Study Of Rupture Risk Of Intracranial Aneurysms With Multimodal Magnetic Resonance Imaging

With the advancement of magnetic resonance imaging technology(MRI),MRI can display not only the structural characteristics of intracranial aneurysms(IAs),but also the functional characteristics of IAs.Gadolinium-enhanced vessel wall MRI(VW-MRI)has ultra-high spatial resolution,which could indicate the pathological inflammatory response of IAs wall.Dynamic contrast enhanced magnetic resonance imaging(DCE-MRI)has an ultra-high time resolution and could evaluate the pharmacokinetic information for example,the permeability of the IAs wall.4D-flow-MRI could evaluate hemodynamic information for example wall shear stress(WSS)of IAs.This study was based on the latest advances in MRI technology described above to study the functional characteristics of IAs through multi-modality MRI.On the one hand,this study evaluated the ability of advanced MRI technology to screen for rupture risk of IAs,etc.;on the other hand,it was based on advanced MRI technology to further understand the functional changes in the process of IAs occurrence,development and rupture.This study was divided into the following three parts.Part 1 Contrast-Enhanced Vessel Wall Magnetic Resonance Imaging Study of Intracranial Aneurysms StabilityObjective:The prevalence of unruptured intracranial aneurysms(UIAs)was 7%in Chinese population aged 35-75.With the widespread use of non-invasive imaging techniques such as magnetic resonance angiography(MRA)and computed tomography angiography(CTA)in clinical practice,the detection rate of UIAs has been continuing to increase.UIAs rupture causes 80%of non-traumatic subarachnoid hemorrhage(SAH).The cumulative morbidity,mortality of UIAs surgery or intravascular treatment was between 3%and 10%,while the risk of UIAs spontaneous rupture was less than 1%per year.Therefore,individualized rupture risk assessment is of great significance to UIAs clinical management.Unstable IAs(symptomatic UIAs,increased UIAs during follow-up,and ruptured IAs,etc.)had a higher risk of rupture than stable IAs(IAs discovered by chance or IAs that did not increase during imaging follow-up)and needed to accept more active treatment.In summary,it was also important to screen for UIAs with a low or relatively stable risk of rupture to avoid unnecessary surgery.This study was based on gadolinium-enhanced VW-MRI,to explore the value of aneurysm wall enhancement(AWE)and wall enhancement index(WEI)in screening UIAs stability,and to further understand IAs’ functional changes during occurrence,development and rupture.Methods:1.A retrospective analysis of IAs patients who underwent 3.0T magnetic resonance gadolinium-enhanced VW-MRI examination and digital silhouette angiography(DSA)from October 2014 to October 2019 in our hospital was performed.Through multidisciplinary consultation(MDT),symptomatic UIAs and ruptured IAs were identified and defined as unstable IAs,and we accidentally discovered IAs were defined as stable IAs.A total of 283 patients were enrolled,including 174 patients with stable IAs and 109 patients with unstable IAs.The aforementioned patients were all tested with 3.0T Siemens magnetic resonance scanner for VW-MRI.Among them,54 patients completed gadolinium-enhanced 2D-VW-MRI examination based on 3.0T Siemens Verio/Skyra magnetic resonance scanner(16-channel head and neck combined coil).229 patients completed gadolinium-enhanced 3D-VW-MRI examination based on 3.0T Siemens Prisma magnetic resonance scanner(64-channel head and neck combined coil).2.Two senior neuroimaging physicians independently conducted blind reading on the image archiving and communication systems(PACS)platform to determine whether AWE appeared in IAs and further strengthened the AWE pattern(AWEP):AWEP 0(none),AWEP 1(focal)and AWEP 2(circumferential).3.Two experienced neuroimaging physicians independently conducted blind analysis of VW-MRI images of IAs through Vessel-MASS 2014-EXP software and calculated the semi-quantitative index of AWE-WEI.4.We used MedCalc 18.2 software for statistical processing.Categorical variables are represented by percentages,and continuous variables are represented by x ± s or median(interquartile range).Kappa test was used to evaluate the consistency of AWE and its classification by 2 physicians.The