体素内不相干运动成像与动态对比增强磁共振成像评价胶质瘤灌注的应用价值

Comparison of perfusion-related parameters of intravoxel incoherent motion and dynamic contrast enhanced-magnetic resonance imaging in gliomas

目的比较体素内不相干运动(IVIM)成像与动态对比增强磁共振成像(DCE-MRI)对于不同级别胶质瘤的诊断效能,并分析两者间的相关性。方法选择2016年6月—2018年1月间在上海交通大学医学院附属仁济医院行手术治疗,术后病理证实为胶质瘤的患者62例,男35例,女27例,年龄16~78岁,平均年龄为(52.8±4.6)岁。根据2016年WHO胶质瘤分类和分级标准将患者分为低级别胶质瘤组(LGG组,24例)和高级别胶质瘤组(HGG组,38例)。所有患者术前均行MRI检查,经工作站和软件后处理后,获得IVIM双指数模型相关参数真实扩散系数(D)、假扩散系数(D*)和灌注分数(f),以及DCE-MRI灌注相关参数容积转运常数(Ktrans)、血管外细胞外间隙容积分数(Ve)、血浆容积分数(Vp),分别在IVIM和DCE-MRI的后处理图像上手动放置感兴趣区(ROI),得到各参数的测量值。采用Mann-Whitney U检验比较两组间各参数值,IVIM灌注相关参数f、D*与DCE-MRI相关灌注参数Ktrans、Ve、Vp之间的相关性分析采用Spearman秩相关检验。采用ROC曲线检验相关参数区分高、低级别胶质瘤的诊断效能,根据约登指数确定最佳诊断界值、诊断灵敏度和特异度。结果 HGG组的IVIM灌注相关参数D*值、f值和DCE-MRI灌注相关参数Ktrans值、Ve值均显著高于LGG组(P值均<0.05),IVIM扩散相关参数D值显著低于LGG组(P<0.05)。两组间DCE-MRI灌注相关参数Vp值的差异无统计学意义(P>0.05)。Spearman相关性分析显示,f与Ktrans具有强相关性(ρ=0.614,P<0.01),与Ve和Vp具有弱相关性(ρ值分别为0.385和0.364,P值均<0.01);D*与Ktrans具有中等相关性(ρ=0.432,P<0.01),与Vp具有弱相关性(ρ=0.394,P<0.01),与Ve无相关性(ρ=0.172,P=0.463)。在区分低级别与高级别胶质瘤的诊断价值方面,诊断准确性最高的参数是f(AUC=0.873),其最佳诊断界值为5.250%,诊断灵敏度和特异度分别为0.902和0.829。结论胶质瘤IVIM相关参数与DCE-MRI灌注相关参数密切相关,IVIM技术可在无需外源性造影剂示踪的情况下在体评价脑胶质瘤的灌注情况,f值在区分高、低级别胶质瘤中的诊断效能最优。

Objective To explore the diagnostic efficiency and correlation between intravoxel incoherent motion （IVIM） and dynamic contrast enhanced-magnetic resonance imaging （DCE-MRI） in identifying the low- and high-grade gliomas. Methods A total of 62 patients with pathologically-confirmed gliomas who underwent surgical treatment between June 2016 and January 2018 were enrolled in this study. There were 35 males and 27 females with a mean age of （52.8±4.6） years （range, 16-78 years）. They were assigned to low-grade glioma （LGG） group （n = 24） and high-grade glioma （HGG） group （n = 38） according to WHO Glioma Classification and Grading 2016. All the patients were examined by MRI before surgery. Diffusion parametes of pure diffusion coefficient （D） value, perfusion parameters of pseudodiffusion coefficient （D＊） value, perfusion fraction （f） value derived from IVlM, and DCE-MRI derived perfusion parameters of volume transfer constant （Ktrans）, extravascular extracellular space volume fraction （Ve） and fraction of plasma volume （Vp） were obtained. Regions of interes! （ROD were put in the solid portion of the tumors. All values were compared between LGG and HGG groups by using Mann-Whitney U test. Correlation between IVIM and DCE-MRI derived perfusion-related parameters was analyzed by using the Spearman rank correlation. The diagnostic performance for tumor grading was evaluated with receiver operating characteristic （ROC） analysis. The optimal cutoff value, diagnostic sensitivity and specificity were identified by Youden Index. Results D value, f value, Ktrans and Ve were significantly higher in HGG group than those in LGG group （all P^0.05）, while D value was significantly lower in HGG group than that in LGG group （P〈0.05）. There was no significant difference in the Vp between two groups （P〉0.05）. Spearman correlation analysis showed that the f value was strongly correlated with Ktrans （p = 0.614, P〈0.01 ） and weakly correlated with Ve and Vp （ρ = 0. 385 and 0. 364, respectively, both P〈0.01）, that D value had medium correlation with Ktrans （ρ= 0. 432, P〈0.01 ） and weak correlation with Vp （ρ = 0. 394, P d0.01 ）, but had no correlation with Ve （ρ=0. 172, P = 0. 463）. Among all the parameters, f value achieved the highest accuracy for differentiating low- from high-grade gliomas, with an area under the ROC curve value （AUC） of 0. 873, and the diagnostic sensitivity and specificity were 0. 902 and 0. 829 at the optimal cutoff value of 5. 250%. Conclusion Perfusion parameters derived from IVIM are closely correlated with perfusion parameters derived from DCE-MRI. IVIM may serve as an alternative to measure tumor perfusion in gliomas in a noninvasive and efficient way, and f value has the best diagnostic performance for tumor grading.