颅外-颅内联合脑血运重建术治疗出血型烟雾病的远期临床预后及脑灌注状态研究

Research of long-term clinical prognosis and cerebral perfusion of hemorrhagic Moyamoya disease after combined extra-intracranial bypass

目的 研究颅外-颅内联合脑血运重建术对出血型烟雾病的远期作用。方法 回顾性连续纳入2018年1月至2019年1月就诊于南京鼓楼医院神经外科行颞浅动脉-大脑中动脉旁路移植术+脑硬膜动脉贴敷术的成人出血型烟雾病患者。收集患者的一般资料及临床资料,包括性别、年龄、高血压病史、糖尿病史、吸烟史、出血位置(蛛网膜下腔、单纯脑室、脑实质)、出血半球、术前是否反复出血、改良Rankin量表(mRS)评分、头部CT灌注(CTP)检查数据及DSA资料等。术后近期(3~6个月)随访时收集MR血管成像(MRA)资料,评价旁路移植血管通畅性;新近一次远期(术后>12个月)随访时收集是否存在新发卒中事件(短暂性脑缺血发作、脑梗死、脑出血等)、mRS评分、头部CTP数据及DSA资料等。通过头部DSA评价手术侧半球改良铃木分期,铃木分期Ⅰ、Ⅱ期为早期,Ⅲ、Ⅳ期为中期,Ⅴ、Ⅵ期为晚期。根据手术侧的大脑半球DSA对烟雾血管浓集程度分级,大脑半球底部未见烟雾血管定义为无,虽在脑底形成了烟雾血管,但比较稀疏定义为稀疏,脑底部可见大量的烟雾血管在颈内动脉末端吻合,形成烟雾网络,并在脑底向各个方向扩展定义为浓集。通过远期随访头部DSA评价旁路移植血管通畅性,并与术后近期随访MRA比较,若旁路移植血管显影纤细或不显影,则定义为纤细或萎缩,若旁路移植血管显影扩张或形态不变,则定义为扩张或不变。基于大脑中动脉(侧位)的DSA,使用松岛分级评价新生代偿血管,以颈外动脉造影的新生血管分布范围占大脑中动脉分布区域比例进行分级,A级为>2/3,B级为1/3~2/3,C级为<1/3。与术前相比,若远期随访时mRS评分降低,则定义为神经功能预后改善,评分不变定义为神经功能无变化,评分增加定义为神经功能恶化;神经功能结局以mRS评分为评价标准,0~1分为无残疾,2~3分为轻度至中度残疾,4~5分为重度残疾,6分为死亡。根据患者术前及术后远期随访时大脑双侧半球的CTP资料分别划分手术侧和手术对侧半球感兴趣区(ROI)。各ROI分别采集脑血容量(CBV)、脑血流量(CBF)、平均通过时间(MTT)和达峰时间(Tmax)4个脑灌注参数。各ROI灌注参数分别与小脑齿状核平面的小脑皮质(ROI 3)相应参数做相对值,得出相对脑血容量(rCBV)、相对脑血流量(rCBF)、相对平均通过时间(rMTT)和相对达峰时间(rTmax),以比较半球间脑灌注情况。结果 (1)本研究共纳入35例成人出血型烟雾病患者,其中男23例,女12例;年龄21~65岁,平均(38±13)岁;高血压病11例(31.4%),糖尿病7例(20.0%),吸烟史9例(25.7%);出血半球左侧17例(48.6%),右侧18例(51.4%);出血位置包括脑实质出血9例(25.7%),单纯脑室出血21例(60.0%),蛛网膜下腔出血5例(14.3%)。(2)纳入患者随访36~48个月,平均随访时间(40±9)个月。随访出现新发脑出血5例(14.3%),包括脑室出血3例,脑实质出血2例;出血时间在手术后15~40个月,其中手术侧出血3例(8.6%),对侧出血2例;另有3例患者出现新发脑缺血性事件,均为短暂性脑缺血发作。远期随访时旁路移植血管通畅25例(71.4%),纤细或萎缩10例(28.6%);新生代偿血管A级12例(34.3%),B级16例(45.7%),C级7例(20.0%)。与术前比较,远期随访神经功能预后改善23例(65.7%),无变化7例(20.0%),恶化5例(14.3%)。与术前比较,远期随访时神经功能结局明显改善(P=0.013),无残疾比例为80.0%;远期随访改良铃木分期分布不同(P<0.01),以晚期比例最大(77.1%),中期比例较少(20.0%);mRS评分(P=0.622)及烟雾血管密度(P=0.155)差异均无统计学意义。(3)与术前比较,远期随访CTP术侧尾状核[(0.94±0.07)比(1.22±0.11),t=-23.664,P=0.04]、纹状体rCBF[(0.99±0.06)比(1.23±0.08),t=-23.654,P=0.03]减少;各ROI rCBV差异均无统计学意义(均P>0.05);胼胝体膝[(1.21±0.10)比(0.96±0.07),t=14.790,P=0.03]、内囊前肢[(1.17±0.07)比(0.95±0.06),t=18.593,P=0.02)] rMTT延长;rTmax差异在前峡部[(0.94±0.03)比(1.03±0.03),t=-17.750,P=0.049]、内囊后肢[(1.04±0.05)比(0.92±0.03),t=14.210,P=0.04)]、额叶[(1.06±0.03)比(1.05±0.04),t=-1.972,P=0.03]和顶叶[(0.97±0.04)比(1.09±0.04),t=-17.750,P=0.01]均有统计学意义。(4)与术前比较,远期随访时非手术侧rCBF在扣带回[(1.17±0.16)比(1.06±0.08),t=4.067,P<0.01]、胼胝体膝[(1.29±0.22)比(1.05±0.07),t=6.454,P<0.01]、前峡部[(1.16±0.15)比(1.05±0.08),t=4.338,P=0.02]、苍白球[(1.13±0.14)比(1.08±0.08),t=2.113,P=0.01]增加,在纹状体显著降低[(0.87±0.06)比(1.05±0.06),t=-17.750,P=0.02)];顶叶rCBV差异有统计学意义[(0.97±0.04)比(1.04±0.05),t=-10.353,P=0.02];各部位rMTT差异均无统计学意义(均P>0.05);苍白球rTmax差异有统计学意义[(1.00±0.02)比(1.01±0.02),t=-2.958,P=0.04]。结论 颅外-颅内联合脑血运重建术可改善出血型烟雾病远期神经功能预后,促进铃木分期进展至高分期,对脑深部血管网络的灌注影响较小。出血型烟雾病手术对侧的脑血流动力学变化仍有待进一步研究。

Objective To investigate the long-term effects of combined extracranial-intracranial bypass on hemorrhagic Moyamoya disease(hMMD).Methods Adults with hMMD treated by superficial temporal artery-middle cerebral artery bypass and encephalo-duro-synangiosis(EDAS)who admitted to the Department of Neurosurgery,Nanjing Drum Tower Hospital from January 2018 to January 2019 were retrospectively reviewed.General and clinical data of the hMMD patients were collected,including gender,age,history of hypertension,history of diabetes mellitus,history of smoking,location of hemorrhage(subarachnoid space,simple ventricle,brain parenchyma),hemisphere of bleeding,history