intraclass correlation coefficient(ICC)was used to evaluate the consistency of WEI in evaluating 2 medical practitioners.The χ2 test was used to compare categorical variables between groups,and the U test was used to compare continuous variables between groups.Variables with P<0.2 between the groups were included in univariate regression analysis.Variables with univariate regression analysis results P<0.05 were include in the multivariate logistic regression analysis to obtain 95%confidence interval(CI)and odds ratio(OR)for predicting UIAs instability.We analyzed categorical or continuous variables through receiver operating characteristic(ROC)curves to obtain sensitivity,specificity,etc.;and clarified cut-off value of the ROC curve through the Yoden index(P<0.05 was considered statistically significant).Result:1.This study included 283 patients with 362 IAs.Inter-reader agreement was excellent for both the presence of AWE(k=0.83[95%CI:0.77 to 0.89]),AWE’s classification(k=0.87[95%CI:0.83 to 0.92])and the calculation of WEI(ICC=0.98[95%CI:0.97 to 0.98]).2.The median of WEI was significantly higher in unstable than in stable UIAs(1.3 versus 0.3,respectively;P<0.001).AWEP also has a different composition ratio in the unstable state of UIAs(P<0.001).3.Multivariate logistic regression revealed that AWEP(OR,2.0;95%CI,1.2 to 3.3;P=0.005)and WEI(OR:2.9,95%CI:1.5 to 5.5;P=0.001)were the only independent factors associated with unstable UIAs.4.The cut-off value of AWEP screening UIAs stable state was AWEP≤0(sensitivity:69.0%,specificity:81.7%),area under curve(AUC)was 0.79(95%CI:0.74 to 0.83),Youden index was 0.51,P<0.001.The cut-off value of WEI screening UIAs stable state was WEI ≤0.98(sensitivity:77.4%,specificity:78.5%),AUC was 0.78(95%CI:0.73 to 0.82),and Youden index was 0.56,P<0.001.The cut-off value of combined AWEP and WEI screening UIAs stable state was AWEP+WEI≤0.20(sensitivity:73.4%,specificity:95.7%),AUC was 0.91(95%CI:0.88 to 0.94),and Yoden index was 0.69,P<0.001.Conclusion:1.AWEP and WEI were independent risk factors for unstable UIAs.2.AWEP 0 could effectively screen the stable state of UIAs.WEI≤0.98 could also effectively screen the stable state of UIAs.The combination of AWEP and WEI screening UIAs stable state had the highest specificity.Part 2 Study on the Instability of Intracranial Aneurysms Wall by Dynamic Contrast-enhanced Magnetic Resonance Imaging and VW-MRIObjective:IAs were pathological enlargements of intracranial arteries,with an incidence of 3%to 5%in adult populations regardless of race or region.Relative to the risk of UIAs spontaneously rupturing,the risks posed by UIAs treatment might be higher.Therefore,in order to balance the risks and benefits of the treatment of UIAs patients,it was of great significance to study individualized IAs rupture risk assessment markers.In daily clinical practice,the risk assessment of IAs was mostly based on morphological factors such as diameter.Pathological markers based on the wall of IAs might help supplement pure morphological markers.At present,most of the noninvasive imaging studies on the walls of IAs were based on VW-MRI.In clinical practice,sentinel headaches with a high risk of rupture of IAs were also considered to be caused by micro-exudation of blood.S AH was caused by a large amount of blood surging into the cerebrospinal fluid of the subarachnoid space caused by the totally rupture of the IAs wall.The aforementioned blood leakage may be detected by gadolinium contrast agent by DCE-MRI and based on the post-processing algorithm to obtain Ktrans for quantitative evaluation.This study used DCE-MRI and VW-MRI to explore the value of Ktrans and AWE in screening UIAs instability,and to further understand the functional changes in the occurrence,development and rupture of IAs.Methods:1.We prospectively collected IAs patients who underwent 3.0 T magnetic resonance gadolinium-enhanced DCE-MRI,VW-MRI and DSA from January 2018 to October 2019 in our hospital.Symptomatic UIAs and ruptured IAs were identified through MDT and defined as unstable IAs.It was Incidentally discovered that IAs were defined as stable IAs.A total of 82 patients were enrolled,including 51 patients with stable IAs and 31 patients with unstable IAs.All patients completed gadolinium-enhanced DCE-MRI and 3D-VW-MRI examinations