【摘要翻译】2007年神经影像学进展

个人翻译 不当之处请指教！

Advances in Imaging 2007
Imaging continues to be a mainstay of stroke diagnosis, management, and investigation, and this was fully evident again in 2007. Stroke continues to command a substantial amount of attention from investigators from a wide variety of disciplines. A broad range of insights and innovations in imaging have occurred in the past year, and the space allotted here remains too small to capture the range of efforts. In this survey we will discuss the ongoing advances in optimal imaging of the acute stroke patient, the role of imaging as a biomarker or surrogate end point, as well as efforts to use imaging to assess the risk of hemorrhage after chemical thrombolysis. Important data about the frequency of incidental stroke have been published. And, in a special focus, we will describe the role imaging has played this past year in building understanding of recovery from stroke.
One ongoing area of investigation regards the manner in which to optimally image the acute stroke patient. Guidelines issued this year 1 provide an excellent summary: x-ray computed tomography (CT) guides most treatment decisions, and therapy should not be delayed to obtain multimodal imaging studies. However, multimodal imaging with CT and with MRI may provide additional information that will improve the diagnosis of acute ischemic stroke. Indeed, recent data suggest that CT angiography and perfusion assessment with CT is both feasible and in the specific setting of hemispheric stroke can provide assessment of vessel status such as site of occlusion, tissue damaged and potentially at risk, and other features with good accuracy.2 Given the widespread availability of CT in many hospitals around the world, CT is highly likely to remain a mainstay of diagnosis, and in settings where, as the guidelines state, therapy is not delayed to obtain multimodal imaging studies, such additional imaging may well improve diagnosis and in turn treatment.
Before considering multimodal imaging, it may be valuable to review new data evaluating diagnostic methods. Data continue to accumulate highlighting the advanced diagnostic power of MRI, especially diffusion-weighted MRI, over CT, with one major prospective study reporting a sensitivity of 83% for MRI relative to the final clinical diagnosis of ischemic stroke compared with 26% for CT.3 This population was among a broad range of patients presenting with possible acute ischemic stroke; certainly in select populations such as patients with large stroke, CT has much higher sensitivity than 26%. Nevertheless, most users of both CT and MRI can confirm from experience the greater reliability of MRI for diagnosis, and there appears to be a trend toward MRI for the initial diagnosis of ischemic stroke whenever possible.4 This increased power is particularly evident in the management of transient ischemic attack patients, as was shown again this year.5,6 Given the widespread prevalence of transient ischemic attacks and guidelines that already suggest including imaging at some point,7,8 these new results suggest diffusion-weighted imaging evaluation should be done urgently after transient ischemic attack. Of course, diagnosis with any imaging modality is better than no imaging, and in the face of a lack of ability to do MRI for any number of reasons (ranging from the absence of availability of emergency MRI to the presence of a pacemaker), certainly CT remains highly capable.