of recurrent hemorrhage,modified Rankin Scale(mRS)scores,hemispheric data of CTP examination,and DSA.MR angiography(MRA)was collected at short-term follow-up(3-6 months after bypass)to evaluate patency of direct bypass graft.Moreover,data of long-term follow-up(LTFU),characterized by over 12 months after bypass,were also collected including presence of new-onset of stroke events(transient ischemic attack,cerebral infarction,cerebral hemorrhage,etc.),mRS scores,hemispheric CTP data,and DSA data.The modified Suzuki stage of the ipsilateral hemisphere was evaluated by DSA,in which Suzuki stagesⅠandⅡwere considered as early stage,Suzuki stagesⅢandⅣas middle stage,and Suzuki stagesⅤandⅥas advanced stage.The degree of Moyamoya vessel concentration was graded according to the DSA of the ipsilateral hemisphere.The absence of Moyamoya vessels at the base of the cerebral hemisphere was defined as absent.Although Moyamoya vessels were formed at the base of the brain,they were sparse and defined as sparse.The dense Moyamoya vessels was defined as a large number of smoke vessels formed at the base of the brain anastomosing at the end of the internal carotid artery to form a network of Moyamoya vessels,which was extended in all directions at the base of the brain.Compared with short-term follow-up of MRA,bypass patency was evaluated by follow-up cerebral DSA which was defined as slender or atrophic if the bypass graft appeared thin or absent,and as dilated or unchanged if the bypass graft appeared dilated or unchanged in morphology.Based on the lateral cerebral angiographic images of the middle cerebral artery(MCA),neovascularization was evaluated using the Matsushima grading system,which was based on the extent of neovascularization of the external carotid artery angiogram as a proportion of the area of distribution of the MCA,with grade A of>2/3,grade B of 1/3-2/3,and grade C of<1/3.Compared to mRS score before surgery,a decrease in the mRS score at LTFU was defined as an improvement in the prognosis of neurological function,a score of no change was defined as no change in neurological function,and an increase of score was defined as deterioration in neurological function.Neurological outcome was also evaluated by the mRS score,with 0 to 1 as no disability,2 to 3 as mild to moderate disability,4 to 5 as severe disability,and 6 as death.The regions of interest(ROIs)were divided according to the CTP data of the bilateral hemispheres at the preoperative and LTFU timepoints.Four cerebral perfusion parameters,cerebral blood volume(CBV),cerebral blood flow(CBF),mean transit time(MTT)and time to peak(Tmax)were collected for each ROI.Each ROI perfusion parameter was made relative to cerebellar cortex(ROI 3)of the cerebellar dentate nucleus to derive the relative cerebral blood volume(rCBV),relative cerebral blood flow(rCBF),relative mean time to passage(rMTT),and relative time to peak(rTmax)for hemispheric comparison.Results(1)A total of 35 adults(23 males and 12 females)with hMMD were included,aged 21-65 years old,mean(38±13)years old.Eleven cases of hypertension(31.4%),7 cases of diabetes mellitus(20.0%),and 9 cases of history of smoking(25.7%)were observed.Seventeen cases(48.6%)of hemorrhagic hemisphere on the left side,and 18 cases(51.4%)of hemorrhagic hemisphere on the right side were recorded.Hemorrhagic location included parenchymal hemorrhage in 9 cases(25.7%),ventricular hemorrhage in 21 cases(60.0%),and subarachnoid hemorrhage in 5 cases(14.3%).