MR弥散张量纤维束示踪及锰增强MRI评价兔坐骨神经挤压伤的实验研究
|
摘要
研究背景
周围神经损伤是严重创伤时常见的合并症,常见的损伤类型有挤压伤、牵拉伤等。损伤的严重程度是临床决定采取何种治疗方法的重要因素,轻度损伤通常采取保守治疗患者即可自行恢复神经功能,而重度损伤通常需要手术修复才能保证神经功能的恢复。通常需要观察患者在数周至数月时间内有无神经再生来判断神经损伤的严重程度,很多患者因此失去最佳的治疗时机,从而导致不可恢复的残疾。因此早期准确判断神经损伤程度的辅助检查方法对于帮助临床医生采取正确治疗神经损伤是很必要的。
目前临床上除病史、体征外主要通过神经电生理检查评估外周神经损伤程度及范围,一直被认为是评价外周神经功能状况的金标准。然而电生理检查受操作者的经验等因素影响较大,判断神经功能缺乏敏感性和特异性,而且无法提供对于神经损伤的术前评估和手术定位都十分重要的准确解剖学信息。
虽然一些研究表明常规的MR成像能够显示周围神经损伤时信号和弛豫时间的异常,但是对于神经的再生缺乏特异性,而且无法显示复杂解剖部位的神经走行。磁共振弥散张量成像(diffusion tensor imaging, DTI)在显示中枢神经纤维束及据此诊断中枢神经系统疾病的作用已被广泛认同,部分研究也显示了DTI在周围神经系统损伤中的作用,然而常规场强下的在体DTI评价周围神经挤压伤作用尚缺乏系统研究。锰增强磁共振成像(manganese enhanced magnetic resonance imaging, MEMRI)被证实能够准确显示脑内神经传导通路,然而MEMRI评价周围神经损伤的应用尚很少见。
本课题采用1.5TMR对兔坐骨神经挤压伤模型进行弥散张量成像和锰增强MR成像,目的在于：①探讨1.5TMR兔坐骨神经弥散张量纤维束示踪的可行性及优化b值。②通过影像与病理及神经功能的对照,探讨DTI在评价外周神经损伤的作用。③影像与组织病理学及神经功能对照,初步探讨MEMRI在评价外周神经损伤及修复过程中的作用。以期为早期评价外周神经损伤及指导治疗方案的制定提供准确可靠的影像学方法。
第一部分兔坐骨神经弥散张量纤维束示踪参数优化及可行性研究
研究目的
通过不同b值之间成像质量的比较探讨b值对弥散张量(DTI)纤维束示踪成像质量的影响；明确1.5T MR兔坐骨神经弥散张量纤维束示踪成像的最优b值；比较不同b值之间各导出量判断b值对导出量的测量有无影响；通过与大体解剖及常规MR扫描相对照探讨1.5TMR兔坐骨神经弥散张量纤维束示踪成像的可行性。
1材料与方法
1.1实验对象：
健康新西兰大白兔10只,体重2.0-3.0Kg.由广州中医药大学提供。
1.2仪器、设备及药品
Philips Achieva 1.5T Nova Dual MR扫描仪
上海辰光医疗科技有限公司生产的实验动物(兔)专用相控阵列射频线圈
Philips EWS (Extended Workspace) v2.6图像后处理工作站,内装FiberTrak
纤维束示踪后处理软件包
3%戊巴比妥钠,速眠新2代
1.3 MR扫描
1.3.1动物准备
新西兰大白兔经耳缘静脉注射3%戊巴比妥钠(30mg/kg)及0.2-0.3ml速眠新2代肌注麻醉后行MR扫描。
1.3.2 MR扫描方法及成像参数
扫描序列包括常规T2WI、T1WI和DTI。
T2WI采用快速自旋回波(Turbo Spin-echo,TSE)序列,参数如下：TR,2500ms；TE,120ms；TSE因子,25；层厚,2mm；层间距,1mm；扫描视野(Field of View,FOV),120mm；扫描矩阵：344×344,平面内像素大小0.35x0.35mm；重建矩阵,512×512；平面内重建像素,0.23×0.23；激励次数(NSA),4次。
T1WI亦采用TSE序列：TR,572ms；TE,17ms；TSE因子：3；脂肪抑制采用频谱预饱和反转恢复序列(SPIR)；余扫描参数同T2WI。
DTI扫描采用单次激发自旋回波回波平面成像(SE-EPI)及SENSE并行采集技术,SENSE因子为2；采用6个不同b值：400、600、800、1000、1200和1400s/mm2,成像参数：TR,1550-3650ms:TE分别为55、59、63、66、68和71ms；6个b值扫描的方向、范围及其他参数均相同：层厚,1.6 mm；间隔,Omm；FOV,130mm；扫描矩阵80×80；平面内像素大小,1.6×1.6；激励次数(NSA),2；弥散敏感梯度方向：32个。
1.4数据处理
DTI数据传入Philips EWS v2.6工作站FiberTrak软件包行图像后处理。纤维束示踪采用多个感兴趣区(ROI)定义法：于股骨大转子层面采用freehand模式划定2个相距5mm的ROI,刚好包含坐骨神经外径,软件自动显示出穿过2个ROI的纤维束。纤维束FA下限阈值取0.4,最大偏转角为27度,最小纤维束长度10mm,然后将纤维束与T2WI定位图融合。
1.5数据分析：
1.5.1通过软件自动计算纤维束数量、平均长度、纤维束FA值、ADC值、3个本征向量值(λ1、λ2、λ3)。
1.5.2测量并计算SNR
软件生成平均弥散加权成像(DWI)图,在坐骨神经上划定1个椭圆形ROI(5个像素),测量坐骨神经信号强度(SIn)；然后在兔肢体外距离肢体边缘1cm处地空气区域划定ROI(20个像素),测量背景噪声信号强度的标准差(SDbg)；按照公式计算信噪比：SNR=sIn/SDbg。
1.6图像质量主观评价
由两名有十年以上MR诊断经验的医师对每只新西兰兔6个不同b值下的纤维束示踪图像质量进行评估。阅片采用盲法评估。评估标准包括：纤维束长度、粗细、均匀度及边缘清晰度；根据图像质量进行排序(1为最好,6为最差)。
将4只实验兔处死后解剖右下肢坐骨神经,与质量最优的纤维束重建图像对比,主观评价纤维束走行与大体解剖的一致性。对侧下肢冰冻后在股骨转子水平垂直股骨长轴截断,重建纤维在T2WI、T1WI图投影图与横断面解剖对照,判断两者坐骨神经位置是否一致。
1.7统计学分析
采用SPSSl3.0统计软件。所有定量数据均采用均数±标准差形式表示。先对不同b值间纤维束数量、纤维束平均长度、FA值数据进行数据分析,符合正态分布时均数比较采用单因素方差分析(One-Way ANOVA),进行方差齐性检验,若方差齐采用Fisher法,若方差不齐采用近似F检验Welch法；若方差分析显著则进行组间多重比较,方差齐时多重比较采用Bonferroni法,方差不齐者采用Dunnett's T3法；若方差分析不显著则不进行多重比较。不同b值间主观评价分数的比较采用多个相关样本非参数检验,两评价者之间一致性检验采用kappa检验。SNR.FA值、ADC值及λ1、λ2、λ3值与b值的相关性检验采用Spearman相关检验。均以P<0.05为有统计学意义。
2结果
2.1 DWI图像信噪比(SNR)与b值的关系
图像信噪比(SNR)随b值增高SNR逐渐下降。SNR与b值呈负相关(P=0.000,r=-0.589)
2.2各组不同b值纤维束数量、平均长度比较
纤维束数量：b=1000 s/mm2时纤维束数量明显较其他组多,b=400 s/mm2和1400 s/mm2组数量最少。方差分析有显著性(F=21.641,P=0.000)；组间多重比较,b=1000 s/mm2组与b=400 s/mm2、b=600 s/mm2、b=1200s/mm2、b=1400s/mm2组间纤维束数量差异均有显著性差异(P值分别为0.000、0.002、0.000和0.000),与b=800 s/mm2纤维束数量无显著性差异。
纤维束平均长度：b=1000 s/mm2时纤维束平均长度较其他b值长,b=400 s/mm2时纤维束平均长度最短,方差分析有统计学意义(F=19.736,P=0.000)。组间多重比较采用Dunnett's T3法,b=1000 s/mm2组与b=400 s/mm2、b=600 s/mm2、b=1200 s/mm2、b=1400 s/mm2组之间差异有统计学意义(P值分别为0.000、0.008、0.013和0.002),与b=800 s/mm2组间差异无统计学意义。
2.3 FA值和ADC值及3个本征向量值(λl、λ2、λ3)与b值的关系
各组间FA值均数介于0.51+0.02和0.53±0.03之间,各组FA值总体组间差别无统计学意义,且与b值无相关性。纤维束的ADC值、3个本征向量值(λ1、λ2、λ3)均随b值增加而逐渐降低,Spearman相关分析显示ADC值、3个本征向量值(λ1、λ2、λ3)均与b值成负相关(P=0.000,r分别为-0.787、-0.898、-0.829和-0.559)。
2.4图像质量主观评价：
b=1000 s/mm2组的图像质量等级评分均在前3名之内,平均秩次最低(图像质量最高),其次为b=800 s/mm2和1200 s/mm2组,b=400组平均秩次最高(图像质量最差),组间差异有统计学意义(χ=33.714,P=0.000)；两评价者之间具有高度一致性(κ=0.76,P=0.000)。
选择b=1000 s/mm2组的纤维束示踪图像与大体解剖和断面解剖对比,结果显示重建纤维束的走行方向和部位与大体解剖一致,基本能够显示坐骨神经全程,坐骨神经的两个分支构成-胫神经和腓神经亦能较清楚显示。
结论
1.不同的b值对DTI外周神经纤维束示踪的信噪比以及纤维束长度和数量产生显著的影响,从而影响重建图像的质量。
2.不同b值对坐骨神经ADC值及本征向量值的测量会产生影响,而FA值的测量不受影响。
3.在1.5TMR系统下进行兔坐骨神经DTI纤维束示踪成像时,取b值为1000s/mm2时可以获得最好的图像质量。
4.采用本研究的扫描及后处理方法在1.5TMR系统下进行兔坐骨神经弥散张量纤维束示踪成像是可行的。
第二部分弥散张量纤维束示踪评价兔坐骨神经挤压伤的实验研究
研究目的
明确坐骨神经损伤远段及近段FA值、ADC值、λ//和λ⊥随时间变化情况。与组织病理学改变及神经功能评分相对照,判断坐骨神经挤压伤远段FA值、ADC值、λ//和λ⊥与组织学改变和神经功能相关性。
材料与方法
1.1研究对象
新西兰大白兔36只,体重2.0-3.0Kg,平均2.4kg.由广州中医药大学提供。36只兔随机分组：4只作为正常组,其余32只为损伤组。损伤组32只兔按照神经损伤后恢复时间随机分为24小时、4天、8天、2周、4周、6周、8周、10周共8组,每组4只。
1.2仪器、设备及药品
荷兰Philips Achieva 1.5T Nova Dual双梯度MR扫描仪。
上海辰光医疗科技有限公司生产的实验动物(兔)专用相控阵列射频线圈,
型号CG RBC 18-H150-AP。
Philips EWS (Extended Workspace) v2.6图像后处理工作站,内装FiberTrak
纤维束示踪后处理软件包。
麻醉药品：3%戊巴比妥钠,速眠新2代。
外科无菌手术包
16cm无菌持针钳1把
1.3坐骨神经急性挤压伤动物模型制作
实验组兔经耳缘静脉注射3%戊巴比妥钠(30mg/kg)及0.2-0.3ml速眠新2代肌注麻醉。在无菌操作下暴露右侧坐骨神经,于股骨转子下方1cm处以大号持针钳垂直夹持坐骨神经,锁至第一扣,夹持5分钟后解除,可见神经纤细变薄但未离断。在夹伤处神经外膜用尼龙线缝一针作为标记。然后以0.2mm丝线分层缝合切口。对侧相同部位行假手术。
1.4 MR扫描
分别于挤压伤模型完成后24小时、4天、8天、2周、4周、6周、8周、10周对每组动物行MR扫描。
1.4.1动物准备：
新西兰大白兔经耳缘静脉注射3%戊巴比妥钠(30mg/kg)及0.2-0.3ml速眠新2代前肢肌注复合麻醉后行MR扫描。
1.4.2 MR扫描方法及成像参数
T1WI和T2WI扫描参数同第1部分。
DTI扫描参数：b=1000 s/mm2;TR/TE:8184/66 ms;层数：35；扫描时间：9分46秒；余参数同第一部分DTI扫描
1.5纤维束重建
使用Philips EWS v2.6工作站的FiberTrak软件包自动生成彩色编码FA图,进行纤维束示踪重建后将T1WI和T2WI解剖定位图与FA图融合。纤维束示踪采用多个感兴趣区(ROI)定义法：结合坐骨神经横断面T2WI定位图及彩色编码FA图,于股骨大转子层面采用freehand模式划定2个相距10mm的ROI,刚好包含坐骨神经外径,软件自动显示出穿过2个ROI的纤维束。纤维束FA下限阈值取0.35,最大偏转角为27度,最小纤维束长度30mm,然后将纤维束与T2WI定位图融合。
1.6数据测量
在彩色编码FA图上分别在损伤侧(右侧)坐骨神经损伤处远端1cm处、损伤近端1cm处以及假手术侧相应位置划定感兴趣区(ROI)。软件自动计算出相应ROI的FA值、ADC值及3个本征向量值(λl、λ2λ3)。计算λ⊥=(λ2+λ3)/2；λ∥=λ1。
1.7组织学检查
在DTI扫描完成之后,分别于24小时、4天、2周、4周、6周、8周和10周处死动物,暴露右侧坐骨神经并于钳夹处远端1cm处截取1mm神经组织置于4%预冷戊二醛溶液中预固定,放置4℃冰箱保存,1%四氧化锇后固定,系列乙醇脱水,Epon812环氧树脂包埋,超薄切片,经醋酸铀和柠檬酸铅双重染色后,用Hitachi H7500电子显微镜观察；远端再截取5mm神经组织10%甲醛溶液固定,石蜡包埋后行横断面及纵切面切片,HE染色后光镜下观察。
1.8功能评价：
每次扫描时对实验模型兔损伤侧肢体坐骨神经功能进行定量评价,趾张反射每个时间段每只兔对应1-4级神经功能分别记1-4分；改良Tarlov评分根据对应的0-4级神经功能分别记0-4分。每只兔满分8分,每组满分32分。
1.8.1趾张反射
手捏住兔颈部皮肤将兔提离地面,迅速使其在空中下降而不让其着地,观察患侧脚趾张开情况,1级：脚趾张开完全不可见；2级：仅可见轻度脚趾分开；3级：明显可见脚趾分开(但幅度仍低于正常水平)；4级：脚趾完全张开达到正常水平。
1.8.2 Tarlov评分
0级：肢体完全瘫痪,针刺无反应；1级：针刺刚刚可见有轻微反应；2级：损伤侧后肢所有关节有自主运动,手压使足部屈曲时无抵抗；3级：手压使足部屈曲时有明显抵抗,但走路步态不稳；4级：行走步态正常。根据0-4级分别记0-4分.