Does this improved diagnostic ability have any impact on patients? The answer from data published this past year appears to be yes. Increasingly, investigators are beginning to use multimodal imaging to attempt to widen the therapeutic window, using certain imaging findings such as the presence of an imaging correlate to the ischemic penumbra to consider treating with thrombolysis outside the conventional 3-hour window. Data comparing typical CT to multimodal MR suggest that MRI methods can be used to safely and effectively identify patients in whom thrombolysis might be safely administered to after the typical 3-hour window.9 Comparative data suggest that perfusion assessment with CT might be used similarly as MRI for treatment decisions in certain select patients.10 There are not yet data directly comparing multimodal CT to multimodal MR or showing that multimodal CT can be used to extend the window of thrombolysis. Nevertheless, it stands to reason that more information is good and whether such information comes from CT, MR, ultrasound, or any other modality, reliable information is the key to improved diagnosis and then treatment. However, the lack of success of the DIAS-II trial has led some to question whether the diffusion-perfusion mismatch is truly a surrogate for patients who might benefit. This remains to be seen; the release of the formal results may shed light on this question. Although some have argued that the diffusion-perfusion mismatch on MRI might be matched by a clinical-diffusion mismatch, 2 studies this year demonstrate that this is not the case.11,12
Multimodal MRI has also been used in the setting of thrombolysis to attempt to predict which patients might experience intraparenchymal hemorrhage. Data suggest that small T2* lesions, thought of as cerebral microbleeds, do not appear to have any increased incidence of hemorrhage after chemical thrombolysis,13 but the presence of a large diffusion lesion 14 or high permeability 15 is indicative of increased risk. These data are consistent with earlier results suggesting that large lesions do not do well with thrombolysis.16
One cautionary note around imaging: there are new risks. Recent estimates of the effects of radiation from CT scanning have suggested that a single CT scan of the head in a 65-year-old is associated with an 0.04% lifetime risk of cancer 17 (the risks are higher for younger patients). Given that a CT perfusion scan can have as much as 50 times the dose of a standard head CT,18,19 and that CT angiography adds further ionizing radiation dosing, the long-term effects of these imaging tests and especially repeat scans must be carefully considered. MRI is also associated with recently identified risks, specifically when gadolinium is administered to patients with very poor renal function.20 Fortunately, dose-lowering approaches are under development for CT, and noncontrast perfusion methods for MRI. While these risks must be considered in the context of the potential severity of acute ischemic stroke, there is clearly evidence to warrant a more cautious attitude.
We now turn to the role of imaging in assessing recovery, perhaps one of the most important areas of stroke investigation, and one receiving increasing attention. Functional neuroimaging has been applied quite frequently for the demonstration of brain areas involved in recovery of function after a stroke and even for the selection of patients for specific rehabilitation strategies. In a series of reviews the various imaging technologies for noninvasive exploration of the reorganization in cerebral networks relevant for recovery were described, but the reported results are still quite controversial.21–23 As clearly stated by Ward,21 a functionally relevant “reorganization can only occur in structurally and functionally intact brain regions”, and therefore is dependent on the location of the primary lesion. In several studies the functional reorganization of the motor network could be documented in patients after capsular stroke and subcortical stroke: good recovery was related to enhanced recruitment of the lateral premotor cortex of the lesional hemisphere and lateral premotor and to a lesser extent primary sensorimotor and parietal cortex of the contralateral hemisphere.24 The mechanisms of motor recovery vary according to location of the lesion: cortical infarcts activate the contralateral primary sensorimotor cortex, whereas subcortical infarcts largely showed activation of bilateral primary sensorimotor cortex.25 While there was an enhancement of the activation in the contralesional cortical network with motor skill challenge,26 several studies indicated that worse motor performance was related to a greater amount of contralesional activation 27 and that patients who activated the ipsilesional primary motor cortex early had a better recovery of hand function.28 Repetitive peripheral magnetic stimulation increases the activation of the parieto-premotor network and thereby might have a positive conditioning effect for treatment.29 In addition to changes in the activation pattern of the motor network, different activation patterns were observed in the proprioceptive system, where the initially observed blood flow increases in SI and SII of the noninfarcted hemisphere vanished during successful rehabilitation and the normal activation patterns were restored, indicating an interhemispheric shift of attention associated with recovery.30
Very few studies discussed results from functional imaging in recovery from poststroke aphasia and their consequence for treatment strategies.31 These studies support the model of 3 phases of language recovery: a strongly reduced activation of remaining left language areas in the acute phase, followed (or substituted) by an upregulation of homologue language zones and finally a normalization of the activation pattern reflecting consolidation in the language systems.32 The pattern of activation and the recruitment of regions during the rehabilitation depend on the available language-related regions, where restoration of the left hemisphere networks seems to be more effective, although in some cases right hemisphere areas are integrated successfully.33 It remains to be shown in future studies whether the inhibition or facilitation of selected areas, eg, by repetitive transcranial magnetic stimulation,34 can improve recovery of poststroke aphasia.
