丘脑底核磁共振成像技术研究进展

孔维诗， 路明宽， 阳青松， 陈智昌， 仇一青， 吴 曦*

1.海军军医大学学员旅，上海 200433 2.海军军医大学附属长海医院影像科， 上海 200433 3.海军军医大学海医系，上海 200433 4.海军军医大学附属长海医院神经外科，上海 200433

丘脑底核磁共振成像技术研究进展

孔维诗1△， 路明宽1△， 阳青松2， 陈智昌3， 仇一青4， 吴 曦4*

1.海军军医大学学员旅，上海 200433 2.海军军医大学附属长海医院影像科， 上海 200433 3.海军军医大学海医系，上海 200433 4.海军军医大学附属长海医院神经外科，上海 200433

丘脑底核(subthalamic nucleus, STN)是我国帕金森病患者接受脑深部电刺激(deep brain stimulation, DBS)治疗的主要核团。磁共振(MRI)影像个体化、精确植入电极至STN的感觉运动部要求MRI成像对STN边界清晰划分，并确保图像保真。目前使用的MRI序列可分为3大类：基于自旋回波序列的T2加权成像、反转恢复序列、扩散张量成像、各向异性分数；基于磁化转移技术的磁敏感加权成像、自由衰减的T2*成像等；基于磁敏感图像重建技术的定量磁敏感图谱。其中，定量磁敏感图谱的STN与周边的信噪比最优、边界最清晰，T2*技术次之；T2加权成像在患者戴框架时几何精度较高，适合戴框架直接定位使用。

丘脑底核; 磁共振成像; 脑深部电刺激; 帕金森病

丘脑底核(subthalamic nucleus, STN)是我国帕金森病患者接受脑深部电刺激(deep brain stimulation, DBS)采用最多的核团[1]。以往对STN的定位是利用脑解剖结构(前联合与后联合)的经验性定位。随着MRI的发展，其逐渐被应用于STN定位，包括戴框架的直接定位和与CT影像融合定位。为了能够个体化、精确地将电极植入至STN的感觉运动部(背外侧区域)，要求在MRI图像上清晰划分核团边界并保证图像的最小几何失真，但目前效果欠理想[2]，原因如下：(1)STN体积较小，呈倾斜的凸透镜形；(2)低场强的MRI下STN边界不够清晰，STN和黑质(substantia nigra, SN)、未定带(zona incerta, ZI)之间边界不清晰[3]；(3)传统的立体定向头部框架和指示器较大，佩戴后无法使用高质量的多通道MRI头部线圈，影响了MRI下直接定位的图像质量。因此，国内大部分临床中心对患者采用术前MRI与戴头架时CT的融合定位，有时辅以术中电生理监测以纠正电极植入误差[1]。但是，如果术前MRI图像存在失真和图像辨识不清，也会影响融合后的靶点计算结果。因此，需要对不同MRI序列对STN的显影效果进行研究。

目前采用的MRI序列可以分为3大类：基于自旋回波序列的T2加权成像(T2-weighted imaging, T2WI)反转恢复序列(inversion recovery, IR)、扩散张量成像(diffusion tensor imaging, DTI)；基于磁化转移技术的磁敏感加权成像(susceptibility weighted imaging, SWI)、自由衰减的T2*成像(T2-weighted magnitude imaging,T2*WI)、 磁敏感加权相位成像(susceptibility-weighted phase imaging, SWPI)；基于图像重建技术的定量磁敏感图谱(quantitative susceptibility mapping, QSM)。本文对此进行总结，以期为神经外科医师进一步了解目前临床应用技术条件下扫描参数的优缺点并做出合理选择提供参考。

1 基于自旋回波技术的序列

1.1 T2WI序列 T2WI是STN MRI研究中涉及最多的序列。在此序列下，STN较周围组织显示出更低的信号[4]。但在区分STN与其周围组织交界时，容易出现图像失真且分辨不清的情况，尤其是在低场强(1.5 T)下更加明显。根据T2WI对STN直接定位虽然比经验性解剖定位更准确[2]，但是图像失真和分辨率较低可能使定位出现偏差[5-6]。当患者佩戴立体定位框架行T2WI扫描时，失真程度可能更大[7]。术中实时MRI虽然也使用的是低场强的T2WI，但是电极植入时，可以通过多次扫描获得电极植入过程的动态位置，从而调整电极图像与STN的相对关系，减少图像失真引起的误差以及脑脊液减少造成的脑移位[8-9]。