1.9统计学分析
采用SPSS13.0统计软件。所有定量数据均采用均数±标准差形式表示。损伤后8个不同时间点及正常组之间坐骨神经FA值、ADC值及λ//和λ⊥均数的比较采用单因素方差分析(One-Way ANOVA),先进行方差齐性检验,若方差齐采用Fisher法,若方差不齐采用近似F检验Welch法；方差齐时多重比较采用LSD法,方差不齐者采用Dunnett's T3法；若方差分析不显著则不进行多重比较。均以P<0.05为有统计学意义。
2结果
2.1挤压伤后坐骨神经各段FA值、ADC值及本征向量值的变化
2.1.1损伤远段FA值、ADC值及本征向量值的变化
单因素方差分析显示,损伤远段各时间段FA值有显著性差异(F=47.283, P=0.000),LSD法组间多重比较显示：正常组FA与其余各组间均有显著性差异(P<0.01)；其余两相邻时间段之间FA值有显著性差异的有：24小时组与4天组(P=0.000)、4周组与6周组(P=0.000)。
损伤后各时间段损伤远段神经ADC值之间无显著性差异。
损伤后各时间段损伤远段神经λ//方差分析无显著性。损伤后各时间段损伤远段神经λ⊥方差分析有显著性(F=11.847,P=0.000).组间多重比较显示：正常组与4天、8天、2周、4周、6周组之间λ⊥有显著性差异(P<0.05),与24小时、8周、10周组之间无显著性差异；其余相邻两时间段之间λ⊥有显著性差异者为：24小时组与4天组(P=0.000)、4周组与6周组(P=0.007)。
2.1.2挤压伤后损伤侧近段FA值、ADC值及本征向量值的变化
单因素方差分析结果显示,损伤后各时间段FA值、ADC值λ//及λ⊥均无显著性差异。
2.1.3假手术侧FA值、ADC值及本征向量值的变化
单因素方差分析结果显示,损伤后各时间段FA值、ADC值λ//及λ⊥均无显著性差异。
2.2挤压伤后坐骨神经损伤远段组织病理学改变
正常坐骨神经神经束膜中可见神经纤维排列整齐,髓鞘及轴索结构清晰可见。损伤后24小时可见髓鞘部分髓鞘肿胀为主。损伤后4天轴索开始崩解,髓鞘内可见轴索碎片。雪旺细胞增生不明显。损伤后2-4周大量轴索消失,髓鞘广泛崩解,雪旺细胞开始广泛增生,广泛炎症细胞浸润。至损伤6周时可见少量新生轴索,增生的雪旺细胞排列成索状。损伤8周时可见大量新生轴索,明显可见髓鞘形成。损伤10周时大部分神经髓鞘结构形成,其中可见清晰的轴索结构。
2.3坐骨神经损伤后功能恢复情况
损伤后24小时实验兔趾张反射完全消失,但伤侧下肢对针刺多有反应,但患肢无张力(Tarlov 2级)；2周时个别动物患肢稍有张力(Tarlov 3级),趾张反射仍检测不到；4周时个别动物出现患侧脚趾轻微分开,多数动物下肢有张力；6周时多数动物出现脚趾轻微分开,患肢有力但行走步态达不到正常；8周时趾张反射恢复明显,个别达到正常；10周时趾张反射基本正常,几乎全部动物行走步态正常。
假手术侧肢体在术后各时间段有力并活动自如,被毛光整,趾张反射正常。
结论
1.坐骨神经挤压伤远段FA值和λ⊥在损伤后发生了显著变化,其变化程度与组织学及神经功能改变有较好的相关性,能够反映神经Wallerian变性和再生情况。
2.损伤远段ADC值和λ//损伤后各时间段均未出现显著变化。
3.坐骨神经损伤近段在损伤后各时间段FA、ADC、λ//λ(?)均未出现显著变化。
第三部分锰增强MRI评价兔坐骨神经挤压伤的初步研究
研究目的
评价MEMRI坐骨神经挤压伤近端强化程度与组织学和神经功能的相关性,探讨ME-MEI评价坐骨神经挤压伤的价值；探讨周围神经MEMRI锰离子的运输途径。
1材料与方法
1.1研究对象
新西兰大白兔24只,雌雄不限,体重2.0～3.OKg,平均2.4kg.由广州中医药大学动物实验部提供,24只兔随机分组：4只作为正常对照组,其余20只兔作为损伤组,损伤组20只兔按照神经损伤后恢复时间分为24小时组、2周、4周、6周、10周共5组,每组4只。
1.2仪器、设备及药品
磁共振扫描仪、线圈、图像工作站、麻醉药品、外科切开无菌手术包及16cm
持针钳同第二部分；
50ul微量注射器
AR级MnCl2·4H2O晶体(分子量,197.9g/mol;纯度>99%)；
聚乙二醇1540
1.3坐骨神经急性挤压伤动物模型制作
同第二部分
1.4MnCl2溶液配制：
聚乙二醇1540固体配置成50%聚乙二醇1540溶液备用。MnCl2·4H2O晶体加入50%聚乙二醇1540溶液配制成400mM MnCl2溶液。
1.5MnCl2注射
24小时组、2周组、4周组、6周组和10周组分别于损伤后相应的时间行MnCl2溶液神经鞘内注射。
新西兰大白兔麻醉后无菌操作下暴露坐骨神经,然后用50ul微量注射器分别向坐骨神经的两个主要分支——胫神经和腓神经内分别注射30ul和20ul 400mM的MnCl2溶液,注射点距离损伤部位约2cm。对侧下肢行假手术。
1.6 MR扫描
MnCl2注射后24h行T1WI扫描。
T1WI和T2WI扫描方法及参数同第二部分。
1.7数据测量：
在坐骨神经损伤近端选取3个不同的位置：股骨大转子水平、股骨大转子近端1cm处和坐骨结节水平,以下分别称为近段、中段和远段。分别测量3个感兴趣区的坐骨神经信号强度S和对侧坐骨神经信号强度Sc,再测量股骨近端内侧肌肉信号强度Sm作为内参照,计算相对信号强度AS=(S-Sc)/Sm。
1.8组织学检查：
同第二部分相应章节
1.9功能评价：
同第二部分相应章节
1.10统计学分析
采用SPSS 13.0统计软件。所有定量资料均采用均数±标准差形式表示。坐骨神经在损伤后不同时间段相对信号强度△S的比较及各时间段近段与中段、中段与远段相对信号强度的比较分别采用单因素方差分析和双因素方差分析,先进行正态性和方差齐性检验,若方差齐采用Fisher法方差分析,若方差不齐采用近似F检验Welch法,方差分析有显著性者进行多重比较,若方差齐则采用LSD法,方差不齐则采用Dunnett's T3法；若方差分析不显著,则不进行多重比较。规定P<0.05为有统计学意义。
2结果
2.1正常组及损伤后各时间段的坐骨神经MEMRI表现
正常组坐骨神经在远端注射400mM MnCl2溶液后可在MRTIWI图像上观察到神经干的明显强化,强化范围均达到垂直段以上,强化程度由远端到近段逐渐减低。挤压伤24小时后的动物坐骨神经损伤点近端绝大多数未见明确强化；损伤后2周和4周组损伤近端可见由远及近的轻度强化,,但范围较短。损伤后6周时和10周时坐骨神经近段可见明显强化,但强化程度不如正常组,强化程度亦由远端到近段逐渐减低。
2.2坐骨神经损伤近侧近段、中段和远段在损伤后不同时间段锰增强后的相对信号强度(△S)比较
单因素方差分析显示：
损伤近侧近段、中段及远段在不同损伤时间段的相对信号强度(△S)均有显著性差异(F=5.819, P=0.002; F=26.987, P=0.000; F=45.406, P=0.000)。
近段损伤后各时间段相对信号强度(△S)多重比较显示,正常组与24小时、2周、4周和6周组间差异有显著性(P<0.05),正常组与10周组无显著差异；其余各相邻时间段之间均无显著性差异。
中段损伤后各时间段相对信号强度(△S)多重比较显示,正常组与其余各时间段之间均有显著性差异(P≤0.001)；损伤后24小时、2周、4周组之间无显著性差异,6周组和10周组与24小时、2周、4周组之间均有显著性差异(P≤0.011,P<0.01),6周与10周组之间无显著性差异。
远段损伤后各时间段相对信号强度(△S)多重比较显示,正常组与其余各时间段之间均有显著性差异(P=0.000)；损伤后24小时、2周、4周组之间无显著性差异,6周组和10周组与24小时、2周、4周组之间均有显著性差异(P≤0.001,P=0.000),6周与10周组之间无显著性差异。
2.3双因素方差分析显示不同时间段损伤近侧之近段、中段及远段相对信号强度(△S)之间有显著性差异(F=90.928, P=0.000); LSD法多重比较显示：近段与中段(P=0.000)、近段与远段(P=0.000)以及中段与远段(P=0.001)之间相对信号强度(△S)均有显著性差异。
2.4挤压伤后坐骨神经损伤远段组织病理学改变
见第二部分相应章节
2.5坐骨神经损伤后功能恢复情况
第二部分相应章节
2.6 MEMRI坐骨神经损伤近侧相对信号强度(△S)与神经功能评分的相关性
Spearman相关检验显示,坐骨神经损伤近侧的近段、中段和远段相对信号强度(△S)随损伤后恢复时间的变化与坐骨神经功能评分均呈正相关,相关系数分别为r=0.670(P=0.000),r=0.888(P=0.000)和r=0.899(P=0.000).
结论
1.坐骨神经挤压伤MEMRI显示损伤后各时间段损伤近段强化程度与组织学和神经功能有较好的相关性,MEMRI能够反映坐骨神经损伤后再生情况。
2.坐骨神经MEMRI的强化程度有远及近逐渐减弱,一定程度说明Mn2+的逆向传递有被动弥散的因素。
Background
Peripheral nerve injury(PNI) is a common complication of severe trauma, the most common types including crush injury and strech injury,et al. Severity of the injury is an important factor for clinicians to determine what kind of therapy to apply. Conservative approach is sufficient for mild nerve injury, but in severe injury, surgical repair is often needed to ensure the fuctional recovery of the nerve. To judge the severity of PNI, observation of nerve regeneration for several weeks to months is needed, and many patients lost the best opportunity of therapy and result in irreversible disability. Therefore, an auxiliary method that can accurately detect the severity and regeration of the injuried nerve is necessary to help clinicians to apply proper therapy for PNI.
Electrophysiological method is mainly used to detect the degree of PNI in addition to physical examination in clinical practice nowadays, which is known as gold standard to evaluate the function of peripheral nerve. However, Electrophysiological method is largely dependent on the experience of the operator, lack of sensitivity and specificity and can not obtain the anatomical information which is very important for preoperative evaluation and intraoperative localization of PNI.
Although some research showed that conventional MRI is useful to reveal abnormality of signal and time of relaxation of PNI, but it lacks of specificity in detection of nerve regeneration and can not display the running course of nerves with complex anatomy. It is generally accepted that diffusion tensor MR imaging (DTI) can display nerve fibers and diagnose diseases in central nervers system, some researches also showed it is helpful in PNI. Nevertheless, systemic research of in vivo DTI evaluating PNI using conventional MR scanner has never been reported. It is reported that manganese enhanced magnetic resonance imaging(MEMRI) can accurately reveal pathways in the brain, but research of MEMRI evaluating PNI is very few.
The purpose of this study lies in:①To explore the feasibility of in vivo diffusion tensor fiber tractography of rabbit sciatic nerve at 1.5 T MR. (2)To establish animal model of peripheral nerve crush injury and investigate the value of DTI in evaluation of peripheral nerve crush injury and regeneration by comparing with histopathology and nerve function assessment.③To investigate the value of MEMRI in evaluation of peripheral nerve crush injury and regeneration by comparing with histopathology and nerve function assessment. This study aims at providing a reliable imaging method to accurately evaluate PNI and direct therapy.
Part One:Study on Optimal Parameter and Feasibility of Diffusion Tensor Tractography of Rabbit Sciatic Nerve
Objective
To determine the effect of b value on the image quality of DTI tractography by comparing the image quality under different b values and optimal b value of diffusion tensor tractography of rabbit sciatic nerve at 1.5 T MR. To determine the effect of b value on measurement of the derived DTI paramertes; To investigate the feasibility of diffusion tensor tractography of rabbit sciatic nerve at 1.5 T MR by comparing reconstructed sciatic nerve fiber with gross anatomy and conventional MR images of sciatic nerve.
1 Materials and methods
1.1 Subjects
10 healthy New Zealand white rabbits weighting 2.0-3.OKg provided by Guangzhou traditional Chinese Medicine University
1.2 Instruments, equipment and drugs
Philips Achieva 1.5T Nova Dual MR scanner
Dedicated rabbit coil manufactured by Chenguang medical technology company
Ltd. Of Shanghai.
Philips EWS (Extended Workspace) v2.6 post processing workstation with
FiberTrak software package for fiber tracking.
3% sodium pentobarbital and Su-Mian-XinⅡ
1.3 MR scan
1.3.1 Animal preparation:
New Zealand rabbits underwent MR scan after anesthetized by 3% sodium pentobarbital (30mg/kg) intravenously and 0.2-0.3ml Su-Mian-XinⅡintramuscularly
1.3.2 MR scan method and parameters
MR sequence including conventional T2WI,T1WI and DTI.
TSE sequence were used for T2WI scan, following parameters were used:TR, 2500ms; TE,120ms; TSE factor,25; slice thickness,2mm; gap, 1mm; FOV, 120mm; matrix:344×344, in plane pixel size,0.35×0.35mm; NSA,4.
TSE sequence were used for T1WI scan:TR,572ms; TE,17ms; TSE factor,3; SPIR were used for fat suppression, other parameters were the same as T2WI.
Single shot SE-EPI sequence and SENSE parallel scanning technique were used for DTI scan. Six different b values were used:400、600、800、1000、1200和1400 s/mm2, scan parameters were as follow:TR,1550-3650ms; TE,55,59,63,66,68 and 71ms respectively; Direction, range and other parameters were same under 6 b values:slice thickness,1.6 mm; gap,0mm; FOV,130mm; matrix,80×80; in plane pixel size,1.6×1.6; NSA,2; MPG:32.
1.4 Data processing
Philips EWS (Extended Workspace) v2.6 post processing workstation with FiberTrak software package were used. Multiple ROI method was used:2 ROIs of 5mm apart were drawn in freehand mode at the level of greater trochanter of femur, including the outer diameter of sciatic nerve. Fibers running through the 2 ROIs were computed automatically by the software. The lower limit of FA value was defined as 0.4, maximum angle change was 27 degree, minimum fiber length was 10mm, then the fibers was fused with the T1 WI or T2WI images.
1.5 Data analysis:
1.5.1 Number, mean length, FA, ADC and three Eigen value (λ1、λ2、λ3) of the fibers were computed by the software automatically
1.5.2 Measurement and computation of signal to noise ratio (SNR) DWI images were produce by the software, a ROI (5 pixels) was drawn on the sciatic nerve to measure the signal intensity of the nerve (SIn); ROI of 20 pixels was drawn on the air region that lcm outside the body margin of rabbit limb to measure the standard deviation of background noise (SDbg); SNR was computed according to the following formula:SNR=SIn/SDbg。
1.6 subjective assessment of image quality
Assessment of the quality of fiber tractography under 6 different b values was conducted by 2 radioglogists with more than10 years experience of MR diagnosis. Blind method was adopted. The principle to consider including:fiber lenth, diameter, homogeneity and smoothnesss of fiber margin. Sorting of the image quality was made afterwards(l is defined as the best quality and 6 the worst)
Four rabbits were sacrificed; the sciatic nerve was dissect and compared with the reconstructed fibers of the best quality to evaluate the consistency between the tractography and the gross anatomy. Contralateral legs were frozen and transversely desected at the level of great trochanter of femur. Reconstructed fibers fused with the T2WI and T1WI images were compared with sectional anaotomy to test whether the position of fibers was in accordance.
1.7 Statistical analysis
SPSS 13.0 software was used. All data was presented as the mean±standard deviation, Test of normality was done in advance, One-way Anova was used to compare means of number, mean length, FA, ADC and three Eigen value (λ1、λ2、λ3) of the fibers under different b values if the data is of normal distribution. Fisher testwas used if variance was homogeneous, otherwise Welch test was used; If anova was significant, multiple comparison was conducted, Bonferroni test was used when the variance was homogeneous, Dunnett's T3 test was used otherwise. Multiple related sample nonparametic test was used in the comparison of scores of subjective quality assessment, and kappa consistency test was used to check interobserver agreement on tractographic image. Correlation between SNR, FA, ADC,λ1,λ2,λ3 and b value was conducted using the Spearman correlation test. P<0.05 was defined as statistical significant.
2 Result
2.1 Relation between SNR of DWI image and b value
SNR decreased with the rising of b value, SNR was negtively correlated with the b value (P=0.000,r=-0.589)
2.2 Number and mean length of fibers under different b value
Number of fibers of b=1000 s/mm2 was more than that of other b values, anova was significant (F=21.641, P=0.000); Multiple comparison showed significant difference between number of fibers of b= 1000 s/mm2 group and that of b=400,600, 1200,1400 s/mm2 group (P<0.001), but no significant difference between b=800 s/mm2 group.
Mean length of fibers of b=1000 s/mm2 was longer than that of other b values, anova is significant (F=19.736, P=0.000). Multiple comparison showed significant difference between mean length of b=1000 s/mm2 group and that of b=400,600,1200,1400 s/mm2 group (P=0.000,0.008,0.013,0.002, respectively), but no significant difference between b=800 s/mm2 group.
2.3 Relationship between FA, ADC,λ1,λ2,λ3 and b value
Means of FA in all of the groups was between 0.51±0.02 and 0.53±0.03, there was no significant difference between FAs of different groups, and there was no correlation between FA and b value.ADC,λ1,λ2.λ3 decreased with the rising of b value. Negative correlation between ADC,λ1,λ2,λ3 and b value were detected by Spearman correlation test (P=0.000, r=-0.787,-0.898,-0.829,-0.559, respectively).
2.4 Subjective assessment of image quality
Image quality ranking of b=1000 s/mm2 was always in the first 3 position with the lowest score(best quality), and score of b=400 s/mm2 group was the highest(worst quality)(X2=33.714,P=0.000). Interobserver agreement on tractographic image quality ranking was good (k=0.76, P=0.000)
Reconstructed fiber of b=1000 s/mm group was compared with the gross anatomy and transactional anatomy, comparison showed that the running course and location of the tractographic fiber is consistant with that of sciatic nerve anatomy, the two main branch within the sciatic nerve-tibial nerve and peroneal nerve was clearly revealed.
Conclusion
1. Different b value had significant effect on the SNR, number and length of fibers of peripheral nerve, thereby effect the image quality of DTI tractography..
2. Different b values had significant effect on ADC and eigenvalues measurement, but had no effect on FA measurement.
3. The optimal b value for DTI and fiber tractography of rabbit sciatic nerve at 1.5 T was 1000 s/mm2.
4. In vivo diffusion tensor fiber tractography of rabbit sciatic nerve at 1.5 T MR was feasible with parameters we used.
Part Two:Experimental Study on MR Diffusion Tensor Tractography in Evaluating Sciatic Nerve Crush Injury of Rabbit
Objective
To investigate the FA, ADC,λ//andλ(?) of distal and proximal to the crush injury changes with the different time course after injury. Comparing with histopathology and functional assessment, to determine the relationship between FA, ADC,λ//,λ(?) of distal part crushed sciatic nerve and histopathology and nerve function.
1 Materials and methods
1.1 Subjects
36 New Zealand white rabbits weighting 2.0-3.0Kg provided by Guangzhou traditional Chinese Medicine University were used.36 rabbits were randomly divided into normal control group(4 rabbits) and crush injury group(32 rabbits), the crush injury group(32 rabbits) were randomly divided into 8 groups(24hours,4 days, 8 days,2 weeks,4 weeks,6 weeks,8weeks and 10weeks) according to the time course after crush injury.
1.2 Instruments, equipment and drugs
MR scanner, coil, and anesthetic agent were the same as part I.