These efforts at understanding stroke and especially stroke recovery have been made even more relevant by a study of 2000 subjects over age 45 demonstrating a 7% incidence of incidental brain infarcts.35 The true burden of disease of stroke has yet to be fully understood, and imaging continues to play an important role in its diagnosis and treatment.

2007年影像学进展
影像仍然是卒中诊断和管理、调查的主要手段，这在2007年再次体现出来。卒中仍然吸引着全世界各个学科研究者的注意。在过去的一年中，一系列的见解和革新都发生在影像学中，这里所给的篇幅显然不足够表达这些成果。在这一研究中，我们将会讨论急性卒中病人最佳影像的持续进展，除了使用影像评估化学溶栓后出血的风险外，影像还作为一种生物标志或者终点的代用品。卒中事件发生的频率的资料已经发表。在某些点上，我们将会叙述影像在过去一年卒中患者恢复中扮演的角色。
如何优化急性卒中患者的医学影像是目前正在深入研究的方向之一。这一年的有关影像学方面的指南给出了精彩的概括：CT检查结果指导大多数治疗决策的制定，但不能因影像检查而推迟治疗的开始，然而，多样的CT和MRI可能为急性卒中的诊断和治疗提供更多的信息。事实上，最近的数据表明：急性卒中患者实施CTA和CT灌注成像检查具备可行性，其能够评估半球梗死患者的血管异常状态，例如血管闭塞的部位、组织损伤和潜在的治疗风险，且准确率高。随着CT在全世界各大医院的广泛应用，CT仍然很有可能是诊断的主要手段。在某些州制定的指南规定，不能为了取得多形的影像研究而耽误治疗，尽管这些额外的影像可以改善诊断和治疗。
在考虑多形影像学检查之前，回顾资料对于提高诊断方法是很有价值的。累积的资料显示，MRI在卒中的诊断方面的优越性超过了CT，特别是MRI灌注加权成像（即DWI）。一项前瞻性的研究表明，在缺血性卒中的最终诊断方面，MRI的敏感性为83％，而CT为26％。这一个研究的病例包括了大部分表现可能为急性缺血性卒中的患者。当然，选择有大面积的脑卒中的病人，CT的敏感性将会比26％高的多。大多数同时应用CT和MRI进行急性卒中诊断的医生承认：从经验来看，MRI诊断的可信度比较高。因此，对于急性缺血性卒中的病人都尽可能使用MRI检查，这是一种趋势。MRI对于卒中诊断价值的提高表现在今年有关MRI检查于TIA处理的相关研究中。考虑到短暂性脑缺血发作存在的广泛性，今年的指南就已经明确建议MRI要作为TIA病人首要检查项目，新的研究建议：急性短暂性脑缺血发作的病人发作之后尽快做DWI检查和评估。当然，任何形式的影像诊断都比没有影像诊断好，由于种种原因（从急诊做MRI不适宜到存在有起搏器的病人）不能行急诊MRI检查的情况下，CT仍然是非常有可行的。
这些影像学检查诊断能力提高对病人有什么影响吗？答案是肯定的，从这一年各方面研究的资料可以看出来。目前越来越多的研究者开始利用多元化的影像检查以帮助延长治疗的时间窗，利用某些影像的发现，诸如与缺血半暗带相关的影像，探讨在常规的3小时治疗时间窗外溶栓治疗。通过比较传统的CT和多元化的MRI可知：在传统的3小时治疗时间窗外实施溶栓治疗，病人的安全性评估方面MRI更加安全而有效。CT灌注检查可能在特定病人治疗决策的制定上起到与MRI类似的作用。目前尚无直接比较多元化CT检查和多元化MRI检查的研究数据，也没有多元化CT用于延长溶栓治疗时间窗的研究数据。但总的来说，获得的信息越多越好，无论这些信息是来自于CT、MRI、超声，还是来源与其他检查方式，可靠的信息是改善诊断和治疗的关键。然而，缺少成功的DIAS-II试验而导致大家提出质疑：DWI和PWI图像差异是否提示病人可能从溶栓治疗中获益。这一个问题仍然值得探讨。尽管有些人提出MRI上DWI和PWI的差异可能与临床和DWI的差异相吻合，但今年的研究却得出相反的结果。
