),其中良性患者TIC曲线I类占79.17%,较恶性患者更高,而III类占0.00%,较恶性患者更低,差异有统计学意义(P<0.05);良性患者的肿瘤直径为(2.39±0.64) cm,明显小于恶性患者的(3.98±1.06) cm,差异有统计学意义(P<0.05);良性患者速率常数(rate constant,Kep),容量转移常数(volume transfer constant,Ktrans),血管外细胞外化间隙容积分数(extravascular extracellular volume fraction,Ve)分别为(0.339±0.123) min−1,(0.060±0.034) min−1和0.531±0.136,其中Kep,Ktrans低于恶性患者,Ve显著高于恶性患者(P<0.05);通过绘制ROC曲线发现,Kep预测恶性鼻腔鼻窦肿瘤的最佳截断值为0.509 min−1,敏感度为81.3%,特异度为89.6%;Ktrans预测的最佳截断值为0.205 min−1,敏感度为84.4%,特异度为89.6%;Ve预测的最佳截断值为0.385,敏感度为71.9%,特异度为82.2%;DCE-MRI诊断恶性鼻腔鼻窦肿瘤的敏感度、特异度、阳性预测值、阴性预测值分别为88.37%,91.49%,90.00%,90.48%,89.58%,与病理诊断一致性Kappa=0.799。结论:DCE-MRI的TIC曲线及定量参数能为鼻腔鼻窦肿瘤性质的鉴别提供更多信息,诊断效果良好,值得临床推广。
关键词:
鼻腔鼻窦肿瘤;动态增强磁共振;时间-信号曲线;病理诊断
Value of DCE-MRI time signal curve and related quantitative parameters in the diagnosis of sinonasal neoplasms
CorrespondingAuthor: LIU Jinkang Email: 13908482001@139.com
DOI: 10.3978/j.issn.2095-6959.2018.04.009
Abstract
Objective: To investigate the diagnostic value of dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) time signal curve (TIC) and related quantitative parameters in sinonasal neoplasms. Methods: A total of 180 patients with nasal sinus in Xiangya Hospital Central South University from February 2015 to February 2017 were enrolled, all patients were pathologically diagnosed. The patients obtained preoperative DCE-MRI scanning, gained TIC and related quantitative parameters, with surgical pathology diagnosis as the gold standard, the effect of the TIC of DCE-MRI and the related quantitative parameters of diagnosis were analyzed. Receiver-operating characteristic (ROC) curve was used to analyze the sensitivity and specificity of each parameter’s prediction on nasal cavity and sinus tumors; the Kappa test was used to analyze the consistency of DCE-MRI and the pathological diagnosis. Results: Ninety-six cases (53.33%) of benign tumors were detected by DCE MRI, and 84 cases (46.67%) of malignant, including 79.17% of patients with benign TIC curve I class, higher than that in patients with malignant, and III class accounted for 0.00%, lower than that in patients with malignant, the difference was statistically significant (P<0.05). The average tumor diameter of benign patients was (2.39±0.64) cm, which was significantly smaller than that of malignant patients (3.98±1.06) cm, and the difference was statistically significant (P<0.05). Among benign patients, rate constant (Kep), volume transfer constant (Ktrans) and extravascular extracellular volume fraction (Ve) were respectively (0.339±0.123) min−1, (0.060±0.034) min−1, 0.531±0.136, among which Kep and Ktrans were lower than those of malignant patients, and Ve significantly higher than malignant patients (P<0.05). By ROC curve, Kep predicted that the optimal truncation value of malignant nasal sinus tumors was 0.509 min−1, with a sensitivity of 81.3% and 89.6% specificity. The best truncation value predicted by Ktrans was 0.205 min−1, with a sensitivity of 84.4% and a specificity of 89.6%. The best truncation value of Ve was 0.385, sensitivity was 71.9%, and the specificity was 82.2%. The sensitivity, specificity, positive predictive value and negative predictive value of DCE-MRI diagnosis of malignant nasal sinus tumors were 88.37%, 91.49%, 90.00%, 90.48%, 89.58%, and the pathological diagnosis consistency Kappa=0.799. Conclusion: The TIC curve and quantitative parameters of DCE-MRI can provide more information for the identification of nasal sinus tumor, and it is well diagnosed and worthy of clinical promotion.
Keywords:
rhinosinus tumor; dynamic contrast enhanced magnetic resonance imaging; time signal curve; pathological diagnosis