分享:
分享到微信朋友圈
X
临床研究
脑脊液伪影在3.0 T MR二维和三维液体衰减反转恢复图像上的表现
钱银锋 张诚 吴津民 柏亚 余永强

钱银锋,张诚,吴津民,等.脑脊液伪影在3.0 T MR二维和三维液体衰减反转恢复图像上的表现.磁共振成像, 2012, 3(3): 179-183. DOI:10.3969/j.issn.1674-8034.2012.03.005.


[摘要] 目的 通过与1.5 T MRI比较,探讨3.0 T MR上2D和3D FLAIR图像脑脊液伪影的表现。方法 50例受试者行1.5 T MR 2D液体衰减反转恢复(FLAIR)成像,另外50例行3.0 T MR 2D和3D FLAIR成像,按4级分类法记录各受试者脑室系统、鞍上池和桥前池内脑脊液伪影:0级,无伪影;1级,伪影信号低于灰质;2级,伪影信号与灰质相似;3级,伪影信号高于灰质。将2级和3级合称为高级别。结果 1.5 T MR 2D FLAIR图像上,9例未见脑脊液伪影,余41例中共发现94处脑脊液伪影,其中2级最常见(54.3%),最常见于四脑室。3.0 T MR 2D FLAIR图像上,每例受试者至少有2处脑脊液伪影,其中3级最多见,占总伪影数50.0%。3.0 T MR 2D FLAIR上各部位高级别脑脊液伪影例数均明显高于1.5 T. 3D FLAIR图像上均未见脑脊液伪影。结论 2D FLAIR成像时,同1.5 T MRI相比,3.0 T MRI上高级别脑脊液伪影明显增加;但3D FLAIR图像上脑脊液信号抑制完全、无伪影。
[Abstract] Objective: To study cerebrospinal fluid (CSF) artifacts on 2D and 3D fluid-attnuated inversion-recovery (FLAIR) images at 3.0 T MR by comparing with 1.5 T system.Materials and Methods: Fifty individuals were examined with 2D FLAIR images with a 1.5 T system, another fifty cases were examined with 2D and 3D FLAIR images with a 3.0 T system. CSF artifacts within ventricles, suprasellar cistern and prepontine cistern were recorded using a 4-point scale (0=no artifacts, 1=artifacts intensity lower than that of gray matter, 2=artifacts intensity equal to that of gray matter, 3= artifacts intensity higher than that of gray matter), 2 and 3 scales were defined as high grade.Results: On 2D FLAIR images with 1.5 T system, 9 individuals had no CSF artifacts, and there were 94 regions with artifacts in the other 41 cases, among which, 2 scale was the most frequency and occupied 54.3% of all artifacts, and forth ventricle was the most common region. On 2D FLAIR images with 3.0 T system, artifacts were seen at least two regions in every cases, and 3 scale was the most frequency and occupied 50.0%, forth ventricle and prepontine cistern were the most common regions. Compared with 1.5 T system, the ratio of high grade artifacts in ventricle and subarachnoid space was higher on 2D FLAIR images with 3.0 T system. However, there were no artifacts on 3D FLAIR images with 3.0 T system.Conclusion: Compared with 1.5 T system, high grade CSF artifacts are more often on 2D FLAIR images with 3.0 T system, but 3D FLAIR imaging can complete null CSF signal and eliminate CSF artifacts.
[关键词] 液体衰减反转恢复;脑脊液伪影;蛛网膜下腔;磁共振成像
[Keywords] Fluid-attenuated inversion-recovery;Cerebrospinal fluid artifact;Subarachnoid space;Magnetic resonance imaging

钱银锋 安徽医科大学第一附属医院放射科,合肥 230022

张诚 安徽医科大学第一附属医院放射科,合肥 230022

吴津民 安徽医科大学第一附属医院放射科,合肥 230022

柏亚 安徽医科大学第一附属医院放射科,合肥 230022

余永强* 安徽医科大学第一附属医院放射科,合肥 230022

通讯作者:余永强,E-mail: shenyx@ustc.edu.cn


收稿日期:2011-12-20
接受日期:2012-01-31
中图分类号:R445.2 
文献标识码:A
DOI: 10.3969/j.issn.1674-8034.2012.03.005
钱银锋,张诚,吴津民,等.脑脊液伪影在3.0 T MR二维和三维液体衰减反转恢复图像上的表现.磁共振成像, 2012, 3(3): 179-183. DOI:10.3969/j.issn.1674-8034.2012.03.005.