多元化的MRI检查被用作溶栓治疗的评估，多元化得MRI被用作试图预测哪些病人可能出现脑实质内出血。数据表明，T2加权像上的微小病灶，被认为是大脑的微小出血，不能提示化学溶栓治疗后出血风险的增加，但如果存在一个较大的弥撒损害病灶或高渗透性预示出血风险的增加。这些研究数据与早期的研究数据均提示大病灶不适宜进行化学溶栓。
有关影像学方面需要注意的是：有新的风险存在。据最有关CT扫描辐射的研究指出，在一个65岁的老人,接受单次的头颅CT扫描伴随患0.04%癌症的发生风险（这些风险在年轻患者中更高）。鉴于CT灌注扫描接受放射剂量是常规头颅CT标准剂量的50倍以及CTA额外增加的电离辐射的剂量，这些影像检查带来的长期的副作用来看，特别是重复扫描，必须仔细考虑。MRI的检查同样伴随着危险：特别是当钆增强剂应用于肾功能较差的病人时。幸运的是，随着CT的发展，人们接受放射剂量越来越少，磁共振成像出现了非增强的灌注方法。在严重性急性缺血性中风病人中，这些风险必须同时考虑，现在有明确的证据说明要以更加审慎的态度来对待。
我们现在转到另外一个话题，有关影像在脑卒中恢复中的评估作用，也许这可能是脑卒中研究最重要的领域之一，这方面的研究越来越受到重视。功能性神经影像学已应用得相当广泛，比如在脑卒中后局部脑功能恢复的评估，甚至是在选择病人进行特殊的康复治疗方面。在一系列的综述性的文章中，各种成像技术为无创性的探索与康复相关脑功能网状结构进行了描述，但报告的结果仍然是相当有争议的。Ward指出，功能相关的脑重构只能发生在结构和功能完好的大脑区域发生，因此是依赖于原发灶的部位。几项研究中报道内囊和皮层下卒中患者运动网状组织的重构：良好的恢复，是与病变半球的外侧运动前区皮层和外侧运动前区的复原增强、并在较小的程度上与原始感觉运动区和对侧半球顶叶皮质的延伸有关.运动的康复程度依照病灶部位而不同的：皮质梗死激活对侧原始运动感觉的皮层，而皮层下梗死在很大程度上表明，激活了是双侧的原始运动感觉皮质.有一项激活对侧的皮质网网状结构可以提高运动技巧，有几项研究表明，病人动作功能加重与病变对侧的活化过度有关，较早的激活健侧初级运动皮层的病人手功能有较好的恢复。除了运动网状结构激活的方式改变外，在本体感觉系统中不同的激活方式的不同。经过成功的康复治疗后，非梗塞侧半球SI和SII脑血流增加的消失以及正常激活模式的重新建立，提示半球间转移注意力与康复有关。
卒中后失语恢复的功能影像以及影像与治疗决策间的关系，这方面的研究比较少.这些研究支持的语言康复的3个阶段：在急性期，残余左侧语言功能区激活程度明显降低;接着,上调同源语言区;最终,实现激活模式的正常化反映出语言系统的重新建立.在康复过程中出现的激活模式以及区域的复原很大程度上依赖于残余的语言相关的地区，虽然在某些情况下，右半球半球能够成功的整合，但是，左侧半球的网络结构的恢复可能更有效。在未来的研究是否可以通过抑制或促进某些领域，如通过重复经颅磁刺激，改善卒中后失语的恢复，这将成为未来的研究课题。
一项有2000例45岁以上的人群研究，脑梗塞的发生率为7％。这个事实表明以上的研究对于卒中和卒中恢复的重要意义。对于卒中是公众的负担仍然未得到充分的理解。卒中的影像学在卒中的诊断和治疗方面继续发挥着重要作用。