Sterile operating instrument set
A 16cm needle holder
1.3 Establishment of rabbit sciatic nerve crush injury model.
Rabbit was anesthetized with 3% sodium pentobarbital (30rng/kg) intravenously and 0.2-0.3ml Su-Mian-Xin II intramuscularly, sciatic nerve of right leg was exposed under sterile operation, sciatic nerve was crushed using a 16cm needle forceps at the level 1cm distal to the greater trochanter of femur, the forceps was locked to the first point of its ratchets and retained for 5 minutes before releasing. An epineurial suture was placed right above the crush site to exactly localize the position of the crush.Sham operation was performed on the contralateral leg.
1.4 MR scan
MR scan were performed 24 hours,4 days,8 days,2 weeks,4 weeks,6 weeks, 8weeks and 10weeks after injury,as well as the normal control group.
Animal preparation were the same as part I.
MR scan sequence and parameters were the same as part I except for some parameters of DTI scan:b=1000 s/mm2; TR/TE:8184/66 ms; slices:35; scan time:9 minute and 46 seconds.
1.5 Post processing of Data.
Philips EWS (Extended Workspace) v2.6 post processing workstation with FiberTrak software package were used. Multiple ROI method was used:2 ROIs of 10mm apart were drawn in freehand mode at the level of greater trochanter of femur(proximal to the injury site), including the outer diameter of sciatic nerve. Fibers running through the 2 ROIs were computed automatically by the software. The lower limit of FA value was defined as 0.35, maximum angle change was 27 degree, minimum fiber length was 30mm, then the fibers was fused with the T1WI or T2WI images.
1.6 Data measurement
Three ROIs were drawn in freehand mode on the color coded FA map generated by the software, two of the ROIs was drawn on the sciatic nerve of the injuried side,one of which is lcm distal and the other lcm proximal to the injury site; the third ROI was placed on the sham operated sciatic nerve on the site corresponding to the distal ROI of the injuried side.FA, ADC,and the three eigenvalues(λ1、λ2、λ3) were computed automatically by the software,λ(?) andλ//were computed as following formula:λ(?)=(λ2+λ3)/2,λ//=λ1
1.7 Histopathological examination
Animals were sacrificed after MR scan, and sciatic nerves of the injuried side that 1cm distal to the injured site were harvested. One part of the nerve were fixed in 4% paraformaldehyde, dehydrated, osmificated, and embedded in plastic resin. Semithin sections were then made and stained with uranyl acetate and lead citrate for transmission electron microscopic examination (Hitachi H7500). Another part of the sample were fixed in 10% paraformaldehyde and sectioned longitudinally at a 2-mm thickness. The sections were stained with standard hematoxylineosin staining for light microscopic examination.
1.8 Functional assessment
Functional assessment was performed befor every MR scan.
Toe-spreading reflex:grade 1,barely visible spreading of toes; grade 2, readily discernible though slight spreading of the toes; grade 3, unequivocal spreading of toes (though less forceful than normal); and grade 4, full spreading of the toes equal to normal.
Modified Tarlov score:grade 0, complete paraplegia of hind limbs; grade 1, hind limbs barely respond to hind limb pinch (barely detectable movement); grade 2, spontaneous movement at all hind limb joints but no resistance when flexing the foot; grade 3, obvious resistance when flexing the foot with abnormal gait; and grade 4, walks with normal gait.
1-4 marks were recorded for one rabbit of corresponding toe spreading reflex grade;0-4 marks were recorded for one rabbit of corresponding Tarlov score grade. Full marks for a single rabbit is 8, and for a group is 32 marks.
1.9 Statistical analysis
SPSS13.0 software was used. All data was presented as the mean±standard deviation, One-way Anova was used to compare FA, ADC,λ//andλ(?) of sciatic nerve on the eight time point after injury as well as the normal control group. Fisher test was used if variance was homogeneous, otherwise Welch test is used; If anova was significant, multiple comparison was conducted, LSD test is used when the variance is homogeneous, Dunnett's T3 test was used otherwise. P<0.05 is defined as statistical significant.
2 Results
2.1 FA, ADC,λ//andλ(?) changes of sciatic nerve after crush injury
2.1.1 FA, ADC,λ,//andλ(?) changes of distal portion of injuried sciatic nerve.
One-way Anova showed that FA value of the distal portion injuried sciatic nerve at different time point is different significantly. (F=47.283, P=0.000), LSD multiple comparison test showed:FA of normal control group is significantly different from that of all the other group (P<0.01);Other significant differences of FA between consecutive groups were:24 hours group and 4 days group(P=0.000),4 weeks group and 6 weeks group(P=0.000).
There was no significant difference between ADC of every time point of distal portion injuried sciatic nerve.
There was no significant difference betweenλ//of every time point of distal portion injuried sciatic nerve.
There was significant difference betweenλ(?) of time points of distal portion injuried sciatic nerve (F=11.847, P=0.000). LSD multiple comparison test showed:λ(?) of normal control group is significantly different from that of 4 days,8 days,2 weeks,4 weeks,6 weeks group (P<0.05), but not significantly different from that of 24 hours,8weeks and 10 weeks group, Other significant differences ofλ(?) between consecutive groups were:24 hours group and 4 days group (P=0.000),4 weeks group and 6 weeks group(P=0.007)
2.1.2 FA, ADC,λ//andλ(?) changes of proximal portion of the injuried injuried sciatic nerve.
One-way Anova showed that There was no significant difference between FA、ADCλ//andλ(?) of every time point of proximal portion injuried sciatic nerve.
2.1.3 FA, ADC,λ//andλ(?) changes of sham operated sciatic nerve.
One-way Anova showed that There was no significant difference between FA、ADC,λ//andλ(?) of every time point of sham operated sciatic nerve.
2.2 Histopathological changes of distal portion of the injuried sciatic nerve.
Normal sciatic nerve showed that nerve fibers lines up in order, myelin sheath and axon was clearly seen.Swelling of the axon an myelin sheath was seen 24 hours after injury; axon began degeneration 4 days after injury, and fragment of axons can be seen.Proliferation of Schwann cells was not obvious; large amounts of axons disappeared and myelin sheath generally disintegrate 2-4 weeks after injury,and Schwann cells began to proliferate generally, inflammatory cell infiltration is obvious. Small amount of regenerated axons were detected until 6 weeks after injury, which wass very thin and amyelinated. Large amount of regenerated axons were revealed 8 weeks after injury.10 weeks after injury, most regenerated axons-were myelinated.
2.3 Functional recovery of sciatic nerve after injury
Toe-spreading reflex disappered 24 hours after injury, but the injuried leg respond to pinch but have no tension (Tarlov 2); Slight tension of the injured leg can be felt in individual animals 2 weeks after injury (Tarlov 3),but Toe-spreading reflex was still undetectable.Indivadule animals can be seen slight spreading of injuried side toes, strong force of the injuried leg can be detected in most animals; 6 weeks after injury, most animals revealed slight spreading and strong force of the injured leg, but the gait was still abnormal; 10 weeks after injury, some animals showed normal toe-spreading reflex, almost all the animals had normal gait.
Sham-operated legs was strong and moved freely after injury, and the fur of the legs was smooth and intacted.The toe spreading reflex was normal.
Conclusion:
1. FA andλ(?) of distal portion of injuried sciatic nerve changed significantly after injury, and the change were in good correlation with the histopathology and fuctional recovery. FA andλ(?) distal portion of injuried sciatic nerve can reflect Wallerian degeneration and regeneration after crush injury of sciatic nerve.
2. ADC andλ,//of distal portion of injuried sciatic nerve had no significant change after injury.
3. FA, ADC,λ,//andλ(?). of proximal portion of the injuried injuried sciatic nerve had no significant change after injury.
Part Three:Preliminary Study of Manganese Enhanced MRI in Evaluating Sciatic Nerve Crush Injury of Rabbit
Objective
To investigate the value of MEMRI in evaluation of nerve regenertation after sciatic nerve crush infury of rabbit by comparing with histopathology and functional assessment, crush injury; To explore the transporting pathway of Mn2+ in MEMRI of peripheral nerve.
1 Materials and methods
1.1 Subjects
24 New Zealand white rabbits weighting 2.0-3.OKg provided by Guangzhou Traditional Chinese Medicine University were used.24 rabbits were randomly divided into normal group(4 rabbits) and crush injury group(20 rabbits), the crush injury group were randomly divided into 5 groups(24hours,2 weeks,4 weeks,6 weeks and 10weeks) according to the recovery time after injury. There were 4 rabbits in each group.
1.2 Instruments, equipment and drugs
MR scanner, coil, and anesthetic agent were the same as part I. Sterile operating instrument set for crush injury modle was the same as part II. AR class MnCl2·4H2O crystal (molecular weight 197.9g/mol; purity>99%) polyethylene glycol(PEG) 1540
1.3 Establishment of rabbit sciatic nerve crush injury model
The procedure was the same as part II.
1.4 Preparation of MnCl2 solution
50% PEG 1540 solution was prepared by PEG 1540 solid melted and dissolved in 0.9% NaCl solution. MnCl2·4H2O crystal was dissolved in 50% PEG 1540 to form 400mM MnCl2 solution.
1.5 MnCl2 injection
After corresponding recovery time, rabbits were anesthetized again in the same manner as the initial operation. The sciatic nerve of the right leg was exposed, then 30 and 20ul of 400 mM MnCl2 were injected into the two major branches of the sciatic nerve, the tibial nerve and peroneal nerves, respectively. The injection site was 2cm distal to the injury site. Sham operation were performed on the contralateral legs. 1.6 MR scan
T1WI and T2WI sequence were scanned after 24 hours after MnCl2 injection.The technique and parameters used were the same as part II 1.7 Data measurement
ROIs were placed on three different portion of injuried sciatic nerve proximal to injury site:Greater trochanter of femur, 1cm proximal to Greater trochanter of femur and ischial tuberosity, which were defined as distal, middle and proximal segment respectively in this study. Signal intensity of the three ROIs on the sciatic nerve were recorded as S and the signal intensity of the contralateral sciatic nerve was recorded as Sc, the signal intensity of muscle medial to the proximal femur (Sm) was recorded as a reference. Relative signal intensity of injuried sciatic nerve was computed as AS, which was calculated according to the formula:ΔS=(S-Sc)/Sm.
1.8 Histopathologic examination
The same as partⅡ.
1.9 Functional assessment
The same as partⅡ.
1.10 Statistical analysis
SPSS 13.0 software was used. All data was presented as the mean±standard deviation, One-way Anova and Two-way Avova were used respectively to compare AS of different recovery time after injury as well as the normal control group andAS between three different portion of injuried sciatic nerve. Fisher test was used if variance was homogeneous, otherwise Welch test was used; If anova was significant, multiple comparison was conducted, LSD test was used when the variance is homogeneous, Dunnett's T3 test was used otherwise. If anova was not significant, multiple comparison would not be conducted. P<0.05was defined as statistical significant.
2 Results
2.1 MEMRI manifestation of normal group and injuried groups of sciatic nerve
Remarkable enhancement of the sciatic nerve on the T1WI images was revealed in the normal group after MnCl2 injection, the extent of enhancement was generally beyond the vertical portion of the sciatic nerve, and the enhancement decreased gradually from distal to proximal portion. There was no enhancement of the nerve in most animals of the 24 hours group after injury.2-4 weeks after injury, the sciatic nerve proximal to injury site revealed slight and short enhancement that decreased gradually from distl to proximal part.6 and 10 weeks after injury, the proximal part of the injuried sciatic nerve showed remarkable enhancement,but not as strong as normal group, the enhancement was also decreased gradually from distal to proximal portion.
2.2 Relative signal intensity (ΔS) changes of sciatic nerve after injury
One-way Anova showed that there were significant difference betweenAS of injuried sciatic at different time point after injury at proximal, middle and distal portion (F=5.819, P=0.002; F=26.987, P=0.000; F=45.406, P=0.000)
In the proximal portion, multiple comparison showed thatΔS of normal control group was significantly different from that of 24 hours,2 weeks,4 weeks and 6 weeks group (P<0.05), there was no significant difference betweenAS of normal group and 10 weeks group. There was no significant differences ofΔS between other consecutive groups.
In the middle portion, multiple comparison showed that AS of normal control group was significantly different from that of all the other groups (P<0.001),there was no significant difference betweenAS of 24 hours,2 weeks and 4 weeks groups.ΔS of 6 weeks and 10 weeks group was significantly different from that of 24 hours, 2 weeks and 4 weeks group(P<0.001,P=0.000). There was no significant differences ofΔS between 6 weeks and 10 weeks group.
In the distal portion, multiple comparison showed thatΔS of normal control group is significantly different from that of all the other groups (P=0.000), there was no significant difference betweenAS of 24 hours,2 weeks and 4 weeks groups. AS of 6 weeks and 10 weeks group was significantly different from that of 24 hours,2 weeks and 4 weeks group (P≤0.011,P=0.000). There was no significant differences ofΔS between 6 weeks and 10 weeks group.
2.3 Two-way Anova showed that there was significant difference between the three portion of injuried sciatic nerve at different recovery time point (F=90.928, P=0.000); LSD multiple comparision showed:There were significant differences betweenAS of proxilmal and distal portion (P=0.000), proxilmal and middle portion (P=0.000) as well as middle portion and distal portion (P=0.001)
2.4 Histopathological changes of distal portion of the injuried sciatic nerve. The same as partⅡ
2.5 Functional recovery of sciatic nerve after injury The same as partⅡ
2.6 Correlation between MEMRI relative signal intensity of sciatic nerve proximal to injury site and functional assessment.
Speraman correlation test showed that the proximal,middle and distal portion of scitic nerve promximal to injuried site was significantly correlated with functional scores(r=0.670,P=0.000;r=0.888,P=0.000,r=0.899(P=0.000).
Conclusion
1. The MEMRI signal intensity of sciatic nerve crush injury was in good correlation with histopathology and nerve function recovery, MEMRI was able to reflect the regeneration of crushed sciatic nerve.
2. Enhancement of sciatic nerve decreased gradually from distal to proximal portion, infered that there was passive diffusion factor in the retrograde Mn2+ transportation.
