Differentiation between cerebral metastases , meningiomas , and primary gliomas by dynamic contrast-enhanced MRI

Huarui Du,Bing Fan,Ya Cao,Xiaoying Wang,Jue Zhang,Jing Fang
2012-01-01
Abstract:Purpose: Conventional magnetic resonance (MR) imaging has a limited capacity to differentiate intraaxial brain masses[1,2]. The purpose of this study is to evaluate the capability of the volume transfer coefficient (K) and the interstitial space volume fraction (ve) derived from Reference Tissue (RR) model[3] to differentiate intraaxial brain masses. Materials and Methods : A total of 59 patients with pathologically confirmed intraaxial brain masses (high -grade glioma, HGG, n =5; low-grade glioma, LGG, n =6; meningiomas, n =10; cerebral metastases, n =20; brain abscess,n=3) underwent dynamic contrast-enhanced MR imaging. Images were acquired using a three-dimensional (3D) fast gradient echo sequence. K and ve were calculated from the DCE MRI data by RR model. Results were compared with pathologic findings. Kruskal-Wallis H test and independent T-test were used to compare differences between the 5 groups. Receiver operating characteristic curve analysis was performed for each of the variables in differentiation between LGG and HGG. Results: K of 59 cases of intraaxial brain masses in the parenchyma: LGG is 0.054±0.015, HGG is 0.210±0.032, brain abscess is 0.234±0.032, cerebral metastases is 0.202±0.021, and meningiomas is 0.267±0.048. (FIG.1) It is observed that there are significant differences among the five group with Kruskal-Wallis H test (Chi-Square =17.37,p<0.05) and significant differences exist between LGG and other four kinds of brain masses with independent T-test (p<0.05, especially for HGG p< 0.01). The optimal cutoff value of K for differentiating LGG from HGG is 0.072, which implies a sensitivity of 90% and a specificity of 83.3%. In contrast, the optimal cutoff value of ve is 0.231, implying a sensitivity of 90% and a specificity of 67.7%. But there are no significant differences among HGG, brain abscess, cerebral metastases and meningiomas. However, the K values in the peripheral edema of the HGG and brain metastasis are 0.063±0.052 and 0.010±0.007 respectively. Meanwhile, there shows a significant difference (FIG. 3) between the K values in the peritumoral edema of HGG and cerebral metastases (p<0.001).
What problem does this paper attempt to address?