       液体衰减反转恢复(fluid-attenuated inversion-recovery, FLAIR)序列通过抑制正常脑脊液信号,可更好地显示较小且靠近脑脊液的病变[1];并可显示出脑脊液成分变化导致的信号改变,在脑膜炎、蛛网膜下腔出血的诊断中具有作用。但由于脑脊液的流动性等原因可导致其有时不能被完全抑制,在FLAIR图像上表现为有一定信号的脑脊液伪影。另一方面,近年来,3.0 T MR机的临床应用逐渐增多,对其常见伪影文献已有一定的报道[2],但有关3.0 T MR FLAIR图像上脑脊液伪影表现的文献报道较少,笔者通过3.0 T MRI上2D、3D FLAIR和1.5 T MRI上2D FLAIR的对比研究,探讨3.0 T MRI FLAIR上脑脊液伪影的表现,以期为3.0 T MR上FLAIR图像的分析和序列选择提供依据。

1 材料与方法

1.1 临床资料

       搜集我院2011年1月至2011年10月间行MRI头颅检查者100例,其中男56例,女44例,年龄20~ 84岁,平均年龄52岁。所有患者常规MRI检查均无异常发现,无颅脑外伤、手术和精神病史。其中50例行1.5 T MRI 2D FLAIR检查,男33例,女17例,年龄21~82岁,平均年龄53岁。另50例行3.0 T MRI 2D和3D FLAIR检查,其中男23例,女27例,年龄20~84岁,平均年龄51岁。

1.2 MR检查方法

       1.5 T超导型MR扫描仪(GE Signa Horizon),正交头线圈。所有检查者均常规扫描轴面T1WI、T2WI和FLAIR, FLAIR序列成像参数为:TR 8000 ms, TE 150 ms, TI 1900 ms,层厚5.0 mm,层间距1.5 mm,矩阵192×160,FOV22 cm ×22 cm,NEX为1。3.0 T超导型MR扫描仪(GE HDx),8通道头颅线圈,常规扫描颅脑轴面T1WI、T2WI、2D FLAIR和3D FLAIR,2D FLAIR的成像参数为:TR 9000 ms,TE 154 ms,TI 2250 ms;层厚5.0 mm,层间距1.5 mm,矩阵320×192,FOV 22 cm×22 cm,NEX为1,成像时间1 min 49 s;3D FLAIR采用矢状面成像:TR 6000 ms,TE 125 ms,TI 1865 ms,FOV 22 cm×22 cm,层厚1.0 mm,层间距-0.5 mm,矩阵256×224,NEX为1,成像时间5 min 15 s。扫描结束后将3D FLAIR图像传送至Adw 4.4工作站,以与2D FLAIR相同的层面、层厚和层间距进行轴面重建。

1.3 MR图像评价方法

       大脑凸面的脑沟裂内脑脊液处于相对静止状态,在FLAIR上可被完全抑制;因此笔者将脑室系统和蛛网膜下腔分为以下几个部分进行观察:左右侧脑室、第三脑室、第四脑室、鞍上池和桥前池。将脑脊液伪影分为4级:0级为无伪影,脑脊液被完全抑制;11级为伪影信号强度高于背景噪声,但低于脑灰质;2级为伪影信号强度与脑灰质呈等信号;3级为伪影信号强度高于脑灰质;将2级和3级合称为高级别伪影。

       所有图像评价由2名高级诊断医师独立完成,当两者不一致时则通过协商决定。

1.4 统计学处理

       对各组间的高级别脑脊液伪影发生率进行Chisquare检验,将P<0.05定义为差异有统计学意义,所有统计学处理用SPSS 16.0软件完成。

2 结果

       50例行1.5 T MRI检查者,其中9例FLAIR图像上未见脑脊液伪影,余41例中共发现94处脑脊液伪影,其中2级最常见(54.3%),1级和3级分别占全部伪影的20.2%和25.5%。

       脑脊液伪影最常见于第四脑室(图1),发生率为60.0%;其次为桥前池,发生率为36.0%。两侧脑室内脑脊液伪影多见于左侧脑室,仅位于左侧脑室者8例,仅位于右侧脑室者2例,7例两侧脑室内均见脑脊液伪影者中5例两侧脑室脑脊液伪影级别相同(图2)。