周围神经损伤是严重创伤时常见的合并症,常见的损伤类型有挤压伤、牵拉伤等。损伤的严重程度是临床决定采取何种治疗方法的重要因素,轻度损伤通常采取保守治疗患者即可自行恢复神经功能,而重度损伤通常需要手术修复才能保证神经功能的恢复。通常需要观察患者在数周至数月时间内有无神经再生来判断神经损伤的严重程度,很多患者因此失去最佳的治疗时机,从而导致不可恢复的残疾。因此早期准确判断神经损伤程度的辅助检查方法对于帮助临床医生采取正确治疗神经损伤是很必要的。
目前临床上除病史、体征外主要通过神经电生理检查评估外周神经损伤程度及范围,一直被认为是评价外周神经功能状况的金标准。然而电生理检查受操作者的经验等因素影响较大,判断神经功能缺乏敏感性和特异性,而且无法提供对于神经损伤的术前评估和手术定位都十分重要的准确解剖学信息。
虽然一些研究表明常规的MR成像能够显示周围神经损伤时信号和弛豫时间的异常,但是对于神经的再生缺乏特异性,而且无法显示复杂解剖部位的神经走行。磁共振弥散张量成像(diffusion tensor imaging, DTI)在显示中枢神经纤维束及据此诊断中枢神经系统疾病的作用已被广泛认同,部分研究也显示了DTI在周围神经系统损伤中的作用,然而常规场强下的在体DTI评价周围神经挤压伤作用尚缺乏系统研究。锰增强磁共振成像(manganese enhanced magnetic resonance imaging, MEMRI)被证实能够准确显示脑内神经传导通路,然而MEMRI评价周围神经损伤的应用尚很少见。
本课题采用1.5TMR对兔坐骨神经挤压伤模型进行弥散张量成像和锰增强MR成像,目的在于：①探讨1.5TMR兔坐骨神经弥散张量纤维束示踪的可行性及优化b值。②通过影像与病理及神经功能的对照,探讨DTI在评价外周神经损伤的作用。③影像与组织病理学及神经功能对照,初步探讨MEMRI在评价外周神经损伤及修复过程中的作用。以期为早期评价外周神经损伤及指导治疗方案的制定提供准确可靠的影像学方法。
第一部分兔坐骨神经弥散张量纤维束示踪参数优化及可行性研究
研究目的
通过不同b值之间成像质量的比较探讨b值对弥散张量(DTI)纤维束示踪成像质量的影响；明确1.5T MR兔坐骨神经弥散张量纤维束示踪成像的最优b值；比较不同b值之间各导出量判断b值对导出量的测量有无影响；通过与大体解剖及常规MR扫描相对照探讨1.5TMR兔坐骨神经弥散张量纤维束示踪成像的可行性。
1材料与方法
1.1实验对象：
健康新西兰大白兔10只,体重2.0-3.0Kg.由广州中医药大学提供。
1.2仪器、设备及药品
Philips Achieva 1.5T Nova Dual MR扫描仪
上海辰光医疗科技有限公司生产的实验动物(兔)专用相控阵列射频线圈
Philips EWS (Extended Workspace) v2.6图像后处理工作站,内装FiberTrak
纤维束示踪后处理软件包
3%戊巴比妥钠,速眠新2代
1.3 MR扫描
1.3.1动物准备
新西兰大白兔经耳缘静脉注射3%戊巴比妥钠(30mg/kg)及0.2-0.3ml速眠新2代肌注麻醉后行MR扫描。
1.3.2 MR扫描方法及成像参数
扫描序列包括常规T2WI、T1WI和DTI。
T2WI采用快速自旋回波(Turbo Spin-echo,TSE)序列,参数如下：TR,2500ms；TE,120ms；TSE因子,25；层厚,2mm；层间距,1mm；扫描视野(Field of View,FOV),120mm；扫描矩阵：344×344,平面内像素大小0.35x0.35mm；重建矩阵,512×512；平面内重建像素,0.23×0.23；激励次数(NSA),4次。
T1WI亦采用TSE序列：TR,572ms；TE,17ms；TSE因子：3；脂肪抑制采用频谱预饱和反转恢复序列(SPIR)；余扫描参数同T2WI。
DTI扫描采用单次激发自旋回波回波平面成像(SE-EPI)及SENSE并行采集技术,SENSE因子为2；采用6个不同b值：400、600、800、1000、1200和1400s/mm2,成像参数：TR,1550-3650ms:TE分别为55、59、63、66、68和71ms；6个b值扫描的方向、范围及其他参数均相同：层厚,1.6 mm；间隔,Omm；FOV,130mm；扫描矩阵80×80；平面内像素大小,1.6×1.6；激励次数(NSA),2；弥散敏感梯度方向：32个。
1.4数据处理
DTI数据传入Philips EWS v2.6工作站FiberTrak软件包行图像后处理。纤维束示踪采用多个感兴趣区(ROI)定义法：于股骨大转子层面采用freehand模式划定2个相距5mm的ROI,刚好包含坐骨神经外径,软件自动显示出穿过2个ROI的纤维束。纤维束FA下限阈值取0.4,最大偏转角为27度,最小纤维束长度10mm,然后将纤维束与T2WI定位图融合。
1.5数据分析：
1.5.1通过软件自动计算纤维束数量、平均长度、纤维束FA值、ADC值、3个本征向量值(λ1、λ2、λ3)。
1.5.2测量并计算SNR
软件生成平均弥散加权成像(DWI)图,在坐骨神经上划定1个椭圆形ROI(5个像素),测量坐骨神经信号强度(SIn)；然后在兔肢体外距离肢体边缘1cm处地空气区域划定ROI(20个像素),测量背景噪声信号强度的标准差(SDbg)；按照公式计算信噪比：SNR=sIn/SDbg。
1.6图像质量主观评价
由两名有十年以上MR诊断经验的医师对每只新西兰兔6个不同b值下的纤维束示踪图像质量进行评估。阅片采用盲法评估。评估标准包括：纤维束长度、粗细、均匀度及边缘清晰度；根据图像质量进行排序(1为最好,6为最差)。
将4只实验兔处死后解剖右下肢坐骨神经,与质量最优的纤维束重建图像对比,主观评价纤维束走行与大体解剖的一致性。对侧下肢冰冻后在股骨转子水平垂直股骨长轴截断,重建纤维在T2WI、T1WI图投影图与横断面解剖对照,判断两者坐骨神经位置是否一致。
1.7统计学分析
采用SPSSl3.0统计软件。所有定量数据均采用均数±标准差形式表示。先对不同b值间纤维束数量、纤维束平均长度、FA值数据进行数据分析,符合正态分布时均数比较采用单因素方差分析(One-Way ANOVA),进行方差齐性检验,若方差齐采用Fisher法,若方差不齐采用近似F检验Welch法；若方差分析显著则进行组间多重比较,方差齐时多重比较采用Bonferroni法,方差不齐者采用Dunnett's T3法；若方差分析不显著则不进行多重比较。不同b值间主观评价分数的比较采用多个相关样本非参数检验,两评价者之间一致性检验采用kappa检验。SNR.FA值、ADC值及λ1、λ2、λ3值与b值的相关性检验采用Spearman相关检验。均以P<0.05为有统计学意义。
2结果
2.1 DWI图像信噪比(SNR)与b值的关系
图像信噪比(SNR)随b值增高SNR逐渐下降。SNR与b值呈负相关(P=0.000,r=-0.589)
2.2各组不同b值纤维束数量、平均长度比较
纤维束数量：b=1000 s/mm2时纤维束数量明显较其他组多,b=400 s/mm2和1400 s/mm2组数量最少。方差分析有显著性(F=21.641,P=0.000)；组间多重比较,b=1000 s/mm2组与b=400 s/mm2、b=600 s/mm2、b=1200s/mm2、b=1400s/mm2组间纤维束数量差异均有显著性差异(P值分别为0.000、0.002、0.000和0.000),与b=800 s/mm2纤维束数量无显著性差异。
纤维束平均长度：b=1000 s/mm2时纤维束平均长度较其他b值长,b=400 s/mm2时纤维束平均长度最短,方差分析有统计学意义(F=19.736,P=0.000)。组间多重比较采用Dunnett's T3法,b=1000 s/mm2组与b=400 s/mm2、b=600 s/mm2、b=1200 s/mm2、b=1400 s/mm2组之间差异有统计学意义(P值分别为0.000、0.008、0.013和0.002),与b=800 s/mm2组间差异无统计学意义。
2.3 FA值和ADC值及3个本征向量值(λl、λ2、λ3)与b值的关系
各组间FA值均数介于0.51+0.02和0.53±0.03之间,各组FA值总体组间差别无统计学意义,且与b值无相关性。纤维束的ADC值、3个本征向量值(λ1、λ2、λ3)均随b值增加而逐渐降低,Spearman相关分析显示ADC值、3个本征向量值(λ1、λ2、λ3)均与b值成负相关(P=0.000,r分别为-0.787、-0.898、-0.829和-0.559)。
2.4图像质量主观评价：
b=1000 s/mm2组的图像质量等级评分均在前3名之内,平均秩次最低(图像质量最高),其次为b=800 s/mm2和1200 s/mm2组,b=400组平均秩次最高(图像质量最差),组间差异有统计学意义(χ=33.714,P=0.000)；两评价者之间具有高度一致性(κ=0.76,P=0.000)。
选择b=1000 s/mm2组的纤维束示踪图像与大体解剖和断面解剖对比,结果显示重建纤维束的走行方向和部位与大体解剖一致,基本能够显示坐骨神经全程,坐骨神经的两个分支构成-胫神经和腓神经亦能较清楚显示。
结论
1.不同的b值对DTI外周神经纤维束示踪的信噪比以及纤维束长度和数量产生显著的影响,从而影响重建图像的质量。
2.不同b值对坐骨神经ADC值及本征向量值的测量会产生影响,而FA值的测量不受影响。
3.在1.5TMR系统下进行兔坐骨神经DTI纤维束示踪成像时,取b值为1000s/mm2时可以获得最好的图像质量。
4.采用本研究的扫描及后处理方法在1.5TMR系统下进行兔坐骨神经弥散张量纤维束示踪成像是可行的。
第二部分弥散张量纤维束示踪评价兔坐骨神经挤压伤的实验研究
研究目的
明确坐骨神经损伤远段及近段FA值、ADC值、λ//和λ⊥随时间变化情况。与组织病理学改变及神经功能评分相对照,判断坐骨神经挤压伤远段FA值、ADC值、λ//和λ⊥与组织学改变和神经功能相关性。
材料与方法
1.1研究对象
新西兰大白兔36只,体重2.0-3.0Kg,平均2.4kg.由广州中医药大学提供。36只兔随机分组：4只作为正常组,其余32只为损伤组。损伤组32只兔按照神经损伤后恢复时间随机分为24小时、4天、8天、2周、4周、6周、8周、10周共8组,每组4只。
1.2仪器、设备及药品
荷兰Philips Achieva 1.5T Nova Dual双梯度MR扫描仪。
上海辰光医疗科技有限公司生产的实验动物(兔)专用相控阵列射频线圈,
型号CG RBC 18-H150-AP。
Philips EWS (Extended Workspace) v2.6图像后处理工作站,内装FiberTrak
纤维束示踪后处理软件包。
麻醉药品：3%戊巴比妥钠,速眠新2代。
外科无菌手术包
16cm无菌持针钳1把
1.3坐骨神经急性挤压伤动物模型制作
实验组兔经耳缘静脉注射3%戊巴比妥钠(30mg/kg)及0.2-0.3ml速眠新2代肌注麻醉。在无菌操作下暴露右侧坐骨神经,于股骨转子下方1cm处以大号持针钳垂直夹持坐骨神经,锁至第一扣,夹持5分钟后解除,可见神经纤细变薄但未离断。在夹伤处神经外膜用尼龙线缝一针作为标记。然后以0.2mm丝线分层缝合切口。对侧相同部位行假手术。
1.4 MR扫描
分别于挤压伤模型完成后24小时、4天、8天、2周、4周、6周、8周、10周对每组动物行MR扫描。
1.4.1动物准备：
新西兰大白兔经耳缘静脉注射3%戊巴比妥钠(30mg/kg)及0.2-0.3ml速眠新2代前肢肌注复合麻醉后行MR扫描。
1.4.2 MR扫描方法及成像参数
T1WI和T2WI扫描参数同第1部分。
DTI扫描参数：b=1000 s/mm2;TR/TE:8184/66 ms;层数：35；扫描时间：9分46秒；余参数同第一部分DTI扫描
1.5纤维束重建
使用Philips EWS v2.6工作站的FiberTrak软件包自动生成彩色编码FA图,进行纤维束示踪重建后将T1WI和T2WI解剖定位图与FA图融合。纤维束示踪采用多个感兴趣区(ROI)定义法：结合坐骨神经横断面T2WI定位图及彩色编码FA图,于股骨大转子层面采用freehand模式划定2个相距10mm的ROI,刚好包含坐骨神经外径,软件自动显示出穿过2个ROI的纤维束。纤维束FA下限阈值取0.35,最大偏转角为27度,最小纤维束长度30mm,然后将纤维束与T2WI定位图融合。
1.6数据测量
在彩色编码FA图上分别在损伤侧(右侧)坐骨神经损伤处远端1cm处、损伤近端1cm处以及假手术侧相应位置划定感兴趣区(ROI)。软件自动计算出相应ROI的FA值、ADC值及3个本征向量值(λl、λ2λ3)。计算λ⊥=(λ2+λ3)/2；λ∥=λ1。
1.7组织学检查
在DTI扫描完成之后,分别于24小时、4天、2周、4周、6周、8周和10周处死动物,暴露右侧坐骨神经并于钳夹处远端1cm处截取1mm神经组织置于4%预冷戊二醛溶液中预固定,放置4℃冰箱保存,1%四氧化锇后固定,系列乙醇脱水,Epon812环氧树脂包埋,超薄切片,经醋酸铀和柠檬酸铅双重染色后,用Hitachi H7500电子显微镜观察；远端再截取5mm神经组织10%甲醛溶液固定,石蜡包埋后行横断面及纵切面切片,HE染色后光镜下观察。
1.8功能评价：
每次扫描时对实验模型兔损伤侧肢体坐骨神经功能进行定量评价,趾张反射每个时间段每只兔对应1-4级神经功能分别记1-4分；改良Tarlov评分根据对应的0-4级神经功能分别记0-4分。每只兔满分8分,每组满分32分。
1.8.1趾张反射
手捏住兔颈部皮肤将兔提离地面,迅速使其在空中下降而不让其着地,观察患侧脚趾张开情况,1级：脚趾张开完全不可见；2级：仅可见轻度脚趾分开；3级：明显可见脚趾分开(但幅度仍低于正常水平)；4级：脚趾完全张开达到正常水平。
1.8.2 Tarlov评分
0级：肢体完全瘫痪,针刺无反应；1级：针刺刚刚可见有轻微反应；2级：损伤侧后肢所有关节有自主运动,手压使足部屈曲时无抵抗；3级：手压使足部屈曲时有明显抵抗,但走路步态不稳；4级：行走步态正常。根据0-4级分别记0-4分.
1.9统计学分析
采用SPSS13.0统计软件。所有定量数据均采用均数±标准差形式表示。损伤后8个不同时间点及正常组之间坐骨神经FA值、ADC值及λ//和λ⊥均数的比较采用单因素方差分析(One-Way ANOVA),先进行方差齐性检验,若方差齐采用Fisher法,若方差不齐采用近似F检验Welch法；方差齐时多重比较采用LSD法,方差不齐者采用Dunnett's T3法；若方差分析不显著则不进行多重比较。均以P<0.05为有统计学意义。
2结果
2.1挤压伤后坐骨神经各段FA值、ADC值及本征向量值的变化
2.1.1损伤远段FA值、ADC值及本征向量值的变化
单因素方差分析显示,损伤远段各时间段FA值有显著性差异(F=47.283, P=0.000),LSD法组间多重比较显示：正常组FA与其余各组间均有显著性差异(P<0.01)；其余两相邻时间段之间FA值有显著性差异的有：24小时组与4天组(P=0.000)、4周组与6周组(P=0.000)。
损伤后各时间段损伤远段神经ADC值之间无显著性差异。
损伤后各时间段损伤远段神经λ//方差分析无显著性。损伤后各时间段损伤远段神经λ⊥方差分析有显著性(F=11.847,P=0.000).组间多重比较显示：正常组与4天、8天、2周、4周、6周组之间λ⊥有显著性差异(P<0.05),与24小时、8周、10周组之间无显著性差异；其余相邻两时间段之间λ⊥有显著性差异者为：24小时组与4天组(P=0.000)、4周组与6周组(P=0.007)。
2.1.2挤压伤后损伤侧近段FA值、ADC值及本征向量值的变化
单因素方差分析结果显示,损伤后各时间段FA值、ADC值λ//及λ⊥均无显著性差异。
2.1.3假手术侧FA值、ADC值及本征向量值的变化
单因素方差分析结果显示,损伤后各时间段FA值、ADC值λ//及λ⊥均无显著性差异。
2.2挤压伤后坐骨神经损伤远段组织病理学改变
正常坐骨神经神经束膜中可见神经纤维排列整齐,髓鞘及轴索结构清晰可见。损伤后24小时可见髓鞘部分髓鞘肿胀为主。损伤后4天轴索开始崩解,髓鞘内可见轴索碎片。雪旺细胞增生不明显。损伤后2-4周大量轴索消失,髓鞘广泛崩解,雪旺细胞开始广泛增生,广泛炎症细胞浸润。至损伤6周时可见少量新生轴索,增生的雪旺细胞排列成索状。损伤8周时可见大量新生轴索,明显可见髓鞘形成。损伤10周时大部分神经髓鞘结构形成,其中可见清晰的轴索结构。
2.3坐骨神经损伤后功能恢复情况
损伤后24小时实验兔趾张反射完全消失,但伤侧下肢对针刺多有反应,但患肢无张力(Tarlov 2级)；2周时个别动物患肢稍有张力(Tarlov 3级),趾张反射仍检测不到；4周时个别动物出现患侧脚趾轻微分开,多数动物下肢有张力；6周时多数动物出现脚趾轻微分开,患肢有力但行走步态达不到正常；8周时趾张反射恢复明显,个别达到正常；10周时趾张反射基本正常,几乎全部动物行走步态正常。
假手术侧肢体在术后各时间段有力并活动自如,被毛光整,趾张反射正常。
结论
1.坐骨神经挤压伤远段FA值和λ⊥在损伤后发生了显著变化,其变化程度与组织学及神经功能改变有较好的相关性,能够反映神经Wallerian变性和再生情况。
2.损伤远段ADC值和λ//损伤后各时间段均未出现显著变化。
3.坐骨神经损伤近段在损伤后各时间段FA、ADC、λ//λ(?)均未出现显著变化。
第三部分锰增强MRI评价兔坐骨神经挤压伤的初步研究
研究目的
评价MEMRI坐骨神经挤压伤近端强化程度与组织学和神经功能的相关性,探讨ME-MEI评价坐骨神经挤压伤的价值；探讨周围神经MEMRI锰离子的运输途径。
1材料与方法
1.1研究对象
新西兰大白兔24只,雌雄不限,体重2.0～3.OKg,平均2.4kg.由广州中医药大学动物实验部提供,24只兔随机分组：4只作为正常对照组,其余20只兔作为损伤组,损伤组20只兔按照神经损伤后恢复时间分为24小时组、2周、4周、6周、10周共5组,每组4只。
1.2仪器、设备及药品
磁共振扫描仪、线圈、图像工作站、麻醉药品、外科切开无菌手术包及16cm
持针钳同第二部分；
50ul微量注射器
AR级MnCl2·4H2O晶体(分子量,197.9g/mol;纯度>99%)；
聚乙二醇1540
1.3坐骨神经急性挤压伤动物模型制作
同第二部分
1.4MnCl2溶液配制：
聚乙二醇1540固体配置成50%聚乙二醇1540溶液备用。MnCl2·4H2O晶体加入50%聚乙二醇1540溶液配制成400mM MnCl2溶液。
1.5MnCl2注射
24小时组、2周组、4周组、6周组和10周组分别于损伤后相应的时间行MnCl2溶液神经鞘内注射。
新西兰大白兔麻醉后无菌操作下暴露坐骨神经,然后用50ul微量注射器分别向坐骨神经的两个主要分支——胫神经和腓神经内分别注射30ul和20ul 400mM的MnCl2溶液,注射点距离损伤部位约2cm。对侧下肢行假手术。
1.6 MR扫描
MnCl2注射后24h行T1WI扫描。
T1WI和T2WI扫描方法及参数同第二部分。
1.7数据测量：
在坐骨神经损伤近端选取3个不同的位置：股骨大转子水平、股骨大转子近端1cm处和坐骨结节水平,以下分别称为近段、中段和远段。分别测量3个感兴趣区的坐骨神经信号强度S和对侧坐骨神经信号强度Sc,再测量股骨近端内侧肌肉信号强度Sm作为内参照,计算相对信号强度AS=(S-Sc)/Sm。
1.8组织学检查：
同第二部分相应章节
1.9功能评价：
同第二部分相应章节
1.10统计学分析
采用SPSS 13.0统计软件。所有定量资料均采用均数±标准差形式表示。坐骨神经在损伤后不同时间段相对信号强度△S的比较及各时间段近段与中段、中段与远段相对信号强度的比较分别采用单因素方差分析和双因素方差分析,先进行正态性和方差齐性检验,若方差齐采用Fisher法方差分析,若方差不齐采用近似F检验Welch法,方差分析有显著性者进行多重比较,若方差齐则采用LSD法,方差不齐则采用Dunnett's T3法；若方差分析不显著,则不进行多重比较。规定P<0.05为有统计学意义。
2结果
2.1正常组及损伤后各时间段的坐骨神经MEMRI表现
正常组坐骨神经在远端注射400mM MnCl2溶液后可在MRTIWI图像上观察到神经干的明显强化,强化范围均达到垂直段以上,强化程度由远端到近段逐渐减低。挤压伤24小时后的动物坐骨神经损伤点近端绝大多数未见明确强化；损伤后2周和4周组损伤近端可见由远及近的轻度强化,,但范围较短。损伤后6周时和10周时坐骨神经近段可见明显强化,但强化程度不如正常组,强化程度亦由远端到近段逐渐减低。
2.2坐骨神经损伤近侧近段、中段和远段在损伤后不同时间段锰增强后的相对信号强度(△S)比较
单因素方差分析显示：
损伤近侧近段、中段及远段在不同损伤时间段的相对信号强度(△S)均有显著性差异(F=5.819, P=0.002; F=26.987, P=0.000; F=45.406, P=0.000)。
近段损伤后各时间段相对信号强度(△S)多重比较显示,正常组与24小时、2周、4周和6周组间差异有显著性(P<0.05),正常组与10周组无显著差异；其余各相邻时间段之间均无显著性差异。
中段损伤后各时间段相对信号强度(△S)多重比较显示,正常组与其余各时间段之间均有显著性差异(P≤0.001)；损伤后24小时、2周、4周组之间无显著性差异,6周组和10周组与24小时、2周、4周组之间均有显著性差异(P≤0.011,P<0.01),6周与10周组之间无显著性差异。
远段损伤后各时间段相对信号强度(△S)多重比较显示,正常组与其余各时间段之间均有显著性差异(P=0.000)；损伤后24小时、2周、4周组之间无显著性差异,6周组和10周组与24小时、2周、4周组之间均有显著性差异(P≤0.001,P=0.000),6周与10周组之间无显著性差异。
2.3双因素方差分析显示不同时间段损伤近侧之近段、中段及远段相对信号强度(△S)之间有显著性差异(F=90.928, P=0.000); LSD法多重比较显示：近段与中段(P=0.000)、近段与远段(P=0.000)以及中段与远段(P=0.001)之间相对信号强度(△S)均有显著性差异。
2.4挤压伤后坐骨神经损伤远段组织病理学改变
见第二部分相应章节
2.5坐骨神经损伤后功能恢复情况
第二部分相应章节
2.6 MEMRI坐骨神经损伤近侧相对信号强度(△S)与神经功能评分的相关性
Spearman相关检验显示,坐骨神经损伤近侧的近段、中段和远段相对信号强度(△S)随损伤后恢复时间的变化与坐骨神经功能评分均呈正相关,相关系数分别为r=0.670(P=0.000),r=0.888(P=0.000)和r=0.899(P=0.000).
