Susceptibility-weighted Imaging Based Prospective Cohort Study In Adult Patients With Moyamoya Disease

ObjectiveTo investigate the relationship between cerebral microbleeds(CMBs)and morphology of intracranial arteries as well as subsequent cerebrovascular events in adult patients with moyamoya disease.MethodsOne hundred and thirty five adult patients with first catheter-based angiography conirmed moyamoya disease in Jinling hospital from June 2009 to March 2013 were enrolled in the present study.Clinical baseline parameters in each patient include age,gender,history of hypertension,diabetes,hyperlipidemia and plasma fibrinogen were collated.All of the patients underwent susceptibility-weighted imaging of 3.0T MRI system to detect the CMBs foci.According to the angiographies,the relationship between distribution patterns of CMBs and the morphology of intracranial arteries include Suzuki stages of moyamoya vessels at the base of brain;posterior circulation involvement as well as the stages of dilation and extension of anterior choroid artery-posterior communicating arteries(AChA-PComA)were prospectively analyzed.Furthermore,a follow-up protocol with maximal 5 years were carried out,the log rank and cox hazard proportion model were employed to investigate the relationship between CMBs and subsequent ischemic and hemorrhagic events.Statistical Product and Service Solutions package 20.0 were employed.Quantitative variables were presented with mean(standard deviation,SD),enumeration data with frequency(proportion),levelral data with median(interquartile range).The baseline data and angiographic data were compared between the three groups with one-way ANOVA,Kruskal Wallis non-parametric test,chi-square test or Fisher exact test.All variables with P<0.10 were enrolled into further logistic regression analysis.Kaplan-Meier analysis was employed to investigate the risk of subsequent hemorrhage or ischemic events.Log-rank was employed for the hypothesis testing.Furthermore,the corresponding variables were enrolled into Cox hazard proportion regression model to calculate the hazard ratio.All variables with P value less than 0.05 were considered statistical significant.ResultsA total of 51 CMBs foci in 44 hemispheres were separately found among 34 of the entire 135 patients.62.7%(32/51)CMBs foci located in the region of deep and periventricular white matter(DPWM).The total 135 patients were categorized into non-CMBs group,DPWM-CMBs group and other region-CMBs group according to microbleeds anatomy rating scale.Multinomial logistic regression analysis suggested that compared with non-CMBs group hemispheres,extremely dilation and extension of AChA-PComA is an independent predictor for CMBs located in DPWM(OR=3.45,95%CI 1.31?9.10,P=0.012),rather than in other regions(OR=1.68,95%CI 0.48-5.85,P=0.416).In the total 5 years follow-up,the presence,number as well as the distribution of CMBs foci did not correlated with subsequent ischemic or hemorrhage stroke.However,in the post-hoc analysis,the CMBs located in the region of DPWM was related the subsequent intraventricular hemorrhage(x 2=9.815,P=0.007).Further Cox proportional hazards model indicated that DPWM CMBs is an independent imaging predictor for subsequent intraventricular hemorrhage.(HR=4.12,95%CI 1.24-13.71,P=0.021)ConclusionCMBs in patients with moyamoya disease are not rare and usually locate in the region of DPWM.This specific distribution patterns might be related to the morphology variation of AChA-PComA.DPWM CMBs might be an imaging predictor for subsequent intraventricular hemorrhage.ObjectiveTo investigate the relationship between cortical/periventricular venous sign and morphology of intracranial arteries as well as subsequent cerebrovascular events in adult patients with moyamoya disease.MethodsAccording to diagnostic criterion of moyamoya disease from Japanese ministry of health and welfare,consecutive patients confirmed moyamoya diseases by catheter-based angiography were prospectively enrolled.The clinical baseline and angiographic data were collected.Besides,all patients received 3.0 T susceptibility weighted imaging.The cortical venous sign(CVS)and periventricular venous sign(PVS)in two specific imaging slices were respectively rated based on the combination of numeric and length data.Furthermore,utmost 5 years follow-up was carried out.The relationship between venous sign and subsequent cerebrovascular events was analyzed.Statistical Product and Service Solutions package 20.0 were employed.Quantitative variables were presented with mean(standard deviation,SD),enumeration data with frequency(proportion),levelral data with median(interquartile range).The baseline data and angiographic data were compared between the three groups with students-t test,Mann-Whitney U test,chi-square test or Fisher exact test.All variables with P<0.10 combined with theoretical influenced variables were enrolled into further logistic regression analysis.Kaplan-Meier analysis was employed to investigate the risk of subsequent hemorrhage or ischemic events.Log-rank was employed for the hypothesis testing.Furthermore,the corresponding variables were enrolled into Cox hazard proportion regression model to calculate the hazard ratio.All variables with P value less than 0.05 were considered statistical significant.ResultsA total of 121 patients were finally enrolled.SWI indicated that CVS and PVS were present in 45(37.2%)and 23(19.0%)patients.Logistic regression model indicated that Suzuki stages is independently related to CVS(OR 1.50,95%CI 1.07-2.10,P=0.019)as well as PVS(OR 1.52,95%CI 1.02-2.27,P=0.039).During the follow-up period,13 