更大的MRI场强允许增加T1重复时间(repetition time, TR)并减小T2WI的TR，这可以改善图像分辨率和(或)缩短成像时间[10]，从而提高信噪比(signal-to-noise ratio, SNR)[11]。因此，3.0 T下T2WI可更好地显示STN在矢状位、冠状位、水平位的边界[12]。虽然7.0 T和9.4 T较3.0 T更能提高SNR[13-14]，但是更大的场强下，影像的几何失真可能更严重，甚至抵消SNR改善带来的获益。虽然有研究[15]显示，9.4 T下STN影像边界与病理的组织学边界吻合，但是该研究样本量太少，结论仍需要进一步评估。

计算机可自动辅助界定STN边界并重建核团图像[16]，但仍需以影像学数据为基础。研究[17]用快速自旋回波 (fast spin echo, FSE)T2WI重建STN ，术中电生理检查显示电极位置良好，但是该研究没有测量电生理边界与影像学边界的相关性。因此，基于FSE-T2WI的STN计算机重建图像的准确性仍需要评估。

1.2 DTI序列 DTI常用于白质纤维束的检查。在DBS术中，其FA参数可以帮助医师识别STN周围的传导束[18]。通过计算电极触点与白质纤维的距离[19]，调整电极植入轨迹，从而尽可能避免激活邻近白质[20]。而目前在常规手术中，对于局麻清醒或术中唤醒的患者仔细进行宏电极刺激测试，即可以避免电刺激不良反应的发生。因此，使用DTI的获益有待进一步评估。

1.3 IR序列 IR序列应用范围较广，其通过选择性地抑制组织中的特定成分增强SNR，常用于STN定位。IR常用序列包括液体衰减反转恢复序列(fluid attenuated inversion recovery, FLAIR)、短反转恢复时间的反转恢复序列(short tau inversion recovery, STIR)、相位敏感反转恢复序列(phase-sensitive inversion recovery, PSIR)和快速灰质采集T1反转恢复序列(fast gray matter acquisition T1inversion recovery, FGATIR)等。

FLAIR显示STN的优点是使其边界更清晰、扫描时间快，但是STN与SN边界仍难以分清。有研究[21]显示，3D FLAIR较2D 快速场回波(fast-field echo， FFE)-T2*WI、快速自旋回波序列(turbo spin-echo, TSE)-T2WI有更好的对比信噪比(contrast-to noise ratio, CNR)，可以更好地辨识STN的边界。

PSIR显示STN时有良好的SNR和CNR[22]，经过Leksell G立体定位框架验证，仅有小于1%的几何失真[23]。4 000 ms的TR和200 ms的反转时间(TI)可提高STN显像效果[23]。但与T2WI、T2*WI、SWI比较，PSIR显示STN的CNR明显较差[24]。尽管如此，由于SWI和T2*WI缺乏几何精确度相关的数据支持，习惯用MRI直接定位而不是CT/MRI融合技术的医师更倾向于用PSIR[25]。

T1STIR序列与FSE-T2WI结合有助于识别STN的所有边界[26]。STIR在STN与SN边界有更高的CNR，而FSE-T2WI对STN与内囊、ZI的外侧界和上界的区分更加清晰。然而，上述研究是基于健康人群开展的，且STN随着患者年龄增加在STIR上可能逐渐难以辨识[27]，因此上述结论还需要进一步的证据支持。

与T1WI和T2WI FLAIR序列相比，FGATIR显示更优的CNR[28]，但是仍需要进一步研究验证其几何精确性。

2 基于磁敏感相关序列

SWI以T2*三维、速度补偿梯度回波序列作为基础[29]，根据不同组织间的磁敏感性差异使图像对比增强，可同时获得磁距图像和相位图像(SWPI)。T2*WI不是一个序列而是一类序列，大多为梯度回波序列(gradient recalled echo, GRE)，包括快速小角度激发(fast low-angle shot, FLASH)重聚焦梯度回波序列GRE(post-excitation refocused GRE)等。随着患者年龄增长，尤其是发生神经退行性疾病时，STN的铁含量增加[4,30]。T2*WI利用组织之间的磁化率差异，可以实现STN与周围结构之间的良好对比，因此较传统T2WI对暗边界的显示更明显。

