Evaluation And Application Value Of Multiple Magnetic Resonance Imaging Perfusion Analysis Techniques In Microvessel Of Glioblastoma

Background and Objective:Glioblastoma（GBM）is one of the most aggressive gliomas,accounting for 80%of malignant glioma.A large number of irregular microvessels were observed under GBM,and the pathological manifestations were different types of vascular structure formation.These characteristics determine the progress and prognosis of GBM,which can be used as an independent predictor of prognosis of GBM patients.The contrast enhancement of conventional enhanced MRI can not reflect the function changes of tumor microvessels,and it is difficult to diagnose glioma accurately and to judge prognosis.At present,a variety of MRI functional imaging technology can provide various parameters related to vascular morphology and function,so as to achieve non-invasive,dynamic and quantitative evaluation of tumor microvascular permeability,and has been widely applied in scientific research and clinical practice.The correlation between the microvessel pathology parameters of GBM and the three kind of MRI functional imaging parameters of Dynamic susceptibility contrast（DSC）,Dynamic contrast-enhanced（DCE）and Vessel size imaging（VSI）is studied,and the prognosis of tumor is evaluated.Materials and Methods:1.Research subjectDuring April 2013 to April 2017,61 patients,which were find intracranial tumors by the conventional and perfusion MR scan,were recruited in the Department of Neurosurgery,Daping Hospital,Army Medical University.According to the 2016 WHO central nervous system tumor classification and grading standards,there were 61 cases with GBM,including 37 males,24 females,age range of 10⁷9 years,the average age of 51.49±14.66years.As a retrospective study,the paraffin specimens of 29 patients were obtained,underwent CD34 and periodic acid-Schiff（PAS）double staining.All cases were first diagnosed,without any radiation or chemical treatment.2.Equipment and image analysisRoutine magnetic resonance imaging and SE perfusion scan were performed in SIEMENS Magnetom Verio 3.0T MRI.DSC-MRI was used to evaluate the vascular distribution and blood perfusion of the patients before operation.The parameters of cerebral blood volume（CBV）,cerebral blood flow（CBF）,CBV and CBF in the contralateral normal white matter area were measured,and relative cerebral blood volume（rCBV）and relative cerebral blood flow（rCBF）were calculated.The application of GE Signa HDX1.5T MR scanner was used for DCE-MRI,the analysis parameters including semi quantitative parameters of the initial area under curve（IAUC）,and a variety of quantitative parameters of volume transfer constant(K^trans),reverse transfer constant（Kep）,extravascular extracellular volume fraction（Ve）,and blood plasma volume（Vp）.Using GE Signa HDX 1.5T MR scanner,GE-SE-EPI sequence was used for VSI to determine the maximum VSI_MRI and VSI_mean of tumor area.3.Pathological staining and image analysisThe paraffin sections of 29 patients were prepared.CD34 and Periodic acid-Schiff（PAS）double staining was performed,then microvessel density（MVD）,microvessel area（MVA）and morphology were analyzed using computer image analysis software（Image-Pro Plus 6.0）.4.Statistical analysisSPSS 23.0 statistical software was used to analyze the MRI values of GBM.Measurement data were present with the x±s.the parameters of all three MRI functional imaging measured,and between with the survival time of patients were performed Pearson the correlation analysis.The number of different types of GBM microvessels,the average MVA,MVA%and the proportion of them in five types of microvessels were counted.The results were expressed by M（P25,P75）.Kruskal-Wallis test was performed for different types of GBM microvessel pathological parameters.When the overall test was meaningful,comparisons between different types of microvessels were performed by Dunn’s multiple test.Using Pearson correlation analysis was used to evaluate the corelation of the CBV,CBF,rCBV,rCBF,K^trans,Kep,mean VSI_MRI,max VSI_MRI,Ve,IAUC,Vp,and other imaging parameters and various types of microvessels,MVA%and MVA average in five types of microvessels in the proportion.The difference was statistically significant in P<0.05.Results:1.The CBF and CBV pseudo-color maps show the tumor area and the blood flow and blood volume of its peritumoral area.The solid area of the tumor is patchy red and yellow-green,the necrotic area is blue,and the surrounding edema area is blue and yellow-green（low Perfusion）;VSI pseudo-color map shows that the tumor signal is obviously uneven,the parenchyma of the tumor is the spotted red area and the flaky yellow-green,the necrotic area is blue,the surrounding edema area is yellow-green;the color of the DCE picture is uneven,The solid area of the tumor also is red and yellow,the necrotic area is blue,and the edema around it is also blue.2..By correlation analysis between the parameters of the three imaging images of GBM,there was a statistically significant correlation between Ve and rCBV,and K^trans and rCBF（All P<0.05）,and there was no statistically significant correlation between the other parameters of the three imaging parameters of GBM and the parameters of Pathological image and survival time（All P>0.05）.3.The Heterogeneity of GBM increased significantly in morphology,diameter and density.The number of microvascular sprouting（MS）is the largest,but the average MVA is the smallest;and the average MVA of GVP is the lagerest,the proportion of MVA%in the sections,and all kinds of proportion also is the largest,and GVP has a long diameter of128.1±75.1μm and a short diameter of 65.1±37.1μm,which are much larger than normal brain tissue microvessels..4.The larger diameter of vessels is not in the most concentrated area of tumor cells.There is not a large number of microvessels in the largest area of microvascular diameter.That is to say,the maximum area of MVD is opposite to the maximum value of microvessel diameter.5.There was no statistically significant correlation between the pathological parameters of the GBM microvessels and the survival time（All p>0.05）.6.The GBM imaging data and pathological data were analyzed by Pearson correlation analysis,found that CBV and the average area of MS（r=0.165,P=0.027）,rCBF and VG（MVA%）（r=-0.200,P=0.002）and VG（the proportion of the total type）（r=-0.217,P=0.020）,and IAUC and the average area of MS（r=0.421,P=0.041）showed a linear positive correlation.K^trans and the average area of vascular cluster（VC）（r=0.395,P=0.038）,Kep and the average area of VC（r=0.073,P=0.044）showed a negative correlation,no significant correlation between other values.Conclusion:1.DSC-MRI,DCE-MRI and VSI-MRI can perform microvascular imaging of GBM,and evaluate the blood flow,blood volume and vascular endothelial permeability of GBM.The color heterogeneity of perfusion pseudo-color map also reflects the heterogeneity of vascular morphology,diameter,density and distribution of GBM,which is consistent with pathological findings.2.Although GVP is not abundant,its microvascular area is the largest,and the impact on the total microvascular area is also the largest.Because MVA has a guiding and predictive effect on the survival of patients,it also indirectly confirms that GVP influences the prognosis of GBM patients.The values of perfusion function imaging parameters and other pathological microvascular parameters reflecting the survival and prognosis of patients need further verification.3.GBM imaging parameters,including CBV,rCBF,IAUC,K^trans and Kep can partly reflect the pathological parameters,average area of MS,VG（MVA%）,VG（the proportion of total types）,and the average area of VC.