patients experienced recurrent ischemic stroke.In the total of 32 patients who received surgical treatment,no patients had subsequent ischemic stroke.However,in the none-surgical treatment group,the proportion of ischemic stroke is 14.2%.x 2=6.342,Log Rank P=0.012.Of the non-surgical patients,CVS and PVS individually showed significant relationship to subsequent ischemic stroke with the Log Rank P value of both 0.001(x 2=11.579 and 10.901,respectively).After adjusted for Suzuki stages of moyamoya vessels and posterior circulation involvement,CVS(HR 4.26,95%CI:1.29-14.10)and PVS(HR 3.50,95%CI:1.28-9.54)are also equally independent indicators for subsequent ischemic stroke.On the other hand,Twenty-one patients(17.4%)experienced subsequent hemorrhage.No significant relationship between surgical treatment and subsequent hemorrhage,however,could be found(x2=1.875,Log Rank P=0.171)in the total 121 patients with MMD.ConclusionThe combined evaluation of numeric and length data of venous on susceptibility weighted imaging significantly correlated with Suzuki stages of moyamoya vessels at the base of the brain.Surgical procedure is the dominator prevention methods for subsequent ischemic.For patients who did not revive surgical treatment,venous sign could be independent predictors for subsequent ischemic stroke.ObjectiveTo investigate the correlation between the cognitive impairment of adult patients with moyamoya disease and the correlation of susceptibility weighted imaging parameters such as microbleeds,cortical vessel sign and periventricular vessel sign.MethodsAccording to diagnostic criterion of moyamoya disease from Japanese ministry of health and welfare,consecutive patients confirmed moyamoya diseases by catheter-based angiography were prospectively checked.Patients were regularly followed up based on the protocol on part 1 and 2 of our study.All patients were asked for imaging follow-up in our institution after 24 months after onset.Patients who did not experience recurrent cerebrovascular events during follow up were enrolled into our study.In addition,35 age,sex and education matched healthy controls as well as 35 age,sex,education and ischemic/hemorrhage ratio matched patients with chronic events were prospectively enrolled.The cognitive tests recommend by National Institute for Neurological Disorders and Stroke-Canadian Stroke Network was performed in individuals.The difference between MMD patients and controls was analyzed.Furthermore,all patients with MMD were scaned repeatedly by susceptibility weighted imaging(SWI)of magnetic resonance imaging.The relationship between imaging parameters include microbleeds,CVS,PVS and cognitive decline was also analyzed.Z scores were employed to evaluate all kinds of congnitive function.Quantitative variables were presented with mean(standard deviation,SD),enumeration data with frequency(proportion),levelral data with median(interquartile range).Comparasion between baseline data were carried out by student-t test,x 2 test and Fisher exact test.All patients with MMD were ranked by the z scores then classified equally into two groups accordingly.Binary logistic regression was performed to investigate the correlation between the imaging markers and cognitive impairment.P value less than 0.05 was considered significant.ResultsThirty five adult patients with MMD were finally enrolled with mean age 41.51 years old,60%female and 71.4%with more than 9 years of education.SWI indicated that 7 patients were microbleeds positive,15 patients were CVS positive and 5 patients were PVS positive.Magnetic resonance angiography(MRA)indicated the median MRA score of MMD was 5(3)and it correlates significantly with Suzuki results of catheter-based angiography 24 months ago.35 healthy controls and 35 patients with cerebrovascular events were also enrolled.Compared with the healthy controls,in all of the 8 tests,patients with MMD is significantly worse.Compared with cerebrovascular disease control group,executive/activation and language capability of patients with MMD is significantly worse.Visuospatial and memory is equal to that of cerebrovascular control group.Moreover,patients with MMD were ranked according to the individual z score of executive/activation and memory capability and categorized into cognitive impairment group and cognitive reservation group taking the median rank as the cut off points.The baseline and imaging parameters were compared between the two groups.The results indicated that surgical procedure is the only protective factor for cognitive function.(executive/activation impairment 11.8%vs.reservation 55.6%,x 2=7.441,P=0.012,language impairment 11.8%vs.reservation 55.6%,x 2=7.441,P=0.012).Cerebral microbleeds,CVS and PVS correlate with neither executive/activation nor language function.After adjusting for surgical procedures,none of the microbleeds,CVS,PVS independently correlates with cognitive impairment(microbleeds and executive/activation OR 0.61,95%CI 0.10?3.79,P=0.592;microbleeds and language OR 1.50,95%CI 0.22?10.30,P=0.680;CVS and executive/activation OR 2.01,95%CI 0.40-10.30,P=0.402;CVS and language OR 2.01,95%CI 0.40?10.30,P=0.402;PVS and executive/activation OR 2.55,95%CI 0.23?27.71,P=0.443;PVS and language OR 2.55,95%CI 0.23?27.71,P=0.443).Conclusion:The cognitive function of adult patients with MMD is significantly declined.Compared with atherosclerosis cerebrovascular disease,the executive/activation and language function decline is the typical character of cognitive impairment.Surgical treatment is the only protective factors for cognitive function.None of microbleeds,CVS and PVS correlates with cognitive function.