结论
1.坐骨神经挤压伤MEMRI显示损伤后各时间段损伤近段强化程度与组织学和神经功能有较好的相关性,MEMRI能够反映坐骨神经损伤后再生情况。
2.坐骨神经MEMRI的强化程度有远及近逐渐减弱,一定程度说明Mn2+的逆向传递有被动弥散的因素。
Background
Peripheral nerve injury(PNI) is a common complication of severe trauma, the most common types including crush injury and strech injury,et al. Severity of the injury is an important factor for clinicians to determine what kind of therapy to apply. Conservative approach is sufficient for mild nerve injury, but in severe injury, surgical repair is often needed to ensure the fuctional recovery of the nerve. To judge the severity of PNI, observation of nerve regeneration for several weeks to months is needed, and many patients lost the best opportunity of therapy and result in irreversible disability. Therefore, an auxiliary method that can accurately detect the severity and regeration of the injuried nerve is necessary to help clinicians to apply proper therapy for PNI.
Electrophysiological method is mainly used to detect the degree of PNI in addition to physical examination in clinical practice nowadays, which is known as gold standard to evaluate the function of peripheral nerve. However, Electrophysiological method is largely dependent on the experience of the operator, lack of sensitivity and specificity and can not obtain the anatomical information which is very important for preoperative evaluation and intraoperative localization of PNI.
Although some research showed that conventional MRI is useful to reveal abnormality of signal and time of relaxation of PNI, but it lacks of specificity in detection of nerve regeneration and can not display the running course of nerves with complex anatomy. It is generally accepted that diffusion tensor MR imaging (DTI) can display nerve fibers and diagnose diseases in central nervers system, some researches also showed it is helpful in PNI. Nevertheless, systemic research of in vivo DTI evaluating PNI using conventional MR scanner has never been reported. It is reported that manganese enhanced magnetic resonance imaging(MEMRI) can accurately reveal pathways in the brain, but research of MEMRI evaluating PNI is very few.
The purpose of this study lies in:①To explore the feasibility of in vivo diffusion tensor fiber tractography of rabbit sciatic nerve at 1.5 T MR. (2)To establish animal model of peripheral nerve crush injury and investigate the value of DTI in evaluation of peripheral nerve crush injury and regeneration by comparing with histopathology and nerve function assessment.③To investigate the value of MEMRI in evaluation of peripheral nerve crush injury and regeneration by comparing with histopathology and nerve function assessment. This study aims at providing a reliable imaging method to accurately evaluate PNI and direct therapy.
Part One:Study on Optimal Parameter and Feasibility of Diffusion Tensor Tractography of Rabbit Sciatic Nerve
Objective
To determine the effect of b value on the image quality of DTI tractography by comparing the image quality under different b values and optimal b value of diffusion tensor tractography of rabbit sciatic nerve at 1.5 T MR. To determine the effect of b value on measurement of the derived DTI paramertes; To investigate the feasibility of diffusion tensor tractography of rabbit sciatic nerve at 1.5 T MR by comparing reconstructed sciatic nerve fiber with gross anatomy and conventional MR images of sciatic nerve.
1 Materials and methods
1.1 Subjects
10 healthy New Zealand white rabbits weighting 2.0-3.OKg provided by Guangzhou traditional Chinese Medicine University
1.2 Instruments, equipment and drugs
Philips Achieva 1.5T Nova Dual MR scanner
Dedicated rabbit coil manufactured by Chenguang medical technology company
Ltd. Of Shanghai.
Philips EWS (Extended Workspace) v2.6 post processing workstation with
FiberTrak software package for fiber tracking.
3% sodium pentobarbital and Su-Mian-XinⅡ
1.3 MR scan
1.3.1 Animal preparation:
New Zealand rabbits underwent MR scan after anesthetized by 3% sodium pentobarbital (30mg/kg) intravenously and 0.2-0.3ml Su-Mian-XinⅡintramuscularly
1.3.2 MR scan method and parameters
MR sequence including conventional T2WI,T1WI and DTI.
TSE sequence were used for T2WI scan, following parameters were used:TR, 2500ms; TE,120ms; TSE factor,25; slice thickness,2mm; gap, 1mm; FOV, 120mm; matrix:344×344, in plane pixel size,0.35×0.35mm; NSA,4.
TSE sequence were used for T1WI scan:TR,572ms; TE,17ms; TSE factor,3; SPIR were used for fat suppression, other parameters were the same as T2WI.
Single shot SE-EPI sequence and SENSE parallel scanning technique were used for DTI scan. Six different b values were used:400、600、800、1000、1200和1400 s/mm2, scan parameters were as follow:TR,1550-3650ms; TE,55,59,63,66,68 and 71ms respectively; Direction, range and other parameters were same under 6 b values:slice thickness,1.6 mm; gap,0mm; FOV,130mm; matrix,80×80; in plane pixel size,1.6×1.6; NSA,2; MPG:32.
1.4 Data processing
Philips EWS (Extended Workspace) v2.6 post processing workstation with FiberTrak software package were used. Multiple ROI method was used:2 ROIs of 5mm apart were drawn in freehand mode at the level of greater trochanter of femur, including the outer diameter of sciatic nerve. Fibers running through the 2 ROIs were computed automatically by the software. The lower limit of FA value was defined as 0.4, maximum angle change was 27 degree, minimum fiber length was 10mm, then the fibers was fused with the T1 WI or T2WI images.
1.5 Data analysis:
1.5.1 Number, mean length, FA, ADC and three Eigen value (λ1、λ2、λ3) of the fibers were computed by the software automatically
1.5.2 Measurement and computation of signal to noise ratio (SNR) DWI images were produce by the software, a ROI (5 pixels) was drawn on the sciatic nerve to measure the signal intensity of the nerve (SIn); ROI of 20 pixels was drawn on the air region that lcm outside the body margin of rabbit limb to measure the standard deviation of background noise (SDbg); SNR was computed according to the following formula:SNR=SIn/SDbg。
1.6 subjective assessment of image quality
Assessment of the quality of fiber tractography under 6 different b values was conducted by 2 radioglogists with more than10 years experience of MR diagnosis. Blind method was adopted. The principle to consider including:fiber lenth, diameter, homogeneity and smoothnesss of fiber margin. Sorting of the image quality was made afterwards(l is defined as the best quality and 6 the worst)
Four rabbits were sacrificed; the sciatic nerve was dissect and compared with the reconstructed fibers of the best quality to evaluate the consistency between the tractography and the gross anatomy. Contralateral legs were frozen and transversely desected at the level of great trochanter of femur. Reconstructed fibers fused with the T2WI and T1WI images were compared with sectional anaotomy to test whether the position of fibers was in accordance.
1.7 Statistical analysis
SPSS 13.0 software was used. All data was presented as the mean±standard deviation, Test of normality was done in advance, One-way Anova was used to compare means of number, mean length, FA, ADC and three Eigen value (λ1、λ2、λ3) of the fibers under different b values if the data is of normal distribution. Fisher testwas used if variance was homogeneous, otherwise Welch test was used; If anova was significant, multiple comparison was conducted, Bonferroni test was used when the variance was homogeneous, Dunnett's T3 test was used otherwise. Multiple related sample nonparametic test was used in the comparison of scores of subjective quality assessment, and kappa consistency test was used to check interobserver agreement on tractographic image. Correlation between SNR, FA, ADC,λ1,λ2,λ3 and b value was conducted using the Spearman correlation test. P<0.05 was defined as statistical significant.
2 Result
2.1 Relation between SNR of DWI image and b value
SNR decreased with the rising of b value, SNR was negtively correlated with the b value (P=0.000,r=-0.589)
2.2 Number and mean length of fibers under different b value
Number of fibers of b=1000 s/mm2 was more than that of other b values, anova was significant (F=21.641, P=0.000); Multiple comparison showed significant difference between number of fibers of b= 1000 s/mm2 group and that of b=400,600, 1200,1400 s/mm2 group (P<0.001), but no significant difference between b=800 s/mm2 group.
Mean length of fibers of b=1000 s/mm2 was longer than that of other b values, anova is significant (F=19.736, P=0.000). Multiple comparison showed significant difference between mean length of b=1000 s/mm2 group and that of b=400,600,1200,1400 s/mm2 group (P=0.000,0.008,0.013,0.002, respectively), but no significant difference between b=800 s/mm2 group.
2.3 Relationship between FA, ADC,λ1,λ2,λ3 and b value
Means of FA in all of the groups was between 0.51±0.02 and 0.53±0.03, there was no significant difference between FAs of different groups, and there was no correlation between FA and b value.ADC,λ1,λ2.λ3 decreased with the rising of b value. Negative correlation between ADC,λ1,λ2,λ3 and b value were detected by Spearman correlation test (P=0.000, r=-0.787,-0.898,-0.829,-0.559, respectively).
2.4 Subjective assessment of image quality
Image quality ranking of b=1000 s/mm2 was always in the first 3 position with the lowest score(best quality), and score of b=400 s/mm2 group was the highest(worst quality)(X2=33.714,P=0.000). Interobserver agreement on tractographic image quality ranking was good (k=0.76, P=0.000)
Reconstructed fiber of b=1000 s/mm group was compared with the gross anatomy and transactional anatomy, comparison showed that the running course and location of the tractographic fiber is consistant with that of sciatic nerve anatomy, the two main branch within the sciatic nerve-tibial nerve and peroneal nerve was clearly revealed.
Conclusion
1. Different b value had significant effect on the SNR, number and length of fibers of peripheral nerve, thereby effect the image quality of DTI tractography..
2. Different b values had significant effect on ADC and eigenvalues measurement, but had no effect on FA measurement.
3. The optimal b value for DTI and fiber tractography of rabbit sciatic nerve at 1.5 T was 1000 s/mm2.
4. In vivo diffusion tensor fiber tractography of rabbit sciatic nerve at 1.5 T MR was feasible with parameters we used.
Part Two:Experimental Study on MR Diffusion Tensor Tractography in Evaluating Sciatic Nerve Crush Injury of Rabbit
Objective
To investigate the FA, ADC,λ//andλ(?) of distal and proximal to the crush injury changes with the different time course after injury. Comparing with histopathology and functional assessment, to determine the relationship between FA, ADC,λ//,λ(?) of distal part crushed sciatic nerve and histopathology and nerve function.
1 Materials and methods
1.1 Subjects
36 New Zealand white rabbits weighting 2.0-3.0Kg provided by Guangzhou traditional Chinese Medicine University were used.36 rabbits were randomly divided into normal control group(4 rabbits) and crush injury group(32 rabbits), the crush injury group(32 rabbits) were randomly divided into 8 groups(24hours,4 days, 8 days,2 weeks,4 weeks,6 weeks,8weeks and 10weeks) according to the time course after crush injury.
1.2 Instruments, equipment and drugs
MR scanner, coil, and anesthetic agent were the same as part I.
Sterile operating instrument set
A 16cm needle holder
1.3 Establishment of rabbit sciatic nerve crush injury model.
Rabbit was anesthetized with 3% sodium pentobarbital (30rng/kg) intravenously and 0.2-0.3ml Su-Mian-Xin II intramuscularly, sciatic nerve of right leg was exposed under sterile operation, sciatic nerve was crushed using a 16cm needle forceps at the level 1cm distal to the greater trochanter of femur, the forceps was locked to the first point of its ratchets and retained for 5 minutes before releasing. An epineurial suture was placed right above the crush site to exactly localize the position of the crush.Sham operation was performed on the contralateral leg.
1.4 MR scan
MR scan were performed 24 hours,4 days,8 days,2 weeks,4 weeks,6 weeks, 8weeks and 10weeks after injury,as well as the normal control group.
Animal preparation were the same as part I.
MR scan sequence and parameters were the same as part I except for some parameters of DTI scan:b=1000 s/mm2; TR/TE:8184/66 ms; slices:35; scan time:9 minute and 46 seconds.
1.5 Post processing of Data.
Philips EWS (Extended Workspace) v2.6 post processing workstation with FiberTrak software package were used. Multiple ROI method was used:2 ROIs of 10mm apart were drawn in freehand mode at the level of greater trochanter of femur(proximal to the injury site), including the outer diameter of sciatic nerve. Fibers running through the 2 ROIs were computed automatically by the software. The lower limit of FA value was defined as 0.35, maximum angle change was 27 degree, minimum fiber length was 30mm, then the fibers was fused with the T1WI or T2WI images.
1.6 Data measurement
Three ROIs were drawn in freehand mode on the color coded FA map generated by the software, two of the ROIs was drawn on the sciatic nerve of the injuried side,one of which is lcm distal and the other lcm proximal to the injury site; the third ROI was placed on the sham operated sciatic nerve on the site corresponding to the distal ROI of the injuried side.FA, ADC,and the three eigenvalues(λ1、λ2、λ3) were computed automatically by the software,λ(?) andλ//were computed as following formula:λ(?)=(λ2+λ3)/2,λ//=λ1
1.7 Histopathological examination
Animals were sacrificed after MR scan, and sciatic nerves of the injuried side that 1cm distal to the injured site were harvested. One part of the nerve were fixed in 4% paraformaldehyde, dehydrated, osmificated, and embedded in plastic resin. Semithin sections were then made and stained with uranyl acetate and lead citrate for transmission electron microscopic examination (Hitachi H7500). Another part of the sample were fixed in 10% paraformaldehyde and sectioned longitudinally at a 2-mm thickness. The sections were stained with standard hematoxylineosin staining for light microscopic examination.
1.8 Functional assessment
Functional assessment was performed befor every MR scan.
Toe-spreading reflex:grade 1,barely visible spreading of toes; grade 2, readily discernible though slight spreading of the toes; grade 3, unequivocal spreading of toes (though less forceful than normal); and grade 4, full spreading of the toes equal to normal.
Modified Tarlov score:grade 0, complete paraplegia of hind limbs; grade 1, hind limbs barely respond to hind limb pinch (barely detectable movement); grade 2, spontaneous movement at all hind limb joints but no resistance when flexing the foot; grade 3, obvious resistance when flexing the foot with abnormal gait; and grade 4, walks with normal gait.
1-4 marks were recorded for one rabbit of corresponding toe spreading reflex grade;0-4 marks were recorded for one rabbit of corresponding Tarlov score grade. Full marks for a single rabbit is 8, and for a group is 32 marks.