T2*FLASH因为扫描时间短、图像SNR高，被广泛使用和研究。3.0 T和7.0 T下，T2*FLASH较SWI能更好地显示冠状位时STN与ZI、SN的分界[31-32]。T1-T2*WI融合也可更清晰地显示STN边界[33]。但是上述研究没有测试几何精度，因此对STN定位情况还需要进一步研究。

SWPI不依赖于T1和T2弛豫参数，并且较少受到场强不均匀的影响，这是其相对于各种T2*WI的主要优势[34]。但是，SWPI图像在大多数情况下无法与CT图像良好融合，所以其应用受限。

SWI是相位和幅值图像的组合，其对出血、钙化、铁沉积和含有脱氧血红蛋白的慢速静脉血敏感[35]，具有从SN和ZI中显示STN边界的能力[36]，其描绘的边界与电生理记录描记的边界高度吻合[37]。SWI即使在低场强下仍较T2WI有更好的SNR[24,38]。此外，SWI可显示深部脑静脉及跨实质血管，有助于术前制定穿刺轨迹以避开这些血管[39-40]。越来越多的中心选择SWI作为CT/MRI融合定位的序列。

磁敏感相关序列也有一些缺点，包括信号损失、失真和局部场强不均匀[35]。特别是在高场强下，非局部磁敏性效应(造成晕状伪影)可能导致STN的边缘模糊[41]。由于STN倾斜于3个平面中且结构较小，这些晕状伪影需要在精确定位STN之前进行量化和校正。

3 基于图像后处理技术

QSM是可以在SWI序列的GRE相位图像上使用的图像后处理技术，可解决晕状伪影问题[41]。QSM成像的灰度与脑组织中铁浓度的线性关系较SWI有更多的阶梯梯度[41]，对组织铁浓度的估算更准确，因此可以更好地辨识STN与周围富含铁的结构边界[35,41-44]。QSM在STN定位中的应用可能有很好的前景。

4 总结及展望

清晰度高、对比度高、无几何失真的STN及其周边结构的MRI图像对于功能神经外科医师对STN解剖定位来说至关重要。随着技术进步以及多种基于磁敏感成像的新序列出现，MRI对STN定位的误差越来越小。然而，目前使用的序列仍然存在两个主要问题。(1)几何失真：使用MRI和CT融合技术可以减少MRI技术中的几何误差，但是融合过程可能增加了几何误差[45]。虽然有失真校正算法，但直接生成无失真的图像显然更理想。(2)图像质量差：虽然可通过在全麻下延长扫描采集时间改善CNR，但顺磁性头架的头部线圈的限制仍未改善。

对上述序列下STN的成像效果的比较研究显示：目前在基于自旋回波技术的序列中，FSE-T2WI的成像效果最佳[24]；而基于磁敏感的SWI、T2*WI明显优于FSE-T2WI[31]。基于磁敏感的序列中，SWI比T2*FLAIR有更好的CNR[31]；SWPI对STN的SNR是SWI的2倍、T2*WI的3.5倍[36]，仅次于QSM[38]。QSM的CNR优于T2、FSE-T2WI、T2*WI、SWPI和SWI[38]，可清晰显示STN在ZI上方和SN下方的部分，且能减少由GRE序列产生的晕状伪影。因此，基于磁敏感的序列和影像重建序列在STN显像中可能替代T2WI。

[ 1 ] 张建国. 功能神经外科发展十年[J]. 中国现代神经疾病杂志,2010,10(1):117-122.

[ 2 ] ASHKAN K, BLOMSTEDT P, ZRINZO L, et al. Variability of the subthalamic nucleus: the case for direct MRI guided targeting[J]. Br J Neurosurg, 2007, 21(2):197-200.

[ 3 ] PATIL P G, CONRAD E C, ALDRIDGE J W, et al. The anatomical and electrophysiological subthalamic nucleus visualized by 3-T magnetic resonance imaging[J]. Neurosurgery, 2017, 71(6):1089-1095.