based on a 3.0T Siemens Prisma magnetic resonance scanner(64-channel head and neck combined coil).2.Two experienced neuroimaging physicians independently performed blind analysis of the DCE-MRI images of IAs through the Siemens syngo.via workstation TISSUE 4D software and calculate the full quantitative pharmacokinetic index of the corresponding aneurysms-Ktrans.3.Two senior neuroimaging physicians independently performed blind reading on the PACS platform to determine whether AWE appeared,and further classify AWEP:AWEP 0(none),AWEP 1(focal)and AWEP 2(circumferential).4.We used MedCalc 18.2 software for statistical processing.Categorical variables were represented by percentages,and continuous variables were represented by x ± s or median(interquartile range).Kappa test was used to evaluate the consistency of AWE and its classification by 2 physicians.We used Bland-Altman diagram,Passing and Bablok regression,and ICC to evaluate the consistency of Ktrans by two physicians.The χ2 test was used to compare categorical variables between groups,and the U test was used to compare continuous variables between groups.Variables with P<0.2 between the groups were included in univariate regression analysis.Variables with univariate regression analysis results P<0.05 were included in the multivariate logistic regression analysis to obtain 95%CI and OR for predicting the unstable state of UIAs.Analyzed categorical or continuous variables through ROC curve to obtain sensitivity,specificity,etc.The cutoff value of the ROC curve was clarifiea by the Youden index.The χ2 test was used to analyze the categorical variables,and the Jonckheere-Terpstra test was used to analyze the continuous variables.P<0.05 was considered statistically significant.Result1.This study included 82 patients with 101 UIAs.Inter-reader agreement was excellent for both the presence of AWE(k=0.85[95%CI:0.75 to 0.96]),AWE’s classification(k=0.81[95%CI:0.71 to 0.92])and the calculation of Ktrans(ICC=0.98[95%CI:0.97 to 0.99]).2.The median of Ktrans was significantly higher in unstable than in stable UIAs(0.9 versus 0.3 s-1,respectively;P<0.001).There was also a difference in the composition ratio of AWEP in the unstable state of UIAs(P<0.001).3.Multivariate logistic regression revealed that AWEP(OR,4.1;95%CI,2.06 to 8.16;P<0.001)and Ktrans(OR,2.77;95%CI,1.49 to 5.17;P=0.01)were the only independent factors associated with unstable UIAs.4.The cutoff value of AWEP screening UIAs unstable state was AWEP>1(sensitivity:59.3%,specificity:87.0%),AUC was 0.75(95%CI:0.65 to 0.83),Youden index was 0.50,P<0.001.The cut-off value of Ktrans screening UIAs unstable state was Ktrans>0.65 s-1(sensitivity:59.3%,specificity:78.3%),AUC was 0.70(95%CI:0.60 to 0.79),and Youden index was 0.38,P=0.01.The cut-off value of combined with Ktrans and AWEP screening UIAs unstable state was Ktrans+AWEP>0.18(sensitivity:85.2%,specificity:76.8%),AUC was 0.83(95%CI:0.74 to 0.90),and Yoden index was 0.62,P<0.001.Conclusion:1.Both Ktrans and AWEP were independent risk factors for the unstable state of UIAs.2.Ktrans>0.65 s-1 could effectively screen the unstable state of UIAs.AWEP could also effectively screen the unstable state of UIAs.The Ktrans+AWEP had the highest sensitivity to screen the unstable state of UIAs.Part 3 4D-Flow and DCE-MRI Study on the Wall Enhancement of Unruptured Intracranial AneurysmsObjective:UIAs rupture cause about 80%of spontaneous SAH.Gadolinium-enhanced AWE in VW-MRI has been proved to be related to unstable state or rupture risk of UIAs and helped to individualize the risk stratification of the corresponding patients.This discovery expanded the rupture risk assessment method based on aneurysm from UIAs sac anatomy to UIAs wall pathology for the first time and brought a new perspective and ideas to the assessment of UIAs rupture risk.AWE might be closely related to the pathological inflammatory response of the aneurysm wall,and the hemodynamic index of UIAs-WSS was also considered to be closely related to the pathological inflammation of the aneurysm wall.Unstable states such as magnified and ruptured IAs might not only be related to the intense inflammatory reaction in the aneurysm wall,but also be related to the endothelial cell apoptosis in the aneurysm wall.Apoptosis of endothelial cells in the UIAs wall might