定量动态对比增强磁共振成像渗透性与T1灌注多参数联合分析对脑胶质瘤分级的诊断价值

Diagnostic Value of Combining Permeability with T1 Perfusion Parameters in Quantitative Dynamic Contrast-enhanced Magnetic Resonance Imaging for Glioma Grading

目的探讨定量动态对比增强磁共振成像(DCE-MRI)渗透性与T1灌注多参数联合分析对脑胶质瘤分级的诊断价值。方法回顾性分析经病理证实的16例高级别脑胶质瘤(HGG)和12例低级别脑胶质瘤(LGG)患者的磁共振平扫和DCE-MRI资料,通过药代动力学模型定量计算肿瘤的渗透性参数及T1灌注模型计算灌注参数,包括转运常数(Ktrans)、部分细胞外血管外间隙容量(ve)、血液回流常数(kep)、血浆容积(vp)和脑血流量(CBF)、脑血容量(CBV)及平均通过时间(MTT)。以t检验比较HGG和LGG定量分析参数的统计学差异,并以受试者工作特征曲线评估渗透性参数(Ktrans值、ve值)、T1灌注参数(CBF值、CBV值)及联合应用灌注和渗透性参数进行脑胶质瘤的分级诊断的敏感性、特异性和曲线下面积。结果 HGG的Ktrans值、ve值、CBF值和CBV值分别为(0.276±0.164)/min、0.486±0.191、(1.755±1.164)ml/(g·min)和(0.204±0.101)ml/g,明显高于LGG的(0.084±0.044)/min、0.274±0.132、(0.761±0.625)ml/(g·min)和(0.115±0.097)ml/g(t值分别为3.934、3.293、2.672和2.338,P均<0.05),而HGG的kep值、vp值和MTT值分别为(1.632±1.204)/min、0.114±0.107和(0.128±0.070)min,与LGG的(1.537±1.194)/min、0.055±0.039、(0.145±0.066)min比较差异无统计学意义(t值分别为0.208、1.823和0.668,P均>0.05)。单参数中以Ktrans值诊断脑胶质瘤分级的曲线下面积最大,为0.919,取阈值为0.105/min时,诊断高级别脑胶质瘤的敏感性和特异性分别为87.5%和83.3%;而联合运用多参数分级诊断时,以ve-CBF值似然比为0.631时曲线下面积最大,为0.974,敏感性和特异性分别为93.7%和100.0%。结论联合DCE-MRI的渗透性参数与灌注参数能够提高高级别和低级别脑胶质瘤分级的诊断率。

Objective To investigate the diagnostic value of combining permeability with T1 perfusion parameters in quantitative dynamic contrast-enhanced magnetic resonance imaging（ DCE-MRI） in glioma grading.Methods Magnetic resonance imaging was performed in 16 patients with high grade gliomas（ HGG） and 12 patients with low grade gliomas（ LGG） confirmed by pathology. The permeability was quantitatively analyzed and the T1 perfusion parameters of the tumor were calculated by the pharmacokinetic model,including volume transfer constant（ Ktrans）,volume fraction of extravascular extracellular space（ ve）,reflux constant（ kep）,fractional plasma volume（ vp）,cerebral blood flow（ CBF）,cerebral blood volume（ CBV）,and mean transit time（ MTT）. A t-test was used to calculate the statistical significance of quantitative analysis parameters between HGG and LGG. The receiver operating characteristic curve analysis was also performed for evaluating the sensitivity,specificity,and area under curve（ AUC） of the permeability parameters and perfusion parameters and the combination of these parameters. Results The differences of the Ktrans,ve,CBF,and CBV values [（ 0. 276 ±0. 164） / min vs.（ 0. 084 ± 0. 044） / min; 0. 486 ± 0. 191 vs. 0. 274 ± 0. 132;（ 1. 755 ± 1. 164） ml /（ g·min）vs.（ 0. 761 ± 0. 625） ml /（ g·min）;（ 0. 204 ± 0. 101） ml / g vs.（ 0. 115 ± 0. 097） ml / g] were statistically significant（ t = 3. 934,3. 293,2. 672,2. 338,P〈0. 05） between HGG and LGG. The differences of the kep,vp,and MTT value [（ 1. 632 ± 1. 204） /min vs.（ 1. 537 ± 1. 194） /min; 0. 114 ± 0. 107 vs. 0. 055 ± 0. 039;（ 0. 128 ± 0. 070） min vs.（ 0. 145 ± 0. 066） min]were not statistically significant（ t = 0. 208,1. 823,0. 688,P〈0. 05）. When the Ktransvalue was 0. 105 / min,the AUC was the largest（ 0. 919） by the single parameter in glioma grading,and meanwhile the sensitivity and specificity were 87. 5% and 83. 3%,respectively. When the ve-CBF value was 0. 631,the AUC was the largest（ 0. 974） by the multiple parameter,and meanwhile the sensitivity and specificity were 93. 7% and 100. 0%,respectively. Conclusion Combining permeability with perfusion parameters in quantitative DCE-MRI can improve the accuracy of the glioma grading.