       50例行3.0 T MRI检查者,其2D FLAIR图像上,每例受试者至少有2处脑脊液伪影,共发现238处脑脊液伪影,其中3级最多见,占总伪影数50.0%,1级和2级分别占17.2%和32.8%。

       脑脊液伪影最常见于第四脑室、桥前池(图3图4),发生率均为98.0%,且多为3级,分别为82.0%和74.0%;其次常见部位为鞍上池,发生率为92.0%。两侧脑室内脑脊液伪影亦多见于左侧脑室,仅左侧脑室出现者9例(图5),仅右侧脑室出现者3例,22例两侧脑室内均见脑脊液伪影者中16例两侧伪影级别相同(图6)。

       1.5 T和3.0 T MRI的2D FLAIR图像上,各部位的高级别伪影情况见表1。由表1可知,3.0 T MRI上各部位的高级别脑脊液伪影例数均明显多于1.5 T MRI。

       形态和部位上,无论是1.5 T还是3.0 T MRI,侧脑室内脑脊液伪影一般位于孟氏孔区,呈椭圆形(图2图5图6);第四脑室内者多位于其上部,呈小圆形或条片状,以前者常见。第三脑室、鞍上池和桥前池内的脑脊液伪影多呈片状(图3图4图7图8)。所有的脑脊液伪影均无占位效应,邻近结构无受压改变。

       全部50例3D FLAIR的原始像和轴面重建图像上均未见脑脊液伪影(图3,图4,图5,图6图8)。

图1  男,54岁,1.5 T MRI 2D FLAIR图像,鞍上池(1A)内未见脑脊液伪影;第四脑室上部见一2级小圆形脑脊液伪影,桥前池内无伪影(1B)
图2  男,65岁。1.5 T MRI 2D FLAIR图像,左右侧脑室内近孟氏孔处各见一椭圆形2级脑脊液伪影
图3  男,41岁。3.0 T MRI,2D FLAIR图像(3A)示鞍上池内3级脑脊液伪影和四脑室上部2级伪影,鞍上池内结构显示不清。3D FLAIR轴面重建图像(3B)示脑脊液信号抑制完全,鞍上池内结构显示清晰
图4  女,48岁。3.0 T MRI,2D FLAIR图像(4A)示桥前池和第四脑室内3级伪影。3D FLAIR轴面重建图像(4B)示脑脊液信号抑制完全
图5  女,34岁。3.0 T MRI,2D FLAIR图像(5A)示左侧脑室近孟氏孔处3级脑脊液伪影,该伪影在3D FLAIR轴面重建图像(5B)上未见显示
图6  男,37岁。3.0 T MRI,2D FLAIR图像(6A)示双侧脑室内对称性脑脊液伪影,3D FLAIR轴面重建图像(6B)无脑脊液伪影
Fig. 1  2D FLAIR images obtained with a 1.5 T system in a 54-year-old male. There are no CSF artifacts in suprasellar cistern (1A) and prepontine cistern (1B), and one 2 scale round artifacts is found in upper of forth ventricle (1B).
Fig. 2  2D FLAIR image obtained with a 1.5 T system in a 65-year-old male. There is one oval-in-shape 2 scale artifacts in right and left ventricle around the foramen of Monro.
Fig. 3  Comparison FLAIR images obtained with a 3.0 T system in a 41-year-old male. There were 3 scale CSF artifacts in suprasellar cistern and 2 scale artifacts in upper of forth ventricle on 2D FLAIR image (3A), but the artifacts were disappear on reformatted axial 3D FLAIR image (3B) at similar location, and the normal anatomy of suprasellar cistern was clear.
Fig. 4  Comparison FLAIR images obtained with a 3.0 T system in a 48-year-old female. There were 3 scale CSF artifacts in prepontine cistern and upper of forth ventricle on 2D FLAIR image (4A), but the artifacts were disappeared on reformatted axial 3D FLAIR image (4B) at similar location.
Fig. 5  Comparison FLAIR images obtained with a 3.0 T system in a 34-year-old female. There was one of 3 scale artifacts in left ventricle around the foramen of Monro on 2D FLAIR image (5A). However, it disappeared on reformatted axial 3D FLAIR image (5B).
Fig. 6  Comparison FLAIR images obtained with a 3.0 T system in a 37-year-old male. There were symmetrical 2 scale artifacts in both lateral ventricle on 2D FLAIR image (6A), the signal of CSF was null on reformatted axial 3D FLAIR image (6B).
图7  男,60岁。1.5 T MRI,2D FLAIR图像示第三脑室内片状2级脑脊液伪影
图8  男,64岁。3.0 T MRI,2D FLAIR图像(8A)示第三脑室内片状3级脑脊液伪影,3D FLAIR轴面重建图像(8B)无脑脊液伪影
Fig. 7  2D FLAIR image obtained with a 1.5 T system in a 60-year-old male. There was 2 scale artifacts in third ventricle and was slabby in shape.
Fig. 8  Comparison FLAIR images obtained with a 3.0 T system in a 64-year-old male. There was slabby 3 scale artifacts in third ventricle on 2D FLAIR image (8A), the signal of CSF was null on reformatted axial 3D FLAIR image (8B).
表1  1.5 T和3.0 T的2D FLAIR上高级别脑脊液伪影例数
Tab. 1  The number of high grade CSF artifacts on 2D FLAIR images with 1.5 T and 3.0 T system

3 讨论

       FLAIR通过施加特定TI的180°脉冲选择性抑制脑脊液,从而使皮层表面和脑室周围的病变更易被检出,已成为颅脑MRI检查的常规序列。脑膜炎、蛛网膜下腔出血、柔脑膜肿瘤种植等时,脑脊液的T1值发生改变(或同时T2值发生改变),在FLAIR上显示为高信号,因此FLAIR在这些蛛网膜下腔病变的检出中亦具有重要作用[3,4]

3.1 FLAIR图像上脑脊液伪影产生的原因

       脑脊液伪影产生的主要原因是脑脊液处于流动状态,2D FLAIR成像时,在反转脉冲和采集信号的时间间隔(△t)内,上方或下方的脑脊液流入成像层面,因此未能被有效抑制,从而产生信号。此外,脑脊液搏动、局部场强不均匀等亦可产生脑脊液伪影[5]。其最常见于鞍上池、桥前池和脑室的孔处。脑脊液伪影的发生与年龄有一定的相关性,在儿童很少见,所以笔者仅研究了成人的脑脊液伪影表现。

3.2 1. 5 T和3. 0 T FLAIR图像上脑脊液伪影表现

       本组结果表明,2D FLAIR成像时,与1.5 T相比,3.0 T上脑脊液伪影的发生率和伪影级别明显增加。其原因可能为以下两个方面:(1) TI时间延长导致流入成像层面内未被抑制的脑脊液增多;(2) 3.0 T MRI对局部磁场不均匀性更敏感,尤其在鞍上池等近颅底层面。Neema等[6]研究了22例正常成人颅脑在1.5 T和3.0 T MR FLAIR图像上的表现,认为脑脊液伪影在两者间相似,但作者没有对脑脊液伪影进行量化分析和比较。

       尽管3.0 T上脑室内脑脊液伪影的发生率尤其是3级伪影的发生率明显增加,但无论是1.5 T还是3.0 T MRI,脑室内脑脊液伪影均具有相同的特点:(1)部位上,多位于脑室孔附近,侧脑室者一般位于孟氏孔区,第四脑室者常位于其上部;(2)形态上,侧脑室者多呈椭圆形,且常双侧同时存在,较对称,第三脑室者一般呈片状,第四脑室者多呈小圆形或条片状。根据这些特点一般可很容易地将脑室内脑脊液伪影与正常结构和病变相区分。

       蛛网膜下腔病变时,鞍上池和桥前池是FLAIR图像上高信号常见位置之一,因此有效抑制这些部位的脑脊液信号极其重要。由于流速、层厚、采集间隔等因素的影响,脑脊液伪影的信号强度并非固定一致,其中与灰质呈等信号或高信号的高级别伪影易于掩盖或误为蛛网膜下腔病变,而1级伪影一般不会误为蛛网膜下腔病变,容易识别。本组中,在1.5 T MRI,鞍上池和桥前池高级别脑脊液伪影的发生率分别为10.0%和26.0%,而在3.0 T的2D FLAIR上,其发生率明显增加,分别为40.0%和98.0%。提示在3.0 T,2D FLAIR可能不适于检测蛛网膜下腔病变。

3.3 消除FLAIR图像上脑脊液伪影的方法

       为了尽量减少或消除脑脊液伪影,在1.5 T上,研究者们提出了多种方法,包括采用非层面选择性反转脉冲、依赖于反转时间的层面K空间重排(KRISP)技术、隔绝脉冲、增加采集间隔因子等[7,8],可明显减少脑脊液高信号伪影,但这些技术在3.0 T上的应用尚未见报道。

       3D FLAIR成像时,由于采用非层面选择的反转脉冲,全部的脑脊液信号被均匀的抑制。本组中全部50例行3D FLAIR成像者,脑脊液信号抑制完全,脑室和蛛网膜下腔内均未见脑脊液伪影,相应处解剖结构显示清楚,与Chagla等[9]的研究结果相似。

       由于FLAIR的TR时间很长,因此最初临床上在3.0 T上行3D FLAIR成像是不现实的。最近几个方面的进步,明显缩短了3D FLAIR的成像时间:(1)再聚焦射频脉冲反转角的调整极大的增加了回波链长度;(2)多个方位加速的并行采集减少了获得3D数据所需要的回波数;(3)新的3D查看排序技术允许K空间的角落部分被忽略,从而进一步的减少需要的回波数。这些技术的应用使3D FLAIR成像时间缩短至10 min以内,本研究中尽管采用很薄的1 mm层厚,成像时间仅为5 min 15 s,临床应用完全可行。

       本组结果表明,同1.5 T相比,3.0 T MRI明显增加了2D FLAIR上脑脊液伪影的出现几率和级别,可能不利于蛛网膜下腔病变的诊断,3D FLAIR上脑脊液信号抑制完全,疑诊蛛网膜下腔病变时应选择3D FLAIR成像。

[1]
Dai JP, Chen HY. MR pulse sequences in the central nervous system: part 1. Chin J Magn Reson Imaging, 2010, 1(3): 220-226.
戴建平,陈红艳. 磁共振脉冲序列在中枢神经系统中的应用(一). 磁共振成像, 2010, 1(3): 220-226.
[2]
Yin JZ. Common artifacts in 3 T high-field MRI systems: principle, appearance and remedy. Chin J Magn Reson Imaging, 2010, 1(4): 291-294.
尹建忠. 3 T高场磁共振设备的常见伪影:原理、表现与对策. 磁共振成像, 2010, 1(4): 291-294.
[3]
Porto L, Kieslich M, Bartels M, et al. Leptomeningeal metastases in pediatrics: magnetic resonance image manifestations and correlation with cerebral spinal fluid cytology. Pediatr Int, 2010, 52(4): 541-546.
[4]
Shimoda M, Hoshikawa K, Shiramizu H, et al. Problems with diagnosis by fluid-attenuated inversion recovery magnetic resonance imaging in patients with acute aneurysmal subarachnoid hemorrhage. Neurol Med Chir (Tokyo), 2010, 50(7): 530-537.
[5]
Stuckey SL, Goh TD, Heffernan T, et al. Hyperintensity in the subarachnoid space on FLAIR MRI. AJR Am J Roentgenol, 2007, 189(4): 913-921.
[6]
Neema M, Guss ZD, Stankiewicz JM, et al. Normal findings on brain fluid-attenuated inversion recovery MR images at 3 T. AJNR Am J Neuroradiol, 2009, 30(5): 911-916.
[7]
Ding JR, Fan SF, Pan JL, et al. Influence of imaging parameters on high-intensity cerebrospinal fluid artifacts in T2-FLAIR MR imaging. J Med Imaging, 2008, 18(3): 228-230.
丁建荣,樊树峰,潘璟俐, 等. 参数选择对磁共振T2-FLAIR像中脑脊液高信号伪影的影响. 医学影像学杂志, 2008, 18(3): 228-230.
[8]
Wu HM, Yousem DM, Chung HW, et al. Influence of imaging parameters on high-intensity cerebrospinal fluid artifacts in fast-FLAIR MR imaging. AJNR Am J Neuroradiol, 2002, 23(3): 393-399.
[9]
Chagla GH, Busse RF, Sydnor R, et al. Three-dimensional fluid attenuated inversion recovery imaging with isotropic resolution and nonselective adiabatic inversion provides improved three-dimensional visualization and cerebrospinal f l u i d suppression compared to two-dimensional FLAIR at 3 tesla. Inves Radiol, 2008, 43(8): 547-551.

上一篇 脑星形细胞瘤磁敏感效应级别与1H-MR波谱相对定量的相关性探讨
下一篇 12例视神经胶质瘤MRI分析
  
诚聘英才 | 广告合作 | 免责声明 | 版权声明
联系电话:010-67113815
京ICP备19028836号-2