(2)The hMMD patients recruited were followed up for 36-48 months,with a mean follow-up time of(40±9)months.Five(14.3%)new cerebral hemorrhages occurred,including ventricular hemorrhages of 3 cases and parenchymal hemorrhages of 2 cases.The time of hemorrhage ranged from 15 to 40 months after bypass,including 3 hemorrhages(8.6%)on the operated hemisphere and 2 hemorrhages on the contralateral side.Three patients developed new cerebral ischemic events,all of them were transient ischemic attack.At LTFU,the bypass graft was patent in 25 cases(71.4%)and slender or atrophic in 10 cases(28.6%).The neovascularization was 12 cases(34.3%)of grade A,16 cases(45.7%)of grade B,and 7 cases(20.0%)of grade C.Compared with the preoperative mRS,the prognosis of neurological function at LTFU improved in 23 cases(65.7%),no change in 7 cases(20.0%),and deteriorated in 5 cases(14.3%).Compared with the preoperative mRS,the neurological function outcome was significantly improved at the LTFU(P=0.013),and the proportion of no disability was 80.0%.The distribution of the modified Suzuki staging at LTFU had significant difference(P<0.01),with the largest proportion in the advanced stage(77.1%),and smaller proportion of middle stage(20.0%).There was no significant difference in comparison of mRS scores and the density of Moyamoya vessels(P=0.155).(3)Compared with the preoperative CTP data,the caudate nucleus([0.94±0.07]vs.[1.22±0.11],t=-23.664,P=0.04)and striatal rCBF([0.99±0.06]vs.[1.23±0.08],t=-23.654,P=0.03)were reduced on the operated hemisphere in the LTFU.The differences in the rCBV of the respective ROIs were not statistically significant(all P>0.05).The genu of corpus callosum([1.21±0.10]vs.[0.96±0.07],t=14.790,P=0.03),and the forelimb of the internal capsule([1.17±0.07]vs.[0.95±0.06],t=18.593,P=0.02)had a prolonged rMTT.The significant differences of rTmax parameter were found in the anterior isthmus([0.94±0.03]vs.[1.03±0.03],t=-17.750,P=0.049),and the posterior of the internal capsule([1.04±0.05]vs.[0.92±0.03],t=14.210,P=0.04),frontal lobe([1.06±0.03]vs.[1.05±0.04],t=1.972,P=0.03),and parietal lobe([0.97±0.04]vs.[1.09±0.04],t=-17.750,P=0.01).(4)Compared with the preoperative CTP data,the rCBF of non-operative side at LTFU was statistically significant in the cingulate gyrus([1.17±0.16]vs.[1.06±0.08],t=4.067,P<0.01),corpus callosum([1.29±0.22]vs.[1.05±0.07],t=6.454,P<0.01),anterior isthmus([1.16±0.15]vs.[1.05±0.08],t=4.338,P=0.02),pale bulb([1.13±0.14]vs.[1.08±0.08],t=2.113,P=0.01)increased,and significantly decreased in the striatum([0.87±0.06]vs.[1.05±0.06],t=-17.750,P=0.02).The difference in parietal rCBV was statistically significant([0.97±0.04]vs.[1.04±0.05],t=-10.353,P=0.02).The difference in rMTT was not statistically significant at all ROIs(all P>0.05).The difference in globus pallidus rTmax was statistically significant([1.00±0.02]vs.[1.01±0.02],t=-2.958,P=0.04).Conclusions Combined extracranial-intracranial bypass improved the long-term neurological prognosis of adults with hMMD,promoted Suzuki stage to advanced stage,and posed less impact on the perfusion of the deep cerebral vascular network.Cerebral hemodynamics of contralateral hemisphere for hMMD still needed further research.