lead to an increase in the permeability of the aneurysm wall,and this change might be reflected by Ktrans.This study used 4D-flow-MRI,DCE-MRI and VW-MRI to explore the relationship between WSS and Ktrans of UIAs and AWE,and to further understand the functional changes in the occurrence,development and rupture of UIAs.Methods:1.We prospectively collected UIAs patients who underwent 3.0T magnetic resonance 4D-flow-MRI,gadolinium-enhanced DCE-MRI,VW-MRI and DSA in our hospital from January 2018 to October 2019.All patients completed 4D-flow-MRI,gadolinium-enhanced DCE-MRI and VW-MRI based on 3.0T Siemens Prisma magnetic resonance scanner(64-channel head and neck combined coil).2.Two experienced neuroimaging physicians independently performed blind analysis of the 4D-flow-MRI images of UIAs through CVI42 5.11.2 software and calculated the hemodynamic index-WSS of the corresponding aneurysm.3.Two experienced neuroimaging physicians independently performed blind analysis on the DCE-MRI images of UIAs through the Siemens syngo.via workstation TISSUE 4D software and calculated the full quantitative pharmacokinetic index of corresponding aneurysm-Ktrans.4.Two senior neuroimaging physicians independently performed blind reading on the PACS platform to determine whether AWE appeared in UIAs,and further classified AWEP:AWEP 0(none),AWEP 1(focal),and AWEP 2(circumferential).5.Two experienced neuroimaging physicians independently conducted a blind analysis of the VW-MRI images of UIAs through Vessel-MASS 2014-EXP software and calculated the semi-quantitative index of AWE-WEI.6.A neuroimaging physician calculated the PHASES score to quantitatively assess the rupture risk of each UIAs.7.We used MedCalc 18.2 software for statistical processing.Categorical variables were represented by percentages,and continuous variables were represented by x ± s or median(interquartile range).Kappa test was used to evaluate the consistency of AWE and its classification by 2 physicians.We used Bland-Altman diagram,Passing and Bablok regression,and ICC to evaluate the consistency of WSS,Ktrans,and WEI among the 2 medical practitioners.The χ2 test was used to compare categorical variables between groups,and the U test was used to compare continuous variables between groups.Selected target variables and included univariate logistic regression analysis.Variables with a univariate regression analysis result P<0.05 were included in the multivariate logistic regression analysis to obtain 95%CI and OR that predict the AWE variable leading to aneurysm wall.Spearman correlation analysis was used to analyze the correlation between continuous variables.A scatterplot with regression line was used to show the quantity change trend among continuous variables.P<0.05 was considered statistically significant.Result:1.This study included 78 patients with 96 UIAs.Inter-reader agreement was excellent for both the presence of AWE(k=0.85[95%CI:0.75 to 0.96])and the calculation of WSS(ICC=0.94[95%CI:0.91 to 0.96]),Ktrans(ICC=0.97[95%cCI:0.95 to 0.98])and WEI(ICC=0.99[95%CI:0.99 to 1]).2.Compared with UIAs without AWE,the median of WSS of UIAs with AWE was lower(1.8 versus 3.4 Pa;P<0.001),the median Ktrans of UIAs with AWE was higher(2.3 versus 0.9 s-1:P<0.001),the median WEI of UIAs with AWE was higher(4.2 versus 0.8;P<0.001)and the average number of PHASES scores of UIAs with AWE was higher(7.3 versus 3.3;P<0.001).3.Multivariate logistic regression revealed that WSS(OR,0.70;95%CI,0.52 to 0.93;P=0.011)was the independent protective factors associated with AWE and Ktrans(OR,1.69,;95%CI,1.17 to 2.44;P=0.01)was the only independent risk factors associated with AWE.4.There was a negative correlation between the WEI and the WSS(rs=-0.5,P<0.001,95%confidence interval:-0.59 to-0.27),a positive correlation between the WEI and the Ktrans(rs=0.4,P<0.001,95%CI:0.24 to 0.57)and a positive correlation between WEI and PHASES scores(rs=0.5,P<0.001,95%CI:0.31 to 0.62).Conclusion1.WSS with AWE group was lower than that without AWE group.The Ktrans,WEI and PHASES scores of the group with AWE were higher than those without AWE.WEI had a negative correlation with WSS.WEI was positively correlated with Ktrans and PHASES scores.2.WSS was an independent protective factor for AWE in UIAs,and Ktrans was an independent risk factor for AWE in UIAs.