[ 4 ] DORMONT D, RICCIARDI K G, TANDÉ D, et al. Is the subthalamic nucleus hypointense on T2-weighted images? A correlation study using MR imaging and stereotactic atlas data[J]. AJNR Am J Neuroradiol, 2004, 25(9):1516-1523.

[ 5 ] ZONENSHAYN M, REZAI A R, MOGILNER A Y, et al. Comparison of anatomic and neurophysiological methods for subthalamic nucleus targeting[J]. Neurosurgery, 2000, 47(2):282-292.

[ 6 ] ANDRADE-SOUZA Y M, SCHWALB J M, HAMANI C, et al. Comparison of three methods of targeting the subthalamic nucleus for chronic stimulation in Parkinson’s disease[J]. Neurosurgery, 2005, 56 (2 Suppl):360-368.

[ 7 ] SIMON S L, DOUGLAS P, BALTUCH G H, et al. Error analysis of MRI and Leksell stereotactic frame target localization in deep brain stimulation surgery[J]. Stereotact Funct Neurosurg, 2005, 83(1):1-5.

[ 8 ] CHABARDES S, ISNARD S, CASTRIOTO A, et al. Surgical implantation of STN-DBS leads using intraoperative MRI guidance: technique, accuracy, and clinical benefit at 1-year follow-up[J]. Acta Neurochir (Wien), 2015, 157(4):729-737.

[ 9 ] CUI Z, PAN L, SONG H, et al. Intraoperative MRI for optimizing electrode placement for deep brain stimulation of the subthalamic nucleus in Parkinson disease[J].J Neurosurg, 2016, 124(1):62-69.

[10] LU H, NAGAE-POETSCHER L M, GOLAY X, et al. Routine clinical brain MRI sequences for use at 3.0 Tesla[J]. J Magn Reson Imaging, 2005, 22(1):13-22.

[11] CHENG C H, HUANG H M, LIN H L, et al. 1.5T versus 3T MRI for targeting subthalamic nucleus for deep brain stimulation[J].Br J Neurosurg, 2014, 28(4):467-470.

[12] MALLET L, SCHÜPBACH M, N’DIAYE K, et al. Stimulation of subterritories of the subthalamic nucleus reveals its role in the integration of the emotional and motor aspects of behavior[J]. Proc Natl Acad Sci U S A, 2007, 104(25):10661-10666.

[13] MASSEY L A, MIRANDA M A, ZRINZO L, et al. High resolution MR anatomy of the subthalamic nucleus: imaging at 9.4 T with histological validation[J].Neuroimage, 2012,59(3):2035-2044.

[14] ABOSCH A, YACOUB E, UGURBIL K, et al. An assessment of current brain targets for deep brain stimulation surgery with susceptibility-weighted imaging at 7 tesla[J]. Neurosurgery, 2010, 67(6):1745-1756.

[15] AL-HELLI O, THOMAS D L, MASSEY L, et al. Brain stimulation of the subthalamic nucleus: histological verification and 9.4-T MRI correlation[J]. Acta Neurochir (Wien), 2015, 157(12):2143-2147.

[16] XIAO Y, JANNIN P, D’ALBIS T, et al. Investigation of morphometric variability of subthalamic nucleus, red nucleus, and substantia nigra in advanced Parkinson’s disease patients using automatic segmentation and PCA-based analysis[J]. Hum Brain Mapp, 2014, 35(9):4330-4344.

[17] LI B, JIANG C, LI L, et al. Automated segmentation and reconstruction of the subthalamic nucleus in Parkinson’s disease patients[J]. Neuromodulation, 2016, 19(1):13-19.

[18] BARKHOUDARIAN G, KLOCHKOV T, SEDRAK M, et al. A role of diffusion tensor imaging in movement disorder surgery[J]. Acta Neurochir (Wien), 2010, 152(12):2089-2095.

[19] SAID N, ELIAS W J, RAGHAVAN P, et al. Correlation of diffusion tensor tractography and intraoperative macrostimulation during deep brain stimulation for Parkinson disease[J]. J Neurosurg, 2014, 121(4):929-935.

