Formulation And Applied Research Of CT/MRI Diagnostic Strategy Process About Cerebral Ischemia

Objective: To amend a partial understanding of cerebral ischemia anddefine the concept afresh, and to make radiologists and clinicians pay moreattention to it; To comprehend the imaging manifestation that has clinicalsignificance on CT/MRI images, and to explore the relationship between theimaging manifestation and infarction, thus to provide accurate directions thefirst time for clinicians through CT images, shorten the diagnostic andtherapeutic time window effectively, reduce disability rate and mortality, andimprove patients prognosis and quality of life.Methods:385cases of patients with cerebral infarction demonstrated byCT/MRI examination and clinical data were drawn at random and analysedretrospectively. All patients underwent skull CT examination and the resultswere negative or doubtful. Reexamination of the cases was made in7days,that6cases with CT examination,237cases with MRI scan and MRA,2caseswith MRA and DWI,8cases with MRA, and115cases with MRI scan, MRAand DWI. New lesions occurred in all cases through reexamination and allcases were evaluated by CT/MRI diagnostic strategy process about cerebralischemia. The idea of formulating process is to optimize a set of evaluationsystem which is simple, scientific and operational. Thus we could continuewith clinical data mentioned above by restrospective cohort study. In addition,we planed to get the results and evaluation indicators of value from the data,especially the indicator that could forecast occurence of brain events. Allindicators and programes in the process aimed at anatomical information thatcould be identified from CT/MRI images, so that we could evaluate each ofthem. All evaluation steps had their own quality control system andimprovement measures. Application of this process, we observed the densityof blood vessels, the width and boundaries of cortical sulci, the size of ventricles, the demarcation between white and gray matter, as well as theboundares of lentiform nucleus and caput nuclei caudati by analysing theinitial CT images. We classified the patients of cerebral infarction into casegroup. At the same time, we selected200healthy human beings that had beendemonstrated by reexamining with MRI/CT and classified them as controlgroup. We observed the same parameters as case group, including the widthand boundaries of cortical sulci, the size of ventricles,the demarcation betweenwhite and gray matter, as well as the boundaries of lentiform nucleus andcaput nuclei caudati. We not only compared the two groups with statisticalanalysis, and find out which was the most predictive indicator. But alsocompared any combination of the indicators, and got the information aboutranking of positive predictive value; We classified the case group into fourgroups (ACA, MCA,PCA,BA-VA) according to the vascular distribution, andanalysed the display difference of CT features and MRI vasculars signal indifferent infarction sites; We observed M2segment of the right middlecerebral artery located in the right Sylvian fissure, and made sure about thepredictability of this sign to the true situation of M2segment; We messuredCT value of the vessels that supply infarction zone and analysed the range ofCT value, so that CT value among different vessels could be compared andregional difference could be analysed.SPSS16.0package was used for data analysis, when P <0.05consideredto have significant difference. The statistical approach used in comparingamong groups of enumeration data was chi-squared test.Results:1Patients with the ischemic cerebrovascular disease were brought to hospitalmosting due to limb weakness, slurred speech and head discomfort.2Compared with narrowing and obscured cerebral sulcis, small of ventricles,indistinct demarcation between white and gray matter, obscured boundaries oflentiform nucleus and hyperdense artery in sex, there was statistic differencebetween obscured cerebral sulcis(P=0.001) and hyperdenseartery(ACA:P=0.004;MCA:P=0.000; PCA:P=0.021; BA:P=0.000). 3Compared with narrowing and obscured boundaries of cerebral sulcis, smallof ventricles, indistinct demarcation between white and gray matter, obscuredboundaries of lentiform nucleus and hyperdense artery in different infarctionsites(ACA,MCA,PCA,BA-VA), there was statistic difference in hyperdenseartery(P=0.000), indistinct demarcation between white and graymatter(P=0.012), and narrowing cerebral sulcis (P=0.041) existence statisticdifference.4To compared with the density of responsible vessels of the infraction, therewere significant differences in density of different vessels（2=22.224，P=0.000）.The density in sequence from high to low was: BA, MCA, ACA,andPCA.5There was a correlation between the display rate of M2segment located inthe right Sylvian fissure in CT image and the true situation of middle cerebralartery branches, the coefficient of correlation on different side (left and right)was0.466（P=0.000）and0.430（P=0.000）respectively.6The most sensitivity indicator was obscured boundaries of lentiform nucleusand caput nuclei caudati (82.3%), the most specificity indicator was obscuredboundaries of cerebral sulcis (98.5%), the highest correct diagnosis index wasobscured boundaries of lentiform nucleus (71.3%) and the lowest was thesmall ventricles (6.2%) by comparing the changes of cerebral structure whenbrain was in the state of cerebral ischemia. The index of the higest diagnosisvalue was obscured boundaries of lentiform nucleus taking everything intoconsideration.7The positive predictive value when combining three indicators of the brainstructure changes after ischemia could reach98.2%, while combining four orfive indicators could reach up to100%.8In369cases that underwent MRA examination, there were7cases whoesbasilar arteries were complete occlusion. The CT images shows threemanifestations: not clear(2cases,28.57%), rough edge(4cases,57.14%), cleardisplay(1case,14.29%). In addition, there were92cases(29.87%) whose CTimages revealed stiffness of middle cerebral artery. 9The detection of hyperintense vessel sign on T1WI was more common thanthat on FLAIR（MCA：T1WI36.7%＞FLAIR33.93%；BA：T1WI42.45%＞FLAIR15.09%）,and there were statistical differences in differentvessels(P=0.000).10The detection of vessels equal signal changes on T1WI was more commonthan that on FLAIR（ACA：T1WI44.44%＞FLAIR11.11%；MCA：T1WI38.24%＞FLAIR7.14%；BA：T1WI3.77%＞FLAIR2.83%）, and there arestatistical differences in different vessels(T1WI:P=0.048; FLAIR:P=0.000).Conclusions:1The formulation and application of CT/MRI diagnostic strategy process isimportant. It can effectively predict the possibility of cerebral ischemiaoccurs which is helpful to observe cerebral ischemia as early as possibleand to shorten the diagnostic and therapeutic time window.2Hyperintense vessel sign not only occurs in middle cerebral artery, butalso in anterior cerebral artery, basilar artery and posterior cerebral artery.This radiological sign may predict the occurrence of cerebral infarction.3In addition to hyperdense of vessels, narrowing and obscured cerebralsulcis, small ventricles, indistinct demarcation between white and graymatter, obscured boundaries of lentiform nucleus can also give helps forthe predictive diagnosis of infarction.4The predictive value of the cerebral structures change when brain tissue isin the state of ischemia is different in which, the highest correct diagnosisindex is obscured boundaries of lentiform nucleus (71.3%) and the lowestis the small ventricles (6.2%).5The positive predictive value when combining three indicators of the brainstructure changes after ischemia can reach98.2%. From this result, we cancome to a conclusion that brain tissue is in the state of ischemia whenthere are hyperdense artery and at least three changes of brain structures.6The detection of M2segment located in the right Sylvian fissure in CTimages can predict the true situation of middle cerebral artery branches inMRA images to some extent. 7Rough edge of the basilar artery works as an important reminder indiscribing the true situation of vessels to some extent. This imagingmanifestation may predict the posterior fossa lesions being combined withhyperdense artery sign.8The severity of artery stiffness may be helpful to understand the truesituation of vessels.9The sign of hyperintense is not only displayed on FLAIR but also can beseen on T1WI which may be more helpful to the diagnosis of ischemia.10The equal signal of vessels may have the same diagnosis value ashyperintense of vessels.