| PurposeOur aims are to investigate the differences in plaque characteristics between patients with hypoperfusion and without in patients with middle cerebral artery stenosis,to find independent predictors of hypoperfusion,and risk factors of hypoperfusion volume;to assess difference between infarction patients and TIA patients in plaque characteristics,to find independent predictors of infarct size,and to establish predictive models of the extent of neurological impairment in patients.Materials and MethodsPart one:Retrospective analysis of 71 patients with ischemic symptoms on MCA within recent 2 weeks,DWI,HR-VWI,TOF-MRA and DSC-PWI were performed.Plaque characteristics and Pial collaterals(PCs)were measured by 3 radiologists;Tmax>6.0 s delayed volumes were obtained as hypoperfusion volume using RAPID software.According to the presence or absence of Tmax>6.0 s delayed volumes,patients were divided into group with hypoperfusion and without.Plaque characteristics and PCs of the two groups were compared by Chi-square test,Fisher's exact test and so on.Univariate and multivariate logistic regression analysis were used to assess the association between variables and occurrence of hypoperfusion area.Receiver operating characteristic(ROC)curve analysis was used to evaluate the diagnostic efficacy of plaque characteristics in the hypoperfusion are,and the comparison of area under the curve(AUC)of each variable was compared by Z test.The correlations between the plaque characteristics,PCs and Tmax>6.0 s delayed perfusion volume were calculated with Spearman correlation analysis.Part two:Retrospective analysis of 67 patients with acute ischemic symptoms of the middle cerebral artery within 2 weeks.According to DWI or neurological function tests,the patients were divided into Stroke group and TIA group.The examination sequences and image post-processing were the same as in the first part.In addition,three radiologists assessed DWI-ASPECTS scores.Plaque characteristics,PCs and hypoperfusion volume of Stroke group and TIA group were compared by Chi-square test,Fisher's exact test and so on.Univariate and multivariate logistic regression analysis were used to assess the association between variables and occurrence of infarction.ROC curve analysis was used to evaluate the diagnostic efficacy of plaque characteristics in predicting the presence of acute stroke onset.Spearman correlation analysis was used to evaluate association between plaque characteristics,PCs,Tmax>6.0 s volume,DWI-ASPECTS and NIHSS.Multiple linear regression analysis was used to establish a predictive model of the extent of neurological impairment.ResultsPart one:71 patients(mean age:59.90±11.25;44 male,27 female)were included in the study,29 patients in group with hypoperfusion and 42 patients in group without.There were significant differences in percentage of infarction and NIHSS between the two groups(P=0.012 and 0.035).The three evaluators had good consistency of plaque characteristics and collateral scores(0.796-0.956).Between two groups,eccentric index,eccentric plaques,plaque length,and PCs were statistically significant(P=0.016,0.002,0.003,and 0.001).In multivariate logistic regression analysis,eccentricity,plaque length,PCs,and hypertension were independent predictors of intracranial hypoperfusion(P=0.014,0.017,0.035,and 0.018)after adjustment of age,sex,and cerebrovascular risk factors.The AUC of the eccentricity,plaque length,PCs and hypertension as a single index were 0.681,0.704,0.694,0.623,respectively.The AUC of combined eccentricity,plaque length,PCs,and hypertension were 0.865.The sensitivity of diagnosis reached 79.31%and the specificity was 83.33%.The AUC of combined plaque eccentricity,plaque length,PCs,and hypertension model was significantly higher than the each individual variable(P<0.05).In the group with hypoperfusion,the eccentric index was correlated with Tmax>6.0 delayed volume(r_s=-4.08,P=0.028),and there was a statistically significant difference in Tmax>6.0 s delayed perfusion volume between eccentric plaques and concentric plaque patients(H=-2.085,P=0.037),and between different enhanced grade(H=9.464,P=0.009).Part two:67 patients(mean age:58.01±12.40 y;41 male,26 female)were included in the study,45 in the Stroke group and 22 in the TIA group.There were statistical differences in NIHSS between the two groups(Z=-3.102,P=0.002).Three evaluators had good consistency for plaque features and PCs(0.743-0.926).Plaque length,intraplaque hemorrhage(IPH),enhanced grade,PCs,and Tmax>6.0 s volume were significantly different between two groups(P<0.05).The results of multivariate logistic regression analysis showed that IPH and PCs were independent predictors of acute stroke onset(P=0.019 and 0.043),after adjustment for age,sex and cerebrovascular risk factors.The AUC of IPH and PCs on ROC curve were 0.755and 0.759.Combined with two variables was 0.860,and diagnostic sensitivity was73.33%,and specificity was 90.91%.The AUC of ROC curve of the combination model was significantly higher than the individual variables(P<0.01).In Stroke group,plaque length,plaque enhancement,IPH,Tmax>6.0 s volume,and PCs were correlated with DWI-ASPECTS(P<0.05).There was a multiple linear relationship between Tmax>6.0 s volume,IPH,hypertension and PCs and NIHSS score,and the model was significant(F[4,58]=16.108,P<0.001).The multi-correlation coefficient of the model was R=0.725,and the adjusted R~2=0.494.ConclusionIn patients with acute middle cerebral artery ischemic events,plaque eccentricity,plaque length,PCs,and hypertension can affect intracranial hypoperfusion area,and the hypoperfusion volume is related to eccentricity and enhanced grade.In addition,plaque length,IPH,enhanced grade,PCs and hypoperfusion volume affect the occurrence of infarction and infarct volume.The linear regression model based on the plaque characteristics,hemodynamics,and hypertension can predict the neurological impairment of patients,and provide important informations for severity evaluation and clinical treatment in radiography. |
PDF Full Text Request