[20] SHIELDS D C, GORGULHO A, BEHNKE E, et al. Contralateral conjugate eye deviation during deep brain stimulation of the subthalamic nucleus[J]. J Neurosurg, 2007, 107(1):37-42.

[21] HEO Y J, KIM S J, KIM H S, et al. Three-dimensional fluid-attenuated inversion recovery sequence for visualisation of subthalamic nucleus for deep brain stimulation in Parkinson’s disease[J]. Neuroradiology, 2015, 57(9):929-935.

[22] STARR P A, VITEK J L, DELONG M, et al. Magnetic resonance imaging-based stereotactic localization of the globus pallidus and subthalamic nucleus[J]. Neurosurgery, 1999, 44(2):303-313,discussion 313-314.

[23] ISHIMORI T, NAKANO S, MORI Y, et al. Preoperative identification of subthalamic nucleus for deep brain stimulation using three-dimensional phase sensitive inversion recovery technique[J]. Magn Reson Med Sci, 2007, 6(4):225-229.

[24] O’GORMAN R L, SHMUELI K, ASHKAN K, et al. Optimal MRI methods for direct stereotactic targeting of the subthalamic nucleus and globus pallidus[J]. Eur Radiol, 2011, 21(1):130-136.

[25] THANI N B, BALA A, LIND C R. Accuracy of magnetic resonance imaging-directed frame-based stereotaxis[J]. Neurosurgery, 2012, 70 (1 Suppl Operative):114-123.

[26] KITAJIMA M, KOROGI Y, KAKEDA S, et al. Human subthalamic nucleus: evaluation with high-resolution MR imaging at 3.0 T[J]. Neuroradiology, 2008, 50(8):675-681.

[27] SARKAR S N, SARKAR P R, PAPAVASSILIOU E. Subthalamic nuclear tissue contrast in inversion recovery MRI decreases with age in medically refractory Parkinson’s disease[J]. J Neuroimaging, 2015, 25(2):303-306.

[28] SUDHYADHOM A, HAQ I U, FOOTE K D, et al. A high resolution and high contrast MRI for differentiation of subcortical structures for DBS targeting: the Fast Gray Matter Acquisition T1Inversion Recovery (FGATIR)[J]. Neuroimage, 2009, 47 Suppl 2:T44-T52.

[29] HAACKE E M, XU Y, CHENG Y C, et al. Susceptibility weighted imaging (SWI) [J]. Magn Reson Med, 2004, 52(3):612-618.

[30] BRASS S D, CHEN N K, MULKERN R V, et al. Magnetic resonance imaging of iron deposition in neurological disorders[J]. Top Magn Reson Imaging, 2006, 17(1):31-40.

[31] KERL H U, GERIGK L, PECHLIVANIS I, et al. The subthalamic nucleus at 3.0 Tesla: choice of optimal sequence and orientation for deep brain stimulation using a standard installation protocol: clinical article[J]. J Neurosurg, 2012, 117(6):1155-1165.

[32] KERL H U, GERIGK L, PECHLIVANIS I, et al. The subthalamic nucleus at 7.0 Tesla: evaluation of sequence and orientation for deep-brain stimulation[J]. Acta Neurochir (Wien), 2012, 154(11):2051-2062.

[33] XIAO Y, FONOV V, BÉRIAULT S, et al. Multi-contrast unbiased MRI atlas of a Parkinson’s disease population[J]. Int J Comput Assist Radiol Surg, 2015, 10(3):329-341.

[34] SHMUELI K, DE ZWART J A, van GELDEREN P, et al. Magnetic susceptibility mapping of brain tissueinvivousing MRI phase data[J]. Magn Reson Med, 2009, 62(6):1510-1522.

[35] GASPAROTTI R, PINELLI L, LISERRE R. New MR sequences in daily practice: susceptibility weighted imaging. A pictorial essay[J]. Insights Imaging, 2011, 2(3):335-347.

[36] VERTINSKY A T, COENEN V A, LANG D J, et al. Localization of the subthalamic nucleus: optimization with susceptibility-weighted phase MR imaging[J]. AJNR Am J Neuroradiol, 2009, 30(9):1717-1724.