1.9 Statistical analysis
SPSS13.0 software was used. All data was presented as the mean±standard deviation, One-way Anova was used to compare FA, ADC,λ//andλ(?) of sciatic nerve on the eight time point after injury as well as the normal control group. Fisher test was used if variance was homogeneous, otherwise Welch test is used; If anova was significant, multiple comparison was conducted, LSD test is used when the variance is homogeneous, Dunnett's T3 test was used otherwise. P<0.05 is defined as statistical significant.
2 Results
2.1 FA, ADC,λ//andλ(?) changes of sciatic nerve after crush injury
2.1.1 FA, ADC,λ,//andλ(?) changes of distal portion of injuried sciatic nerve.
One-way Anova showed that FA value of the distal portion injuried sciatic nerve at different time point is different significantly. (F=47.283, P=0.000), LSD multiple comparison test showed:FA of normal control group is significantly different from that of all the other group (P<0.01);Other significant differences of FA between consecutive groups were:24 hours group and 4 days group(P=0.000),4 weeks group and 6 weeks group(P=0.000).
There was no significant difference between ADC of every time point of distal portion injuried sciatic nerve.
There was no significant difference betweenλ//of every time point of distal portion injuried sciatic nerve.
There was significant difference betweenλ(?) of time points of distal portion injuried sciatic nerve (F=11.847, P=0.000). LSD multiple comparison test showed:λ(?) of normal control group is significantly different from that of 4 days,8 days,2 weeks,4 weeks,6 weeks group (P<0.05), but not significantly different from that of 24 hours,8weeks and 10 weeks group, Other significant differences ofλ(?) between consecutive groups were:24 hours group and 4 days group (P=0.000),4 weeks group and 6 weeks group(P=0.007)
2.1.2 FA, ADC,λ//andλ(?) changes of proximal portion of the injuried injuried sciatic nerve.
One-way Anova showed that There was no significant difference between FA、ADCλ//andλ(?) of every time point of proximal portion injuried sciatic nerve.
2.1.3 FA, ADC,λ//andλ(?) changes of sham operated sciatic nerve.
One-way Anova showed that There was no significant difference between FA、ADC,λ//andλ(?) of every time point of sham operated sciatic nerve.
2.2 Histopathological changes of distal portion of the injuried sciatic nerve.
Normal sciatic nerve showed that nerve fibers lines up in order, myelin sheath and axon was clearly seen.Swelling of the axon an myelin sheath was seen 24 hours after injury; axon began degeneration 4 days after injury, and fragment of axons can be seen.Proliferation of Schwann cells was not obvious; large amounts of axons disappeared and myelin sheath generally disintegrate 2-4 weeks after injury,and Schwann cells began to proliferate generally, inflammatory cell infiltration is obvious. Small amount of regenerated axons were detected until 6 weeks after injury, which wass very thin and amyelinated. Large amount of regenerated axons were revealed 8 weeks after injury.10 weeks after injury, most regenerated axons-were myelinated.
2.3 Functional recovery of sciatic nerve after injury
Toe-spreading reflex disappered 24 hours after injury, but the injuried leg respond to pinch but have no tension (Tarlov 2); Slight tension of the injured leg can be felt in individual animals 2 weeks after injury (Tarlov 3),but Toe-spreading reflex was still undetectable.Indivadule animals can be seen slight spreading of injuried side toes, strong force of the injuried leg can be detected in most animals; 6 weeks after injury, most animals revealed slight spreading and strong force of the injured leg, but the gait was still abnormal; 10 weeks after injury, some animals showed normal toe-spreading reflex, almost all the animals had normal gait.
Sham-operated legs was strong and moved freely after injury, and the fur of the legs was smooth and intacted.The toe spreading reflex was normal.
Conclusion:
1. FA andλ(?) of distal portion of injuried sciatic nerve changed significantly after injury, and the change were in good correlation with the histopathology and fuctional recovery. FA andλ(?) distal portion of injuried sciatic nerve can reflect Wallerian degeneration and regeneration after crush injury of sciatic nerve.
2. ADC andλ,//of distal portion of injuried sciatic nerve had no significant change after injury.
3. FA, ADC,λ,//andλ(?). of proximal portion of the injuried injuried sciatic nerve had no significant change after injury.
Part Three:Preliminary Study of Manganese Enhanced MRI in Evaluating Sciatic Nerve Crush Injury of Rabbit
Objective
To investigate the value of MEMRI in evaluation of nerve regenertation after sciatic nerve crush infury of rabbit by comparing with histopathology and functional assessment, crush injury; To explore the transporting pathway of Mn2+ in MEMRI of peripheral nerve.
1 Materials and methods
1.1 Subjects
24 New Zealand white rabbits weighting 2.0-3.OKg provided by Guangzhou Traditional Chinese Medicine University were used.24 rabbits were randomly divided into normal group(4 rabbits) and crush injury group(20 rabbits), the crush injury group were randomly divided into 5 groups(24hours,2 weeks,4 weeks,6 weeks and 10weeks) according to the recovery time after injury. There were 4 rabbits in each group.
1.2 Instruments, equipment and drugs
MR scanner, coil, and anesthetic agent were the same as part I. Sterile operating instrument set for crush injury modle was the same as part II. AR class MnCl2·4H2O crystal (molecular weight 197.9g/mol; purity>99%) polyethylene glycol(PEG) 1540
1.3 Establishment of rabbit sciatic nerve crush injury model
The procedure was the same as part II.
1.4 Preparation of MnCl2 solution
50% PEG 1540 solution was prepared by PEG 1540 solid melted and dissolved in 0.9% NaCl solution. MnCl2·4H2O crystal was dissolved in 50% PEG 1540 to form 400mM MnCl2 solution.
1.5 MnCl2 injection
After corresponding recovery time, rabbits were anesthetized again in the same manner as the initial operation. The sciatic nerve of the right leg was exposed, then 30 and 20ul of 400 mM MnCl2 were injected into the two major branches of the sciatic nerve, the tibial nerve and peroneal nerves, respectively. The injection site was 2cm distal to the injury site. Sham operation were performed on the contralateral legs. 1.6 MR scan
T1WI and T2WI sequence were scanned after 24 hours after MnCl2 injection.The technique and parameters used were the same as part II 1.7 Data measurement
ROIs were placed on three different portion of injuried sciatic nerve proximal to injury site:Greater trochanter of femur, 1cm proximal to Greater trochanter of femur and ischial tuberosity, which were defined as distal, middle and proximal segment respectively in this study. Signal intensity of the three ROIs on the sciatic nerve were recorded as S and the signal intensity of the contralateral sciatic nerve was recorded as Sc, the signal intensity of muscle medial to the proximal femur (Sm) was recorded as a reference. Relative signal intensity of injuried sciatic nerve was computed as AS, which was calculated according to the formula:ΔS=(S-Sc)/Sm.
1.8 Histopathologic examination
The same as partⅡ.
1.9 Functional assessment
The same as partⅡ.
1.10 Statistical analysis
SPSS 13.0 software was used. All data was presented as the mean±standard deviation, One-way Anova and Two-way Avova were used respectively to compare AS of different recovery time after injury as well as the normal control group andAS between three different portion of injuried sciatic nerve. Fisher test was used if variance was homogeneous, otherwise Welch test was used; If anova was significant, multiple comparison was conducted, LSD test was used when the variance is homogeneous, Dunnett's T3 test was used otherwise. If anova was not significant, multiple comparison would not be conducted. P<0.05was defined as statistical significant.
2 Results
2.1 MEMRI manifestation of normal group and injuried groups of sciatic nerve
Remarkable enhancement of the sciatic nerve on the T1WI images was revealed in the normal group after MnCl2 injection, the extent of enhancement was generally beyond the vertical portion of the sciatic nerve, and the enhancement decreased gradually from distal to proximal portion. There was no enhancement of the nerve in most animals of the 24 hours group after injury.2-4 weeks after injury, the sciatic nerve proximal to injury site revealed slight and short enhancement that decreased gradually from distl to proximal part.6 and 10 weeks after injury, the proximal part of the injuried sciatic nerve showed remarkable enhancement,but not as strong as normal group, the enhancement was also decreased gradually from distal to proximal portion.
2.2 Relative signal intensity (ΔS) changes of sciatic nerve after injury
One-way Anova showed that there were significant difference betweenAS of injuried sciatic at different time point after injury at proximal, middle and distal portion (F=5.819, P=0.002; F=26.987, P=0.000; F=45.406, P=0.000)
In the proximal portion, multiple comparison showed thatΔS of normal control group was significantly different from that of 24 hours,2 weeks,4 weeks and 6 weeks group (P<0.05), there was no significant difference betweenAS of normal group and 10 weeks group. There was no significant differences ofΔS between other consecutive groups.
In the middle portion, multiple comparison showed that AS of normal control group was significantly different from that of all the other groups (P<0.001),there was no significant difference betweenAS of 24 hours,2 weeks and 4 weeks groups.ΔS of 6 weeks and 10 weeks group was significantly different from that of 24 hours, 2 weeks and 4 weeks group(P<0.001,P=0.000). There was no significant differences ofΔS between 6 weeks and 10 weeks group.
In the distal portion, multiple comparison showed thatΔS of normal control group is significantly different from that of all the other groups (P=0.000), there was no significant difference betweenAS of 24 hours,2 weeks and 4 weeks groups. AS of 6 weeks and 10 weeks group was significantly different from that of 24 hours,2 weeks and 4 weeks group (P≤0.011,P=0.000). There was no significant differences ofΔS between 6 weeks and 10 weeks group.
2.3 Two-way Anova showed that there was significant difference between the three portion of injuried sciatic nerve at different recovery time point (F=90.928, P=0.000); LSD multiple comparision showed:There were significant differences betweenAS of proxilmal and distal portion (P=0.000), proxilmal and middle portion (P=0.000) as well as middle portion and distal portion (P=0.001)
2.4 Histopathological changes of distal portion of the injuried sciatic nerve. The same as partⅡ
2.5 Functional recovery of sciatic nerve after injury The same as partⅡ
2.6 Correlation between MEMRI relative signal intensity of sciatic nerve proximal to injury site and functional assessment.
Speraman correlation test showed that the proximal,middle and distal portion of scitic nerve promximal to injuried site was significantly correlated with functional scores(r=0.670,P=0.000;r=0.888,P=0.000,r=0.899(P=0.000).
Conclusion
1. The MEMRI signal intensity of sciatic nerve crush injury was in good correlation with histopathology and nerve function recovery, MEMRI was able to reflect the regeneration of crushed sciatic nerve.
2. Enhancement of sciatic nerve decreased gradually from distal to proximal portion, infered that there was passive diffusion factor in the retrograde Mn2+ transportation.
引文
[1]Campbell WW. Evaluation and management of peripheral nerve injury [J]. Clin Neurophysiol,2008,119(9):1951-65.
[2]Noble J, Munro CA, Prasad VS, et al. Analysis of upper and lower extremity peripheral nerve injuries in a population of patients with multiple injuries [J]. J Trauma,1998,45(1):116-122.
[3]Winfree CJ. Peripheral nerve injury evaluation and management [J].Curr Surg, 2005,62(5):469-476.
[4]Cudlip SA, Howe FA, Griffi ths JR, et al. Magnetic resonance neurography of peripheral nerve following experimental crush injury, and correlation with functional deficit [J]. J Neurosurg,2002,96 (4):755-759.
[5]Aagaard BD, Lazar DA, Lankerovich L, et al. High-resolution magnetic resonance imaging is a noninvasive method of observing injury and recovery in the peripheral nervous system [J]. Neurosurgery,2003,53 (1):199-203.
[6]Stanisz GJ, Midha R, Munro CA, et al. MR properties of rat sciatic nerve following trauma [J]. Magn Reson Med,2001,45(3):415-420.
[7]Bendszus M, Stoll G. Technology insight:visualizing peripheral nerve injuryusing MRI [J]. Nat Clin Pract Neurol,2005, 1(1):45-53.
[8]Mayer AR, Ling J, Mannell MV, et al. A prospective diffusion tensor imaging study in mild traumatic brain injury [J]. Neurology,2010,74(8):643-50.
[9]Stadlbauer A, Nimsky C, Buslei R, et al. Diffusion tensor imaging and optimized fiber tracking in glioma patients:Histopathologic evaluation of tumor-invaded white matter structures [J]. Neuroimage,2007.,34(3):949-56.
[10]Thomalla G, Glauche V, Koch MA, et al. Diffusion tensor imaging detects early Wallerian degeneration of the pyramidal tract after ischemic stroke[J]. Neuroimage,2004,22(4):1767-74.
[11]Tovar-Moll F, Evangelou IE, Chiu AW, et al. Thalamic involvement and its impact on clinical disability in patients with multiple sclerosis:a diffusion tensor imaging study at 3T [J]. Am J Neuroradiol,2009,30(7):1380-6.
[12]Hiltunen J, Suortti T, Arvela S, et al. Diffusion tensor imaging and tractography of distal peripheral nerves at 3 T [J]. Clin Neurophysiol,2005,116(10):2315-23.
[13]Jambawalikar S, Baum J, Button T, et al. Diffusion tensor imaging of peripheral nerves [J]. Skeletal Radiol,2010,39(11):1073-9.
[14]Lehmann HC, Zhang J, Mori S, et al. Diffusion tensor imaging to assess axonal regeneration in peripheral nerves [J]. Exp Neurol,2010,223(1):238-44.
[15]Takagi T, Nakamura M, Yamada M, et al. Visualization of peripheral nerve degeneration and regeneration:monitoring with diffusion tensor tractography [J]. Neuroimage,2009,44(3):884-92.
[16]Stoll G, Wessig C, Gold R, et al. Assessment of lesion evolution in experimental autoimmune neuritis by gadofluorine M-enhanced MR neurography [J]. Exp Neurol,2006,197(1):150-156.
[17]Wessig C, Jestaedt L, Sereda M, et al. Gadofluorine M-enhanced magnetic resonance nerve imaging:comparison between acute inflammatory and chronic degenerative demyelination in rats [J]. Exp Neurol,210(1):137-143.
[18]Bendszus M, Stoll G. Caught in the act:in vivo mapping of macrophage infiltration in nerve injury by magnetic resonance imaging [J]. J Neurosci,2003, 23(34):10892-10896.
[19]Merritt JE, Jacob R, Hallam TJ. Use of manganese to discriminate between calcium influx and mobilization from internal stores in stimulated human neutrophils[J]. J Biol Chem 1989,264(3):1522-1527.
[20]Murayama Y, Weber B, Saleem KS, et al, Tracing neuralcircuits in vivo with Mn-enhanced MRI [J]. Magn Reson Imaging,2006,24(4):349-358.
[21]Cross DJ, Minoshima S, Anzai Y, et al. Statistical mapping of functional olfactory connections of the rat brain in vivo [J]. Neuroimage,2004,23(4), 1326-1335.
[22]Lee J.W., Park J.A., Lee J.J., et al. Manganese-enhanced auditory tract-tracing MRI with cochlear injection. Magn. Reson. Imaging [J].2007,25(5),652-656.
[23]Lin YJ, Koretsky AP. Manganese ion enhances T1-weighted MRI during brain activation:an approach to direct imaging of brain function [J]. Magn. Reson. Med.1997,38(3):378-388.
[24]Silva AC, Lee JH, Aoki I, et al. Manganese-enhanced magnetic resonance imaging (MEMRI):methodological and practical considerations [J]. NMR Biomed,2004,17(8):532-43.
[25]Einstein A. Investigations on the Theory of the Brownian Movement. New York:Dover; 1956
[26]Schaefer PW, Grant PE, Gonzalez RG. Diffusion-weighted MR imaging of the brain. Radiology [J].2000,217(2):331-345.
[27]Carr HY, Purcell EM. Effects of diffusion on free precession in nuclear magnetic resonance experiments [J]. Phys Rev,1954,94(3):630-638.
[28]Stejskal EO, Tanner JE. Spin diffusion measurements:spin echoes in the presence of a time-dependent field gradient [J]. J Chem Phys,1965,42(1): 288-292.
[29]Le Bihan D. Molecular diffusion nuclear magnetic resonance imaging [J]. Dig Imag Q,1991,7(1):1-30.
[30]Le Bihan D, Mangin JF, Poupon C, et al. Diffusion tensor imaging:concepts and applications [J]. J Magn Reson Imaging,2001,13(4):534-546.
[31]Shimony JS, McKinstry RC, Akbudak E, et al. Quantitative diffusion-tensor anisotropy brain MR imaging:normative human data and anatomic analysis [J]. Radiology 1999,212(3):770-784.
[32]Mori S, Crain BJ, Chacko VP, et al. Three-dimensional tracking of axonal projections in the brain by magnetic resonance imaging [J]. Ann Neurol,1999, 45(2):265-9.
[33]Xue R, van Zijl PC, Crain BJ, et al. In vivo three-dimensional reconstruction of rat brain axonal projections by diffusion tensor imaging [J]. Magn Reson Med, 1999,42(6):1123-7.
[34]Kilickesmez O, Yirik G, Bayramoglu S, et al. Non-breath-hold high b-value diffusion-weighted MRI with parallel imaging technique:apparent diffusion coefficient determination in normal abdominal organs [J]. Diagn Interv Radiol, 2008, (2):83-7.
[35]Pruessmann KP, Weiger M, Scheidegger MB, et al. SENSE:sensitivity encoding for fast MRI [J]. Magn Reson Med,1999,42(5):952-62.
[36]Bammer R, Keeling SL, Augustin M, et al. Improved diffusion-weighted singleshot echo-planar imaging (EPI) in stroke using sensitivity encoding (SENSE) [J]. Magn Reson Med,2001,46(3):548-54.
[37]Bammer R, Auer M, Keeling SL, et al. Diffusion tensor imaging using singleshot SENSE-EPI [J]. Magn Reson Med,2002,48(1):128-36.
