节点文献

兔脑微栓塞模型的建立及血流动力学的CT灌注动态变化研究

Establishment of Cerebral Microembolic Model in Rabbits and Dynamic Changes Study of Hemodynamics by CT Perfusion Imaging

【作者】 张放

【导师】 冯晓源; 姚振威; 刘含秋;

【作者基本信息】 复旦大学 , 影像医学与核医学, 2008, 博士

【摘要】 脑梗死是严重危害人类生命健康的常见病、多发病,具有较高的死亡率和致残率,目前除了溶栓治疗外还没有有效的治疗方法,脑梗死的早期诊断、溶栓的适应症、途径及溶栓前后的病理生理变化等问题也有待进一步研究。这也说明了加强脑梗死基础研究的重要性和紧迫性。栓子脱落阻塞血管是脑梗死发病的重要机制,制作模拟人类脑动脉栓塞的动物模型,可为临床研究提供实验性理论依据。兔脑血管结构与人脑结构相似,相对于大鼠而言,兔脑体积较大,较适合神经影像学研究,且兔体形适中,性情温和,易于手术和标本采集,因此选用兔制作动物模型。以往多采用线栓法制作大脑中动脉闭塞模型,但该方法的局限性是只能引起大脑中动脉供血范围的大面积脑梗死。制作微栓塞模型能够模拟人类微栓子脱落的过程,观察微栓子随血液流动引起的脑损伤,并进一步探讨微栓塞的病理生理学改变及影像学特点。CT灌注成像具有成像速度快,时间和空间分辨率高,操作简单易行等优点,与MR灌注成像相比灌注功能图像质量好,且稳定性高,因此促成了CT灌注成像的研究。第一部分正常兔脑血流动力学的多层螺旋CT灌注成像研究目的:探讨多层螺旋CT灌注成像(CT perfusion imaging,CTPI)技术在正常兔脑血流动力学中的应用。材料与方法:应用GE Light Speed 16多层螺旋CT机对20只健康新西兰种大白兔行CT灌注成像研究,体重2.5~3.0 kg。先行CT普通平扫,然后选择基底节及相邻层面行2cm范围的CT灌注扫描。扫描参数:层厚2.5mm,管电压120kV,管电流150mA,矩阵512×512,FOV 9.6cm,经股静脉途径以1.5 ml/s速率注入5ml对比剂欧乃派克(omnipaque,300mg/ml),扫描延迟时间5s,连续30s动态扫描,图像重建间隔时间0.5s,共获得472幅图像。使用GE perfusion3.0脑灌注软件包进行图像后处理,获得兔脑8个相邻层面的脑血流量(cerebral blood flow,CBF)图、脑血容积(cerebral blood volume,CBV)图和平均通过时间(mean transit time,MTT)图。分别选取左右两侧相对应部位的额叶、颞叶、顶叶、基底节、脑干及小脑,测量各自的CBF、CBV及MTT值,并利用SPSS11.0统计学软件分析左右两侧及不同部位间CBF、CBV及MTT值是否存在显著差异,以P<0.05作为有统计学意义标准。结果:20只兔均可直接进入自动调节阈值界面,生成的CBF,CBV及MTT图伪彩色对称,无异常灌注缺损区,灌注范围包括额叶至小脑大部分脑组织。经配对t检验,脑内各部位CBF值(t=0.517,P=0.608)、CBV值(t=1.581,P=0.120)及MTT值(t=-0.541,P=0.591)在两侧半球间无统计学差异;经单因素方差分析,基底节及脑干灌注高于额叶、颞叶、顶叶及小脑,不同部位间CBF值(F=3.059,P=0.013)及MTT值(F=2.493,P=0.036)差异具有统计学意义,CBV值差异无统计学意义(F=1.189,P=0.320)。正常兔脑平均CBF、CBV、MTT值分别为(115.45±36.83)ml/(min·100g)、(3.85±1.20)ml/100g、(2.43±0.81)s。结论:采用经股静脉注入对比剂的方法可以获得满意的CT灌注图像;MSCT灌注成像技术为兔脑血流动力学研究提供了一种新的无创性手段,所获得的血流动力学参数值具有可重复性,为各种疾病兔脑模型的CT灌注研究奠定了基础。第二部分急性兔脑微栓塞模型的建立及CT灌注评价目的:建立稳定的兔脑微栓塞模型,以期为进一步开展脑微栓塞的病理生理变化及影像学研究提供可靠实用的工具;探讨CTPI技术对早期脑缺血的应用价值。材料与方法:30只健康新西兰种大白兔,体重2.5~3.0 kg,随机分成2组,其中A组5只,为假手术对照组;B组25只,为微栓塞组。分离右侧颈部血管,观察兔颈部血管的解剖及变异,经颈外动脉向颈内动脉注入直径约0.5mm的SiO2颗粒10枚,栓塞后30min行CTPI,24小时处死动物取脑组织,10%甲醛液中固定,一周后行5mm厚冠状位切片,并取材经HE染色及石蜡包埋。CT灌注成像方法及图像后处理方法同第一部分。在灌注功能图上测量灌注异常侧与正常对侧的CBF、CBV及MTT值,并进行统计学分析。对照CTPI及HE染色结果,利用SPSS11.0统计学软件进行CBF、CBV及MTT ROC曲线分析,评价各灌注参数诊断早期脑缺血的准确性。结果:右侧颈总动脉在下颌角处分为颈内动脉和颈外动脉,颈内动脉较颈外动脉细小,且颈内动脉起始部可见膨大的颈动脉窦,本组实验30只兔颈内动脉起源均为Ⅰ型(外侧起源型),其中ⅠA型(枕动脉起源于颈外动脉)占73.3%(22/30),ⅠB型(枕动脉起源于颈内动脉近端)占26.7%(8/30)。颈内动脉及颈外动脉平均直径分别为1.15±0.27mm及1.93±0.32mm。A组CTPI未见明显异常灌注区。B组中有3只因实验意外死亡,1只因下肢静脉穿刺失败导致CTPI检查失败,21只行CTPI检查,其中18只CTPI异常,HE染色显示10只脑梗死,7只脑缺血,1只未见明显异常;3只CTPI未见明显异常,HE染色显示1只轻度脑缺血,2只未见明显异常。灌注异常主要表现为局部CBF降低、MTT延长、CBV无明显变化或轻度上升、下降,包括5只右侧额、颞叶片状低灌注;4只右侧额叶、颞叶、基底节及脑干片状低灌注;7只右侧颞叶小片低灌注;2只双侧额颞叶小片低灌注。经配对t检验,灌注异常侧与正常对侧CBF值(t=-7.286,P=0.000)、MTT值(t=5.556,P=0.000)比较有统计学差异,CBV值(t=-1.080,P=0.287)无统计学差异,患侧CBF平均减少为对侧的38.7%±23.7%(1%~83%),MTT较对侧延长6.22±5.36s(0.08s~18.33s)。ROC曲线分析提示MTT诊断早期脑缺血的准确性略高于CBF,CBF和MTT的准确性均明显高于CBV。CBF<67.06ml/(min·100g),诊断早期脑缺血的灵敏度为84.6%,特异度为84.6%;CBV<3.08ml/100g,诊断早期脑缺血的灵敏度为74.4%,特异度为74.4%;MTT>4.58s,诊断早期脑缺血的灵敏度为87.2%,特异度为87.2%。结论:兔脑血管结构与人脑结构相似,相对于大鼠而言,兔脑体积较大,且兔体形适中,性情温和,易于手术和标本采集,可用兔制作脑缺血模型;采用经颈外动脉向颈内动脉注入微栓子的方法能够建立基本稳定的兔脑微栓塞模型;CBF及MTT诊断早期脑缺血具有较高的准确性,CTPI技术可以准确、无创地反映兔局部脑缺血模型的血流动力学变化。第三部分急性兔脑微栓塞模型CT灌注成像各参数的时间变化规律目的:探讨急性兔脑微栓塞模型CT灌注成像各参数的动态变化规律,评价CTPI技术区分脑缺血和脑梗死的价值。材料与方法:对栓塞后30minCTPI异常的18只兔分别于栓塞后3h、6h、12h及24h行CT灌注成像,根据HE染色结果将模型分为缺血组和梗死组,分别观察缺血组和梗死组兔脑CT灌注成像各参数的时间变化规律。CT灌注成像方法及图像后处理方法同第一部分。在灌注功能图上测量不同时间灌注异常侧与正常对侧的CBF、CBV及MTT值。利用SPSS11.0统计学软件进行CBF、CBV及MTT ROC曲线分析,评价各灌注参数诊断脑梗死的准确性。结果:30min时缺血兔脑不同程度低灌注,表现为CBF值不同程度降低,MTT值不同程度延长,CBV值无显著变化,3~6h低灌注进一步加重,CBV值略降低,12h低灌注不同程度恢复,24h进一步恢复。不同时间之间CBF及MTT值均有显著性差异(F=20.08,P=0.001;F=13.22,P=0.002),CBV值无显著性差异(F=1.34,P=0.267)。30min、3h、6h、12h及24h缺血侧与正常对侧CBV值比较均无显著性差异,30min、3h、6h及12h缺血侧与正常对侧CBF及MTT值比较均有显著性差异,24h缺血侧与正常对侧CBF及MTT值比较无显著性差异。30min梗死兔脑明显低灌注,表现为CBF及CBV值显著降低,MTT值显著延长,3只兔低灌注分别在3h、6h及12h不同程度恢复,然后下一时间又迅速降低并随着时间延长进一步加剧,其余7只兔低灌注程度随时间延长逐渐加剧或在一定水平上波动。不同时间之间CBF及CBV值比较均无显著性差异(F=2.01,P=0.097;F=2.399,P=0.054),MTT值有显著性差异(F=5.39,P=0.001)。30min、3h、6h、12h及24h梗死侧与正常对侧CBF、CBV及MTT值比较均有显著性差异。脑缺血与脑梗死的平均CBF、CBV和MTT值分别为(58.08+20.81)ml/(min·100g)、(3.29±0.62)ml/100g、(6.03±2.50)s及(16.47±14.91)ml/(min·100g)、(1.87±1.04)ml/100g、(13.69±5.15)s,差异具有统计学意义。ROC曲线分析提示CBF诊断脑梗死的准确性明显高于CBV,MTT略高于CBF。CBF<27.07ml/(min·100g),诊断脑梗死的灵敏度为82.8%,特异度为82.3%;CBV<2.48ml/100g,诊断脑梗死的灵敏度为73.8%,特异度为74.2%;MTT>7.77s,诊断脑梗死的灵敏度为89.3%,特异度为85.5%。结论:脑缺血3~6h低灌注最明显,12~24h低灌注不同程度恢复,而脑梗死随时间延长低灌注程度不断加重或一过性恢复后再次加重。脑缺血的特征是CBF和CBV的不匹配,缺血组织CBF显著降低,CBV无显著变化,而脑梗死则表现为这两个参数的一致性下降。CBF<27.07ml/(min·100g),MTT>7.77s诊断脑梗死有很高的准确性,可用于区分脑缺血和脑梗死。

【Abstract】 Establishment of cerebral microembolic model in rabbits and dynamic changes study of hemodynamics by CT perfusion imagingCerebral infarct is the common disease leading to death and disability,there is no effective therapy method except thrombolytic presently.And the diagnosis,time window and approach of thrombolysis and the changes of pathology,physiology after therapy are needed further investigation,which also shows the significance and instancy of investigating cerebral infarct.The cerebral vessel blocked by embolus is the crucial mechanism leading to cerebral infarct.And establishing a stable embolic model of brain could provide a reliable tool for clinical study.The cerebral vascular anatomy of rabbit is similar to that of humans and the volume of brain is larger than rat,so it is suitable for the study of nervous system.In the other hand,the figure of rabbit is moderate and the temper is mild which is convenient for surgery and operation,therefore,the rabbit could be used to establish cerebral ischemic model.The model of middle cerebral artery occlusion by an intraluminal thread was applied broadly;however,the limitation of this method was that it only could lead to broad cerebral infarct in the area supplied by middle cerebral artery.Establishing the cerebral microembolic model could simulate the falling of microembolus in humans,observe the cerebral injury caused by microembolus and evaluate the pathology,physiology and medical imaging of cerebral microembolism.CT perfusion imaging has a number of practical advantages:1) It can be performed easily no special equipment,except a power injector,is required;2) examination time is considerably short;3) it has a higher spatial and temporal resolution.On the other hand, compared with MR perfusion imaging,the functional maps of CT perfusion are better and more stable;therefore,CT perfusion imaging was applied in our study.PartⅠMulti-slice spiral CT perfusion imaging in cerebral hemodynamies of normal rabbitsPurpose:To evaluate the application of multi-slice spiral CT perfusion imaging in normal rabbits of cerebral hemodynamics.Materials and Methods:CT perfusion imaging(CTPI) was performed in twenty normal New Zealand rabbits(2.5~3.0 kg) with GE Light Speed 16 multi-slice spiral CT scanner.Firstly,routine CT scans were performed,and then CTPI study was made by selecting the basal ganglia slice and its adjacent slice.Scanning was performed at 120 kV,150 mA,with a 512×512 matrix, 9.6-cm FOV,and 2.5-mm section thickness.5ml of iodinated contrast agent(omnipaque, 300mg/ml) was bolus injected through femoral vein at rate of 1.5 ml/s,and multilayer of dynamic scan was performed for 30 seconds at a delayed time of 5 seconds.Four hundred and seventy-two maps at an interval of 0.5 second were acquired.The post-processing was performed with GE perfusion3.0 cerebral perfusion software.Cerebral blood flow, cerebral blood volume and mean transit time maps were acquired.The specific region of interest in both sides of frontal,temporal,parietal lobes,basal ganglia,cerebral stem and cerebellum were selected,the values of CBF,CBV,and MTT were measured,and quantitative analysis was performed.The data was analyzed using the software of SPSS 11.0 and P<0.05 was taken to indicate statistical difference.Results:Twenty rabbits all entered the auto-regulated threshold value interface and the acquired maps of CBF, CBV,and MTT had no abnormal perfusion,including mostly cerebral tissue from frontal lobe to cerebellum.By paired-samples t test,there was no significant difference in the both sides of CBV(t=0.517,P=-0.608),CBV(t=1.581,P=0.120),and MTT(t=-0.541, P=-0.591).By one-way ANOVA,the perfusion in basal ganglia and cerebral stem was higher than in frontal,temporal,parietal lobes and cerebellum.And in different lotions, the values of CBF(F= 3.059,P=0.013)and MTT(F= 2.493,P=0.036)have significant difference,while the value of CBV has no significant difference(F = 1.189,P=0.320).In normal brain of rabbits,the values of CBF,CBV,and MTT are(115.45±36.83) ml/(min·100g),(3.85±1.20) ml/100g and(2.43±0.81)s respectively.Conclusion:The satisfying perfusion image could be obtained by injecting contrast material through femoral vein.MSCTPI provides a new noninvasive imaging method for measuring the cerebral hemodynamics in normal rabbits.And the acquired values of cerebral hemodynamics are reproducible which established basis for the CT perfusion investigation in all kinds of diseases cerebral mode of rabbits. PartⅡEstablishment of acute cerebral microembolic model in rabbits and evaluation by CT perfusion imagingPurpose:To establish a highly stable microembolic model of brain in rabbits,provide a reliable tool for investigating the pathology,physiology and medical imaging of cerebral embolism,and to evaluate the application of computed tomography perfusion imaging in acute ischemic cerebrovascular disease.Materials and Methods:Thirty normal New Zealand white rabbits(2.5~3.0 kg) were randomly divided into two groups:Group A(n = 5) underwent sham operation,group B(n = 25) underwent an operation of microembolic brain injury.Right cervical vessels were separated in thirty New Zealand white rabbits,the anatomy and variations of the rabbit carotid artery were recorded,and about ten SiO2 grains(D=0.5mm) were injected into internal carotid artery by external carotid artery in group B.After 30 minutes,CT perfusion imaging was performed.After 24 hours,the brain was removed,fixed in 10%formalin.After one week,the brain was cut into 5-mm sections and HE stain was performed.The CT imaging procedures and post-processing are same to the part one.Values of CBF,CBV,and MTT were measured in the lesions and the normal contra-lateral regions and statistic analysis was performed. The results of CTPI were compared with HE stain,and the curves of receiver operating characteristic of CBV,CBV and MTT were performed by SPSS11.0 statistic software, the accuracy of diagnosing acute cerebral ischemia was evaluated.Results:The right common carotid artery is divided into internal carotid artery and external carotid artery in gonial.Internal carotid artery is thinner than external carotid artery and the dilated carotid sinus is seen in the origination of internal carotid artery.In our study,origins of the internal carotid artery in thirty rabbits are all typeⅠ(lateral origin).Subtype A(the occipital artery origins from the external carotid artery) accounts for 73.3%(22/30),and subtype B(the occipital artery origins proximal on the internal carotid artery) accounts for 26.7%(8/30).The mean diameters of internal carotid artery and external carotid artery are 1.15±0.27mm and 1.93±0.32mm respectively.No abnormity was seen on CTPI in group A.Among group B,three rabbits were died during the experiment,CTPI was defeated in one rabbit because of the failure of puncture of femoral vein,and CTPI was succeeded in twenty-one rabbits,including abnormal perfusion in eighteen rabbits and normal perfusion in three rabbits.By HE stain,among the rabbits of abnormal perfusion,ten rabbits were cerebral infarct,seven rabbits were cerebral ischemia and one rabbit was normal.While among the rabbits of normal perfusion,one rabbit was slightly ischemic and two rabbits were normal.Reduced CBF,prolonged MTT and slightly low or high CBV were found on CT perfusion maps.There was broad low perfusion in right frontal and temporal lobes in five rabbits,broad low perfusion in right frontal and temporal lobes,basal ganglia and cerebral stem in four rabbits,local low perfusion in right temporal lobe in seven rabbits and local low perfusion in both side of frontal and temporal lobes in two rabbits.By paired-samples t test,there are significant difference between abnormal side and normal side for CBF(t=-7.286,P=0.000) and MTT(t=5.556, P=0.000),but for CBV,there is no significant difference(t=-1.080,P=0.287).The CBF value in abnormal side was decreased to 38.7%±23.7%(1%~83%),and MTT was delayed 6.22±5.36s(0.08s~18.33s).The accuracy of MTT in diagnosing acute cerebral ischemia is slightly higher than CBF,and the accuracy of both CBF and MTT are remarkably higher than CBV.If the value of CBF is less than 67.06ml/(min.100g),the sensitivity is 84.6%and the specificity is 84.6%in diagnosing cerebral ischemia.If the value of CBV is less than 3.08 ml/100g,the sensitivity is 74.4%and the specificity is 74.4%.And if the value of MTT is more than 4.58s,the sensitivity is 87.2%and the specificity is 87.2%.Conclusions.The cerebral vascular anatomy of rabbit is similar to that of humans and the volume of brain is larger than rat,furthermore,the figure of rabbit is moderate and the temper is mild which is convenient for surgery and operation, therefore,the rabbit could be used to establish cerebral ischemic model.The stable acute cerebral microembolic model in rabbits can be established by injecting microembolus from external carotid artery to internal carotid artery.CBF and MTT have high accuracy in diagnosing cerebral ischemia and CTPI can be as an accurate and non-invasive method in evaluating the acute ischemic cerebrovascular disease.PartⅢSerial changes in CT perfusion parameters of acute microembolic cerebral ischemic model in rabbits.Purpose:To discuss the serial changes in CT perfusion parameters during the first twenty-four hours of microembolic cerebral ischemic model in rabbits and to evaluate the application of computed tomography perfusion imaging in identifying acute cerebral ischemia and cerebral infarct.Materials and Methods:Eighteen rabbits which had abnormal perfusion in 30 minutes after embolism were further studied.CT perfusion imaging was performed continuously in 3 hours,6 hours,12 hours,and 24 hours after embolism.According to the results of HE stain,the rabbits were divided into cerebral ischemia and cerebral infarct.The serial changes in CT perfusion parameters of cerebral ischemia and cerebral infarct were discussed respectively.The CT imaging procedures and post-processing are same to the part one.Values of CBF,CBV,and MTT were measured in the lesions and the normal contra-lateral regions at each time point.The curves of receiver operating characteristic of CBV,CBV and MTT were performed by SPSS 11.0 statistic software and the accuracy of diagnosing cerebral infarct was evaluated. Results:In 30 minutes,the cerebral ischemia had low perfusion with different degrees, and reduced CBF,prolonged MTT and slightly high or normal CBV were found on CT perfusion maps.During 3~6 hours,the descended perfusion was further aggravated,the value of CBV was slightly descended.During 12~24 hours,the low perfusion was lightened.In different time points,there were remarkable difference for CBF and MTT (F=20.08,P=0.001;F=13.22,P=0.002),but for CBV,the difference had no statistic meaning(F=1.34,P=0.267).For CBV,there was no statistic difference between ischemia side and normal contra-lateral side at 30minutes,3,6,12 and 24hours.And for CBF and MTT,the difference between ischemia side and normal contra-lateral side at 30minutes,3, 6 and 12 hours had statistic meaning,but at 24 hours,the difference had no statistic meaning.The cerebral infarct had evidently low perfusion in 30 minutes which showed remarkably descended CBF and CBV,evidently prolonged MTT.Three rabbits had lightened low perfusion in 3 hours,6 hours and 12 hours respectively and in the next time, the perfusion was descended rapidly again and further aggravated along with time. Another seven rabbits had evidently low perfusion and the degree of low perfusion was aggravated gradually or fluctuated slightly at a certain level.In different time points,the difference of CBF and CBV had no statistic meaning(F=2.01,P=0.097;F=2.399, P=0.054),but for MTT,there was remarkable difference(F=5.39,P=0.001).For CBF, CBV and MTT,there were statistic difference between infarct side and normal contra-lateral side at 30minutes,3,6,12 and 24hours.The mean values of CBF,CBV and MTT in cerebral ischemia and cerebral infarct were(58.08±20.81)ml/(min·100g), (3.29±0.62)ml/100g,(6.03±2.50)s and(16.47±14.91)ml/(min·100g),(1.87±1.04)ml/100g, (13.69±5.15)s respectively,and the difference has statistic meaning.The accuracy of CBF in diagnosing cerebral infarct is remarkably higher than CBV and the accuracy of MTT is slightly higher than CBF.If the value of CBF is less than 27.07ml/(min·100g),the sensitivity is 82.8%and the specificity is 82.3%in diagnosing cerebral infarct.If the value of CBV is less than 2.48 ml/100g,the sensitivity is 73.8%and the specificity is 74.2%.And if the value of MTT is more than 7.77s,the sensitivity is 89.3%and the specificity is 85.5%.Conclusions:The cerebral ischemia has most evidently low perfusion during 3~6 hours and the low perfusion was lightened during 12~24 hours, while the cerebral infarct showed that the low perfusion was gradually aggravated along with time or descended rapidly again after lightened with different degree.The cerebral ischemia and cerebral infarct have different features of CT perfusion imaging.The cerebral ischemia was characterized by mismatch in CBF and CBV values which showed a significantly reduced CBF but slightly elevated or reduced or normal CBV,whereas the cerebral infarct showed a matched significant decrease in both CBF and CBV values.The values of CBF<27.07ml/(min.100g) and MTT>7.77s could identify cerebral ischemia and cerebral infarct,and it has very high accuracy in diagnosing cerebral infarct.

  • 【网络出版投稿人】 复旦大学
  • 【网络出版年期】2009年 06期
节点文献中: 

本文链接的文献网络图示:

本文的引文网络