[37] MCEVOY J, UGHRATDAR I, SCHWARZ S, et al. Electrophysiological validation of STN-SNR boundary depicted by susceptibility-weighted MRI[J].Acta Neurochir (Wien), 2015, 157(12):2129-2134.

[38] LIU T, ESKREIS-WINKLER S, SCHWEITZER A D, et al. Improved subthalamic nucleus depiction with quantitative susceptibility mapping[J].Radiology, 2013, 269(1):216-223.

[39] RAUSCHER A, SEDLACIK J, BARTH M, et al. Nonnvasive assessment of vascular architecture and function during modulated blood oxygenation using susceptibility weighted magnetic resonance imaging[J]. Magn Reson Med, 2005, 54(1):87-95.

[40] BEN-HAIM S, ASAAD W F, GALE J T, et al. Risk factors for hemorrhage during microelectrode-guided deep brain stimulation and the introduction of an improved microelectrode design[J]. Neurosurgery, 2009, 64(4):754-763.

[41] LI J, CHANG S, LIU T, WANG Q, et al. Reducing the object orientation dependence of susceptibility effects in gradient echo MRI through quantitative susceptibility mapping[J]. Magn Reson Med, 2012, 68(5):1563-1569.

[42] LANGKAMMER C, SCHWESER F, KREBS N, et al. Quantitative susceptibility mapping (QSM) as a means to measure brain iron? A post mortem validation study[J]. Neuroimage, 2012, 62(3):1593-1599.

[43] DE ROCHEFORT L, LIU T, KRESSLER B, et al. Quantitative susceptibility map reconstruction from MR phase data using bayesian regularization: validation and application to brain imaging[J]. Magn Reson Med, 2010, 63(1):194-206.

[45] THANI N B, BALA A, SWANN G B, et al. Accuracy of postoperative computed tomography and magnetic resonance image fusion for assessing deep brain stimulation electrodes[J]. Neurosurgery, 2011, 69(1):207-214.

Recent progresses of magnetic resonance imaging of subthalamic nucleus

KONG Wei-shi1△, LU Ming-kuan1△, YANG Qing-song2, CHEN Zhi-chang3, QIU Yi-qing4, WU Xi4*

1. Cadet Brigade, Navy Military Medical University, Shanghai 200433, China 2. Department of Medical Imaging, Changhai Hospital, Navy Military Medical University, Shanghai 200433, China 3. Department of Navy Medicine, Navy Military Medical University, Shanghai 200433, China 4. Department of Neurosurgery, Changhai Hospital, Navy Military Medical University, Shanghai 200433, China

Subthalamic nucleus (STN) is the main target nucleus for deep brain stimulation (DBS) treatment in patients with Parkinson disease. To implant the electrode on the sensorimotor part of STN individually and accurately, the boundary of STN is required to be clarified clearly on the magnetic resonance imaging (MRI) without geometric distortion. At present, there are three categories of MRI sequences: spin echo sequence including T2-weighted imaging (T2WI), inversion recovery (IR), diffusion tensor imaging (DTI), and fractional anisotropy (FA); magnetization transfer technique including magnetic susceptibility weighted imaging (SWI) and T2-weighted magnitude imaging (T2*WI); image reconstruction technique such as quantitative susceptibility mapping (QSM). It is found that QSM can provide optimal signal-noise ratio to identify the boundary of STN, T2*technique comes second. T2WI has high geometric accuracy when the patients wear frame, which is appropriate for direct DBS implantation on STN with frame.

subthalamic nucleus; magnetic resonance imaging; deep brain stimulation; Parkinson disease

2017-06-12接受日期2017-09-18

国家重点研发计划“数字诊疗装备研发”试点专项课题(2016YFC0105900). Supported by “Digital Equipment of Diagnosis and Treatment” Special Program of National Key Research and Development Plan (2016YFC0105900).

孔维诗, 海军军医大学2014级本科学员. E-mail：2941106756@qq.com 路明宽, 海军军医大学2013级本科学员. E-mail：630071751@qq.com

△共同第一作者(Co-first authors).

*通信作者(Corresponding author). Tel: 021-31161789, E-mail：wuxi_smmu@sina.com

10.12025/j.issn.1008-6358.2017.20170502

R 742

A

[本文编辑] 姬静芳