[38]Jaermann T, Pruessmann KP, Valavanis A, et al. Influence of SENSE on image properties in high-resolution single-shot echo-planar DTI [J]. Magn Reson Med, 2006,55(2):335-42.
[39]Jones DK, Horsfield MA, Simmons A. Optimal strategies for measuring diffusion in anisotropic systems by magnetic resonance imaging [J]. Magn Reson Med,1999,42(3):515-25
[40]Papadakis NG, Murrills CD, Hall LD, et al. Minimal gradient encoding for robust estimation of diffusion anisotropy. Magn Reson Imaging [J].2000,18(6): 671-79.
[41]Hasan KM, Parker DL, Alexander AL. Comparison of gradient encoding schemes for diffusion-tensor MRI. J Magn Reson Imaging [J].2001,13(5): 769-80.
[42]Tuch DS, Reese TG, Wiegell MR, et al. Diffusion MRI of complex neural architecture [J]. Neuron,2003,40(5):885-95.
[43]Mukherjee P, Chung SW, Berman JI, et al. Diffusion tensor MR imaging and fiber tractography:technical considerations [J]. Am J Neuroradiol,2008, 29(5):843-52.
[44]Huang H, Zhang J, van Zijl PC, et al. Analysis of noise effects on DTI-based tractography using the brute-force and multi-ROI approach [J]. Magn Reson Med,2004,52(3):559-65.
[45]Lazar M, Alexander AL. An error analysis of white matter tractography methods:synthetic diffusion tensor field simulations [J]. Neuroimage, 20(2):1140-53.
[46]Oouchi H, Yamada K, Sakai K, et al. Diffusion anisotropy measurement of brain white matter is affected by voxel size:underestimation occurs in areas with crossing fibers[J]. Am J Neuroradiol,2007,28(6):1102-6.
[47]van Gelderen P, de Vleeschouwer MHM, DesPres D, et al. Water diffusion and acute stroke [J]. Magn Reson Med,1994,31(2):154-163.
[48]Le Bihan D, Breton E, Lallemand D, et al. Separation of diffusion and perfusion in intravoxel incoherent motion MR imaging [J]. Radiology,1988, 168(2):497-505.
[49]Kim HJ, Choi CG, Lee DH, et al. High-b-value diffusion-weighted MR imaging of hyperacute ischemic stroke at 1.5T [J]. Am J Neuroradiol,2005, 26(2):208-215.
[50]Saupe N, White LM, Stainsby J, et al. Diffusion tensor imaging and fiber tractography of skeletal muscle:optimization of B value for imaging at 1.5 T [J]. AJR Am J Roentgenol,2009,192(6):W282-290.
[51]Melhem ER, Itoh R, Jones L, et al. Diffusion tensor MR imaging of the brain: effect of diffusion weighting on trace and anisotropy measurements [J]. Am J Neuroradiol,2000,21(10):1813-20.
[52]Mukherjee P, Chung SW, Berman JI, et al. Diffusion tensor MR imaging and fiber tractography:technical considerations[J]. AJNR Am J Neuroradiol, 2008,29(5):843-52.
[53]李德军,包尚联,马林.不同b值和扩散张量成像导出量的定量关系研究[J].中华放射学杂志,2004,38：118-123.
[54]Yoshiura T, Wu O, Zaheer A, et al. Highly diffusion-sensitized MRI of brain: dissociation of gray and white matter [J]. Magn Reson Med,2001, 45(5):734-740.
[55]Morisaki S, Kawai Y, Umeda M, et al. In vivo assessment of peripheral nerve regeneration by diffusion tensor imaging[J]. Journal of Magnetic Resonance Imaging,2011,33(3):535-542.
[56]Andreisek G, White LM, Kassner A, et al. Diffusion tensor imaging and fiber tractography of the median nerve at 1.5T:optimization of b value[J]. Skeletal Radiol,2009,38(1):51-59.
[57]Schmitz HC, Beer GM. The toe-spreading reflex of the rabbit revisited--functional evaluation of complete peroneal nerve lesions[J]. Lab Anim.2001,35(4):340-345.
[58]Shen J, Zhou CP, Zhong XM, et al.MR neurography:T1 and T2 measurements in acute peripheral nerve traction injury in rabbits[J]. Radiology,2010, 254(3):729-738.
[59]Bridge PM, Ball DJ, Mackinnon SE, et al. Nerve crush injuries--a model for axonotmesis[J]. Exp Neurol,1994,127(2):284-290.
[60]卜令学,李宁毅,杨学财,等.下牙槽神经挤压伤后再生的实验研究[J].现代口腔医学杂志,2002,16(1)：21-23.
[61]Donaldson J, Shi R, Borgens R. Polyet hylene glycol rapidly restores physiological functions in damaged sciatic nerves of guinea pigs[J]. Neurosurgery,2002,50(1):147-157.
[62]Nawwar AM, Sherif MF, Barakat NG. Instrumented forceps for measurement of nerve compression forces[J]. J Biomech Eng,1995,117 (1):53-58.
[63]Rydevik B, Lundborg G. Permeability of intraneural microvessels and perineurium following acute, graded experimental nerve compression[J]. Scand J Plast Reconstr Surg,1977,11(3):179-87.
[64]Dahlin LB, Thambert C. Acute nerve compression at low pressures has a conditioning lesion effect on rat sciatic nerves[J]. Acta Orthop Scand,1993, 64(4):479-81.
[65]Chen L E, Seaber AV, Glisson RR, et al. The functional recovery of peripheral nerves following defined acute crush injuries[J]. J Ort hop Res,1992, 10(5):657-664.
[66]Matsuda K, Wang HX, Suo C, et al.Retrograde axonal tracing using manganese enhanced magnetic resonance imaging[J]. Neuroimage,2010.,50(2):366-74.
[67]李新春,陈健宇,王欣璐,等；兔坐骨神经急性挤压伤的MRI与病理学对比初步研究[J].中华放射学杂志,2004,38(2)：133-138.
[68]Seddon H. Three types of nerve injury. Brain 1943;66:237-88.
[69]Sunderland S:Nerves and Nerve Injuries, ed 2. London:Churchill Livingston, 1978.
[70]Lewis T, Pickering GW, Rothschild P:Centripetal paralysis arising out of arrested bloodflow to the limb[J]. Heart,1931,16:1-32.
[71]Ochoa J, Fowler TJ, Gilliatt RW. Anatomical changes in peripheral nerves compressed by a pneumatic tourniquet[J]. J Anat,1972,113(Pt 3):433-455.
[72]Koeppen AH. Wallerian degeneration:history and clinical significance[J]. J Neurol Sci,2004,220(1-2):115-117.
[73]Waller A:Experiments on the section of the glossopharyngeal and hypoglossal nerves of the frog, and observations of the alterations produced thereby in the structure of their primitive fibers[J]. Phil Trans Roy Soc,1850,140:423-429.
[74]Hall S. Mechanisms of repair after traumatic injury. In:Dyck PJ, Thomas PK, editors. Peripheral neuropathy. Philadelphia:Elsevier, Saunders; 2005. 1403-33.
[75]Kang H, Tian L, Thompson W. Terminal Schwann cells guide the reinnervation of muscle after nerve injury[J]. J Neurocytol 2003,32:975-85.
[76]Burnett MG, Zager EL. Pathophysiology of peripheral nerve injury:a brief review. Neurosurg Focus[J].2004,15; 16(5)
[77]Mackinnon SE, Dellon AL. Surgery of the peripheral nerve. New York:Thieme; 1988.
[78]Bendszus M, Wessig C, Solymosi L, et al. MRI of peripheral nerve degeneration and regeneration:correlation with electrophysiology and histology[J]. Exp Neurol,2004,188(1):171-177.
[79]Sheikh KA. Non-invasive imaging of nerve regeneration [J]. Exp Neurol.2010, 223(l):72-6.
[80]李青峰,谢峰.周围神经损伤中电生理测量技术的应用[J].整形再造外科杂志,2005,2(2)：102-105.
[81]Filler, A., Maravilla, K., Tsuruda, J. MR neurography and muscle MR imaging for image diagnosis of disorders affecting the peripheral nerves and musculature[J]. Neurol. Clin,2004,22(3),643-682.
[82]Lundborg, G.,2004. Nerve Injury and Repair, second ed. Elsevier, Philadelphia.
[83]Budde MD, Kim JH, Liang HF,et al. Axonal injury detected by in vivo diffusion tensor imaging correlates with neurological disability in a mouse model of multiple sclerosis[J]. NMR Biomed,2008,21(6):589-597.
[84]Wang S, Wu EX, Tam CN, et al. Characterization of white matter injury in a hypoxic-ischemic neonatal rat model by diffusion tensor MRI[J]. Stroke, 2008,39(8):2348-2353.
[85]Valsasina P, Rocca MA, Agosta F, et al. Mean diffusivity and fractional anisotropy histogram analysis of the cervical cord in MS patients[J]. Neuroimage,2005,26(3):822-828.
[86]Zhang J, Jones M, DeBoy CA, et al. Diffusion tensor magnetic resonance imaging of Wallerian degeneration in rat spinal cord after dorsal root axotomy[J]. J Neurosci,2009,29(10):3160-3171.
[87]Ford JC, Hackney DB, Alsop DC, et al. MRI characterization of diffusion coefficients in a rat spinal cord injury model [J]. Magn Reson Med 1994;31(5):488-494.
[88]Sakai K, Yamada K, Oouchi H, et al. Numerical simulation model of hyperacute/acute stage white matter infarction[J]. Magn Reson Med Sci 2008, 7(4):187-194.
[89]Cox LJ, Hengst U, Gurskaya NG, et al. Intraaxonal translation and retrograde trafficking of CREB promotes neuronal survival [J]. Nat Cell Biol 2008; 10(2):149-159.
[90]Davies AM. Regulation of neuronal survival and death by extracellular signals during development[J]. EMBO J,2003,22(11):2537-2545.
[91]Markus A, Patel TD, Snider WD. Neurotrophic factors and axonal growth[J]. Curr Opin Neurobiol,2002,12(5):523-531.
[92]Lauterbur PC. Image formation by induced local interactions:examples employing nuclear magnetic resonance [J]. Nature,1973,242:190-191.
[93]Wolf GL, Baum L. Cardiovascular toxicity and tissue proton T1 response to manganese injection in the dog and rabbit[J]. Am. J. Roentgenol,1983,141(1): 193-197.
[94]Narita K, Kawasaki F, Kita H. Mn and Mg influxes through Ca channels of motor nerve terminals are prevented by verapamil in frogs[J]. Brain Res.1990, 510(2),289-295.
[95]Wibeke Nordhoy, Henrik W, Anthonsen,et al. Manganese ions as intracellular contrast agents:proton relaxation and calcium interactions in rat myocardium[J]. NMR Biomed,2003,16(2):82-95.
[96]Hu TC, Pautler RG, MacGowan GA, et al. Manganeseenhanced MRI of mouse heart during changes in inotropy[J]. Magn. Reson. Med,2001,46(5):884-890.
[97]London RE, Toney G, Gabel SA, et al. Magnetic resonanceimaging studies of the brains of anesthetized rats treated withmanganese chloride [J]. Brain Res. Bull,1989,23(3):229-235.
[98]Lucchini R, Albini E, Placidi D, et al. Brain magnetic resonance imaging and manganese exposure[J]. Neurotoxicology,2000,21(5):769-775.
[99]Watanabe T, Natt O, Boretius S, et al. In vivo 3D MRI staining of mouse brain after subcutaneous application of MnCl[J]. Magn.-Reson. Med,2002,48(5): 852-859.
[100]Aoki I, Wu YJ, Silva AC, et al. In vivo detection of neuroarchitecture in the rodent brain using manganese-enhanced MRI[J]. Neuroimage,2004,22(3): 1046-1059.
[101]Pautler RG, Silva AC, Koretsky AP. In vivo neuronal tract tracing using manganese-enhanced magnetic resonance imaging[J]. Magn. Reson. Med,1998, 40(5):740-748.
[102]Allegrini PR, Wiessner C. Three-dimensional MRI of cerebral projections in rat brain in vivo after intracortical injection of MnCl2[J]. NMR Biomed, 2003,16(5):252-256.
[103]Van der Linden A, Verhoye M, Van Meir V, et al. In vivo manganese-enhanced magnetic resonance imaging reveals connections and functional properties of the songbird vocal control system[J]. Neuroscience,2002,112(2):467-474.
[104]Pautler RG, Mongeau R, Jacobs RE. In vivo trans-synaptic tract tracing from the murine striatum and amygdala utilizing manganese enhanced MRI (MEMRI) [J].Magn Reson Med,2003,50(1):33-39.
[105]Ryu S, Brown SL, Kolozsvary A, et al. Noninvasive detection of radiation-induced optic neuropathy by manganese-enhanced MRI[J]. Radiat Res,2002.157(5):500-505.
[106]Thuen M, Singstad TE, Pedersen, et al. Manganese-enhanced MRI of the optic visual pathway and optic nerve injury in adult rats[J]. J Magn Reson Imaging, 2005,22(4):492-500.
[107]Pautler RG, Koretsky AP. Tracing odor-induced activation in the olfactory bulbs of mice using manganese-enhanced magnetic resonance imaging[J]. NeuroImage,2002,16(2):441-448.
[108]Tjalve H, Mejare C, Borg-Neczak K. Uptake and transport of manganese in primary and secondary olfactory neurones in pike. Pharmacol [J]. Toxicol.1995, 77(1):23-31.
[109]Shea TB, Flanagan LA. Kinesin, dynein and neurofilament transport[J]. Trends Neurosci,2001,24(11):644-648.
[110]Bearer EL, Falzone TL, Zhang X, et al. Role of neuronal activity and kinesin on tract tracing by manganese-enhanced MRI (MEMRI) [J]. Neurolmage,2007, 37 (Suppl.1):S37-S46.
[111]Takeda A. Manganese action in brain function[J]. Brain Res Rev,2003, 41(1):79-87.
[112]Leergaard TB, Bjaalie JG, Devor A, et al. In vivo tracing of major rat brain pathways using manganese-enhanced magnetic resonance imaging and three-dimensional digital atlasing[J]. NeuroImage,2003,20(3):1591-1600.
[2]Noble J, Munro CA, Prasad VS, et al. Analysis of upper and lower extremity peripheral nerve injuries in a population of patients with multiple injuries [J]. J Trauma,1998,45(1):116-122.
[3]Winfree CJ. Peripheral nerve injury evaluation and management [J].Curr Surg, 2005,62(5):469-476.
[4]Cudlip SA, Howe FA, Griffi ths JR, et al. Magnetic resonance neurography of peripheral nerve following experimental crush injury, and correlation with functional deficit [J]. J Neurosurg,2002,96 (4):755-759.
[5]Aagaard BD, Lazar DA, Lankerovich L, et al. High-resolution magnetic resonance imaging is a noninvasive method of observing injury and recovery in the peripheral nervous system [J]. Neurosurgery,2003,53 (1):199-203.
[6]Stanisz GJ, Midha R, Munro CA, et al. MR properties of rat sciatic nerve following trauma [J]. Magn Reson Med,2001,45(3):415-420.
[7]Bendszus M, Stoll G. Technology insight:visualizing peripheral nerve injuryusing MRI [J]. Nat Clin Pract Neurol,2005, 1(1):45-53.
[8]Mayer AR, Ling J, Mannell MV, et al. A prospective diffusion tensor imaging study in mild traumatic brain injury [J]. Neurology,2010,74(8):643-50.
[9]Stadlbauer A, Nimsky C, Buslei R, et al. Diffusion tensor imaging and optimized fiber tracking in glioma patients:Histopathologic evaluation of tumor-invaded white matter structures [J]. Neuroimage,2007.,34(3):949-56.
[10]Thomalla G, Glauche V, Koch MA, et al. Diffusion tensor imaging detects early Wallerian degeneration of the pyramidal tract after ischemic stroke[J]. Neuroimage,2004,22(4):1767-74.
[11]Tovar-Moll F, Evangelou IE, Chiu AW, et al. Thalamic involvement and its impact on clinical disability in patients with multiple sclerosis:a diffusion tensor imaging study at 3T [J]. Am J Neuroradiol,2009,30(7):1380-6.
[12]Hiltunen J, Suortti T, Arvela S, et al. Diffusion tensor imaging and tractography of distal peripheral nerves at 3 T [J]. Clin Neurophysiol,2005,116(10):2315-23.
[13]Jambawalikar S, Baum J, Button T, et al. Diffusion tensor imaging of peripheral nerves [J]. Skeletal Radiol,2010,39(11):1073-9.
[14]Lehmann HC, Zhang J, Mori S, et al. Diffusion tensor imaging to assess axonal regeneration in peripheral nerves [J]. Exp Neurol,2010,223(1):238-44.
[15]Takagi T, Nakamura M, Yamada M, et al. Visualization of peripheral nerve degeneration and regeneration:monitoring with diffusion tensor tractography [J]. Neuroimage,2009,44(3):884-92.
[16]Stoll G, Wessig C, Gold R, et al. Assessment of lesion evolution in experimental autoimmune neuritis by gadofluorine M-enhanced MR neurography [J]. Exp Neurol,2006,197(1):150-156.
[17]Wessig C, Jestaedt L, Sereda M, et al. Gadofluorine M-enhanced magnetic resonance nerve imaging:comparison between acute inflammatory and chronic degenerative demyelination in rats [J]. Exp Neurol,210(1):137-143.
[18]Bendszus M, Stoll G. Caught in the act:in vivo mapping of macrophage infiltration in nerve injury by magnetic resonance imaging [J]. J Neurosci,2003, 23(34):10892-10896.
[19]Merritt JE, Jacob R, Hallam TJ. Use of manganese to discriminate between calcium influx and mobilization from internal stores in stimulated human neutrophils[J]. J Biol Chem 1989,264(3):1522-1527.
[20]Murayama Y, Weber B, Saleem KS, et al, Tracing neuralcircuits in vivo with Mn-enhanced MRI [J]. Magn Reson Imaging,2006,24(4):349-358.
[21]Cross DJ, Minoshima S, Anzai Y, et al. Statistical mapping of functional olfactory connections of the rat brain in vivo [J]. Neuroimage,2004,23(4), 1326-1335.
[22]Lee J.W., Park J.A., Lee J.J., et al. Manganese-enhanced auditory tract-tracing MRI with cochlear injection. Magn. Reson. Imaging [J].2007,25(5),652-656.
[23]Lin YJ, Koretsky AP. Manganese ion enhances T1-weighted MRI during brain activation:an approach to direct imaging of brain function [J]. Magn. Reson. Med.1997,38(3):378-388.
[24]Silva AC, Lee JH, Aoki I, et al. Manganese-enhanced magnetic resonance imaging (MEMRI):methodological and practical considerations [J]. NMR Biomed,2004,17(8):532-43.
[25]Einstein A. Investigations on the Theory of the Brownian Movement. New York:Dover; 1956
[26]Schaefer PW, Grant PE, Gonzalez RG. Diffusion-weighted MR imaging of the brain. Radiology [J].2000,217(2):331-345.
[27]Carr HY, Purcell EM. Effects of diffusion on free precession in nuclear magnetic resonance experiments [J]. Phys Rev,1954,94(3):630-638.
[28]Stejskal EO, Tanner JE. Spin diffusion measurements:spin echoes in the presence of a time-dependent field gradient [J]. J Chem Phys,1965,42(1): 288-292.
[29]Le Bihan D. Molecular diffusion nuclear magnetic resonance imaging [J]. Dig Imag Q,1991,7(1):1-30.
[30]Le Bihan D, Mangin JF, Poupon C, et al. Diffusion tensor imaging:concepts and applications [J]. J Magn Reson Imaging,2001,13(4):534-546.
[31]Shimony JS, McKinstry RC, Akbudak E, et al. Quantitative diffusion-tensor anisotropy brain MR imaging:normative human data and anatomic analysis [J]. Radiology 1999,212(3):770-784.
[32]Mori S, Crain BJ, Chacko VP, et al. Three-dimensional tracking of axonal projections in the brain by magnetic resonance imaging [J]. Ann Neurol,1999, 45(2):265-9.
[33]Xue R, van Zijl PC, Crain BJ, et al. In vivo three-dimensional reconstruction of rat brain axonal projections by diffusion tensor imaging [J]. Magn Reson Med, 1999,42(6):1123-7.
[34]Kilickesmez O, Yirik G, Bayramoglu S, et al. Non-breath-hold high b-value diffusion-weighted MRI with parallel imaging technique:apparent diffusion coefficient determination in normal abdominal organs [J]. Diagn Interv Radiol, 2008, (2):83-7.
[35]Pruessmann KP, Weiger M, Scheidegger MB, et al. SENSE:sensitivity encoding for fast MRI [J]. Magn Reson Med,1999,42(5):952-62.
[36]Bammer R, Keeling SL, Augustin M, et al. Improved diffusion-weighted singleshot echo-planar imaging (EPI) in stroke using sensitivity encoding (SENSE) [J]. Magn Reson Med,2001,46(3):548-54.
[37]Bammer R, Auer M, Keeling SL, et al. Diffusion tensor imaging using singleshot SENSE-EPI [J]. Magn Reson Med,2002,48(1):128-36.
[38]Jaermann T, Pruessmann KP, Valavanis A, et al. Influence of SENSE on image properties in high-resolution single-shot echo-planar DTI [J]. Magn Reson Med, 2006,55(2):335-42.
[39]Jones DK, Horsfield MA, Simmons A. Optimal strategies for measuring diffusion in anisotropic systems by magnetic resonance imaging [J]. Magn Reson Med,1999,42(3):515-25
[40]Papadakis NG, Murrills CD, Hall LD, et al. Minimal gradient encoding for robust estimation of diffusion anisotropy. Magn Reson Imaging [J].2000,18(6): 671-79.
[41]Hasan KM, Parker DL, Alexander AL. Comparison of gradient encoding schemes for diffusion-tensor MRI. J Magn Reson Imaging [J].2001,13(5): 769-80.
[42]Tuch DS, Reese TG, Wiegell MR, et al. Diffusion MRI of complex neural architecture [J]. Neuron,2003,40(5):885-95.
[43]Mukherjee P, Chung SW, Berman JI, et al. Diffusion tensor MR imaging and fiber tractography:technical considerations [J]. Am J Neuroradiol,2008, 29(5):843-52.
[44]Huang H, Zhang J, van Zijl PC, et al. Analysis of noise effects on DTI-based tractography using the brute-force and multi-ROI approach [J]. Magn Reson Med,2004,52(3):559-65.
[45]Lazar M, Alexander AL. An error analysis of white matter tractography methods:synthetic diffusion tensor field simulations [J]. Neuroimage, 20(2):1140-53.
[46]Oouchi H, Yamada K, Sakai K, et al. Diffusion anisotropy measurement of brain white matter is affected by voxel size:underestimation occurs in areas with crossing fibers[J]. Am J Neuroradiol,2007,28(6):1102-6.
[47]van Gelderen P, de Vleeschouwer MHM, DesPres D, et al. Water diffusion and acute stroke [J]. Magn Reson Med,1994,31(2):154-163.
[48]Le Bihan D, Breton E, Lallemand D, et al. Separation of diffusion and perfusion in intravoxel incoherent motion MR imaging [J]. Radiology,1988, 168(2):497-505.
[49]Kim HJ, Choi CG, Lee DH, et al. High-b-value diffusion-weighted MR imaging of hyperacute ischemic stroke at 1.5T [J]. Am J Neuroradiol,2005, 26(2):208-215.
[50]Saupe N, White LM, Stainsby J, et al. Diffusion tensor imaging and fiber tractography of skeletal muscle:optimization of B value for imaging at 1.5 T [J]. AJR Am J Roentgenol,2009,192(6):W282-290.
[51]Melhem ER, Itoh R, Jones L, et al. Diffusion tensor MR imaging of the brain: effect of diffusion weighting on trace and anisotropy measurements [J]. Am J Neuroradiol,2000,21(10):1813-20.
[52]Mukherjee P, Chung SW, Berman JI, et al. Diffusion tensor MR imaging and fiber tractography:technical considerations[J]. AJNR Am J Neuroradiol, 2008,29(5):843-52.
[53]李德军,包尚联,马林.不同b值和扩散张量成像导出量的定量关系研究[J].中华放射学杂志,2004,38：118-123.
[54]Yoshiura T, Wu O, Zaheer A, et al. Highly diffusion-sensitized MRI of brain: dissociation of gray and white matter [J]. Magn Reson Med,2001, 45(5):734-740.
[55]Morisaki S, Kawai Y, Umeda M, et al. In vivo assessment of peripheral nerve regeneration by diffusion tensor imaging[J]. Journal of Magnetic Resonance Imaging,2011,33(3):535-542.
[56]Andreisek G, White LM, Kassner A, et al. Diffusion tensor imaging and fiber tractography of the median nerve at 1.5T:optimization of b value[J]. Skeletal Radiol,2009,38(1):51-59.
[57]Schmitz HC, Beer GM. The toe-spreading reflex of the rabbit revisited--functional evaluation of complete peroneal nerve lesions[J]. Lab Anim.2001,35(4):340-345.
[58]Shen J, Zhou CP, Zhong XM, et al.MR neurography:T1 and T2 measurements in acute peripheral nerve traction injury in rabbits[J]. Radiology,2010, 254(3):729-738.
[59]Bridge PM, Ball DJ, Mackinnon SE, et al. Nerve crush injuries--a model for axonotmesis[J]. Exp Neurol,1994,127(2):284-290.
[60]卜令学,李宁毅,杨学财,等.下牙槽神经挤压伤后再生的实验研究[J].现代口腔医学杂志,2002,16(1)：21-23.
[61]Donaldson J, Shi R, Borgens R. Polyet hylene glycol rapidly restores physiological functions in damaged sciatic nerves of guinea pigs[J]. Neurosurgery,2002,50(1):147-157.
[62]Nawwar AM, Sherif MF, Barakat NG. Instrumented forceps for measurement of nerve compression forces[J]. J Biomech Eng,1995,117 (1):53-58.
[63]Rydevik B, Lundborg G. Permeability of intraneural microvessels and perineurium following acute, graded experimental nerve compression[J]. Scand J Plast Reconstr Surg,1977,11(3):179-87.
[64]Dahlin LB, Thambert C. Acute nerve compression at low pressures has a conditioning lesion effect on rat sciatic nerves[J]. Acta Orthop Scand,1993, 64(4):479-81.
[65]Chen L E, Seaber AV, Glisson RR, et al. The functional recovery of peripheral nerves following defined acute crush injuries[J]. J Ort hop Res,1992, 10(5):657-664.
[66]Matsuda K, Wang HX, Suo C, et al.Retrograde axonal tracing using manganese enhanced magnetic resonance imaging[J]. Neuroimage,2010.,50(2):366-74.
[67]李新春,陈健宇,王欣璐,等；兔坐骨神经急性挤压伤的MRI与病理学对比初步研究[J].中华放射学杂志,2004,38(2)：133-138.
[68]Seddon H. Three types of nerve injury. Brain 1943;66:237-88.
[69]Sunderland S:Nerves and Nerve Injuries, ed 2. London:Churchill Livingston, 1978.
[70]Lewis T, Pickering GW, Rothschild P:Centripetal paralysis arising out of arrested bloodflow to the limb[J]. Heart,1931,16:1-32.
[71]Ochoa J, Fowler TJ, Gilliatt RW. Anatomical changes in peripheral nerves compressed by a pneumatic tourniquet[J]. J Anat,1972,113(Pt 3):433-455.
[72]Koeppen AH. Wallerian degeneration:history and clinical significance[J]. J Neurol Sci,2004,220(1-2):115-117.
[73]Waller A:Experiments on the section of the glossopharyngeal and hypoglossal nerves of the frog, and observations of the alterations produced thereby in the structure of their primitive fibers[J]. Phil Trans Roy Soc,1850,140:423-429.
[74]Hall S. Mechanisms of repair after traumatic injury. In:Dyck PJ, Thomas PK, editors. Peripheral neuropathy. Philadelphia:Elsevier, Saunders; 2005. 1403-33.
[75]Kang H, Tian L, Thompson W. Terminal Schwann cells guide the reinnervation of muscle after nerve injury[J]. J Neurocytol 2003,32:975-85.
[76]Burnett MG, Zager EL. Pathophysiology of peripheral nerve injury:a brief review. Neurosurg Focus[J].2004,15; 16(5)
[77]Mackinnon SE, Dellon AL. Surgery of the peripheral nerve. New York:Thieme; 1988.
[78]Bendszus M, Wessig C, Solymosi L, et al. MRI of peripheral nerve degeneration and regeneration:correlation with electrophysiology and histology[J]. Exp Neurol,2004,188(1):171-177.
[79]Sheikh KA. Non-invasive imaging of nerve regeneration [J]. Exp Neurol.2010, 223(l):72-6.
[80]李青峰,谢峰.周围神经损伤中电生理测量技术的应用[J].整形再造外科杂志,2005,2(2)：102-105.
[81]Filler, A., Maravilla, K., Tsuruda, J. MR neurography and muscle MR imaging for image diagnosis of disorders affecting the peripheral nerves and musculature[J]. Neurol. Clin,2004,22(3),643-682.
[82]Lundborg, G.,2004. Nerve Injury and Repair, second ed. Elsevier, Philadelphia.
[83]Budde MD, Kim JH, Liang HF,et al. Axonal injury detected by in vivo diffusion tensor imaging correlates with neurological disability in a mouse model of multiple sclerosis[J]. NMR Biomed,2008,21(6):589-597.
[84]Wang S, Wu EX, Tam CN, et al. Characterization of white matter injury in a hypoxic-ischemic neonatal rat model by diffusion tensor MRI[J]. Stroke, 2008,39(8):2348-2353.
[85]Valsasina P, Rocca MA, Agosta F, et al. Mean diffusivity and fractional anisotropy histogram analysis of the cervical cord in MS patients[J]. Neuroimage,2005,26(3):822-828.
[86]Zhang J, Jones M, DeBoy CA, et al. Diffusion tensor magnetic resonance imaging of Wallerian degeneration in rat spinal cord after dorsal root axotomy[J]. J Neurosci,2009,29(10):3160-3171.
[87]Ford JC, Hackney DB, Alsop DC, et al. MRI characterization of diffusion coefficients in a rat spinal cord injury model [J]. Magn Reson Med 1994;31(5):488-494.
[88]Sakai K, Yamada K, Oouchi H, et al. Numerical simulation model of hyperacute/acute stage white matter infarction[J]. Magn Reson Med Sci 2008, 7(4):187-194.
[89]Cox LJ, Hengst U, Gurskaya NG, et al. Intraaxonal translation and retrograde trafficking of CREB promotes neuronal survival [J]. Nat Cell Biol 2008; 10(2):149-159.
[90]Davies AM. Regulation of neuronal survival and death by extracellular signals during development[J]. EMBO J,2003,22(11):2537-2545.
[91]Markus A, Patel TD, Snider WD. Neurotrophic factors and axonal growth[J]. Curr Opin Neurobiol,2002,12(5):523-531.
[92]Lauterbur PC. Image formation by induced local interactions:examples employing nuclear magnetic resonance [J]. Nature,1973,242:190-191.
[93]Wolf GL, Baum L. Cardiovascular toxicity and tissue proton T1 response to manganese injection in the dog and rabbit[J]. Am. J. Roentgenol,1983,141(1): 193-197.
[94]Narita K, Kawasaki F, Kita H. Mn and Mg influxes through Ca channels of motor nerve terminals are prevented by verapamil in frogs[J]. Brain Res.1990, 510(2),289-295.
[95]Wibeke Nordhoy, Henrik W, Anthonsen,et al. Manganese ions as intracellular contrast agents:proton relaxation and calcium interactions in rat myocardium[J]. NMR Biomed,2003,16(2):82-95.
[96]Hu TC, Pautler RG, MacGowan GA, et al. Manganeseenhanced MRI of mouse heart during changes in inotropy[J]. Magn. Reson. Med,2001,46(5):884-890.
[97]London RE, Toney G, Gabel SA, et al. Magnetic resonanceimaging studies of the brains of anesthetized rats treated withmanganese chloride [J]. Brain Res. Bull,1989,23(3):229-235.
[98]Lucchini R, Albini E, Placidi D, et al. Brain magnetic resonance imaging and manganese exposure[J]. Neurotoxicology,2000,21(5):769-775.
[99]Watanabe T, Natt O, Boretius S, et al. In vivo 3D MRI staining of mouse brain after subcutaneous application of MnCl[J]. Magn.-Reson. Med,2002,48(5): 852-859.
[100]Aoki I, Wu YJ, Silva AC, et al. In vivo detection of neuroarchitecture in the rodent brain using manganese-enhanced MRI[J]. Neuroimage,2004,22(3): 1046-1059.
[101]Pautler RG, Silva AC, Koretsky AP. In vivo neuronal tract tracing using manganese-enhanced magnetic resonance imaging[J]. Magn. Reson. Med,1998, 40(5):740-748.
[102]Allegrini PR, Wiessner C. Three-dimensional MRI of cerebral projections in rat brain in vivo after intracortical injection of MnCl2[J]. NMR Biomed, 2003,16(5):252-256.
[103]Van der Linden A, Verhoye M, Van Meir V, et al. In vivo manganese-enhanced magnetic resonance imaging reveals connections and functional properties of the songbird vocal control system[J]. Neuroscience,2002,112(2):467-474.
[104]Pautler RG, Mongeau R, Jacobs RE. In vivo trans-synaptic tract tracing from the murine striatum and amygdala utilizing manganese enhanced MRI (MEMRI) [J].Magn Reson Med,2003,50(1):33-39.
[105]Ryu S, Brown SL, Kolozsvary A, et al. Noninvasive detection of radiation-induced optic neuropathy by manganese-enhanced MRI[J]. Radiat Res,2002.157(5):500-505.
[106]Thuen M, Singstad TE, Pedersen, et al. Manganese-enhanced MRI of the optic visual pathway and optic nerve injury in adult rats[J]. J Magn Reson Imaging, 2005,22(4):492-500.
[107]Pautler RG, Koretsky AP. Tracing odor-induced activation in the olfactory bulbs of mice using manganese-enhanced magnetic resonance imaging[J]. NeuroImage,2002,16(2):441-448.
[108]Tjalve H, Mejare C, Borg-Neczak K. Uptake and transport of manganese in primary and secondary olfactory neurones in pike. Pharmacol [J]. Toxicol.1995, 77(1):23-31.
[109]Shea TB, Flanagan LA. Kinesin, dynein and neurofilament transport[J]. Trends Neurosci,2001,24(11):644-648.
[110]Bearer EL, Falzone TL, Zhang X, et al. Role of neuronal activity and kinesin on tract tracing by manganese-enhanced MRI (MEMRI) [J]. Neurolmage,2007, 37 (Suppl.1):S37-S46.
[111]Takeda A. Manganese action in brain function[J]. Brain Res Rev,2003, 41(1):79-87.
[112]Leergaard TB, Bjaalie JG, Devor A, et al. In vivo tracing of major rat brain pathways using manganese-enhanced magnetic resonance imaging and three-dimensional digital atlasing[J]. NeuroImage,2003,20(3):1591-1600.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |