分享:
分享到微信朋友圈
X
基础研究
MR弹性成像与钆塞酸二钠增强T1 mapping定量评估兔肝纤维化的对比研究
张豪 邹立秋 钟文新 麦晓飞 石桥

Cite this article as: ZHANG H, ZOU L Q, ZHONG W X, et al. Quantitative evaluation of liver fibrosis by MRE and Gd-EOB-DTPA-enhanced T1 mapping magnetic resonance imaging in a rabbit model[J]. Chin J Magn Reson Imaging, 2024, 15(8): 172-178.本文引用格式:张豪, 邹立秋, 钟文新, 等. MR弹性成像与钆塞酸二钠增强T1 mapping定量评估兔肝纤维化的对比研究[J]. 磁共振成像, 2024, 15(8): 172-178. DOI:10.12015/issn.1674-8034.2024.08.026.


[摘要] 目的 评价MR弹性成像(magnetic resonance elastography, MRE)、钆塞酸二钠(gadolinium ethoxybenzyl diethyle netriamine pentaacetic acid, Gd-EOB-DTPA)增强T1 mapping在肝纤维化(liver fibrosis, LF)早期定量的价值。材料与方法 实验材料为新西兰大白兔(共120只),分为正常对照组(n=20)与肝纤维化组[LF组,50%四氯化碳(carbon tetrachloride, CCL4)油溶液模型,n=100],分别在第4、5、6、15周末取5只对照组与25只LF组兔行MRI肝脏轴位T1WI、MRE及Gd-EOB-DTPA增强T1 mapping扫描,测量肝脏弹性硬度(liver stiffness, LS)、平扫T1弛豫时间(T1native)及Gd-EOB-DTPA增强T1 mapping扫描20 min后T1弛豫时间(T120min),计算T1弛豫时间减少率(ΔT120min)及1/T1弛豫时间增加值(ΔR120min);采用Scheuer评分系统进行LF病理学分期,单因素方差分析对比各定量参数与LF分期间的差异,Spearman相关性分析比较各定量参数与LF病理分期相关性,受试者工作特征(receiver operating characteristic, ROC)曲线评估各定量参数对LF分期诊断效能。结果 共纳入96只活兔模型,病理学评分:F0期15只、F1期22只、F2期22只、F3期18只、F4期19只。LS、T1native、T120min、ΔT120min、ΔR120min于LF各期间差异均有统计学意义(P<0.05)。LS、T1native、T120min、ΔT120min、ΔR120min均与LF分期相关(r=0.935、0.559、0.770、-0.418、-0.686,P<0.001)。F0 vs. F1~F4、F0 vs. F1~F2、F0 vs. F3~F4、F1~F2 vs. F3~F4的LS值ROC曲线下面积(area under the curve, AUC)分别为0.988、0.979、1.000、0.995,T120min值AUC分别为0.914、0.852、0.987、0.896。结论 在评估LF早期诊断中,MRE及Gd-EOB-DTPA增强T1 mapping成像均显示出较好的诊断价值,MRE优于Gd-EOB-DTPA增强T1 mapping。
[Abstract] Objective To compare the accuracy of MR elastography (MRE) and Gd-EOB-DTPA-enhanced T1 mapping in the quantitative evaluation of liver fibrosis (LF) staging.Materials and Methods One hundred and twenty rabbits were randomly divided into control group (n=20), which were injected subcutaneously with normal saline solution, and LF group (n=100), which were received 50% (carbon tetrachloride) CCl4 oil solution. The control group (n=5) and LF group (n=25) underwent MRI axial scan, T1WI, MRE, Gd-EOB-DTPA-enhanced T1 mapping at the end of the 4th, 5th, 6th, 15th week. The pathological LF staging was based on Scheuer staging system. The quantitative parameter included liver stiffness (LS), pre- and post-contrast T1 values of the liver (T1native and T120min), the reduction rate of T1 relaxation time (ΔT120min) and the increase in T1 relaxation rate (ΔR120min), were compared the differences by one-way ANOVA analysis. Spearman correlation coefficients, Receiver operating characteristic (ROC) analysis was used respectively to determine the correlation and diagnostic performance between quantitative parameters and pathological LF staging.Results A total of 96 rabbits were included in F0 (n=15), F1 (n=22), F2 (n=22), F3 (n=18) and F4 (n=19). LS, T1native, T120min, ΔT120min, ΔR120min showed significant differences among all LF staging (P<0.05). There were correlation between LS, T1native, T120min, ΔT120min, ΔR120min and LF stage (r=0.935, 0.559, 0.770, -0.418 -0.686, P<0.001), respectively. LS exhibited the largest area under the curve (AUC), which were 0.988, 0.979, 1.000, 0.995 for F0 vs. F1~F4, F0 vs. F1~F2, F0 vs. F3~F4, F1~F2 vs. F3~F4, respectively. Secondly, the AUC of T120min were 0.914, 0.852, 0.987, and 0.896, respectively.Conclusions In the early quantitative evaluation of LF staging, MRE and Gd-EOB-DTPA enhanced T1 mapping had demonstrated significant diagnostic value, with MRE outperforming Gd-EOB-DTPA enhanced T1 mapping.
[关键词] 肝纤维化;磁共振弹性成像;肝脏硬度;钆塞酸二钠;T1 mapping成像
[Keywords] liver fibrosis;magnetic resonance elastography;liver stiffness;Gd-EOB-DTPA;T1 mapping

张豪 1   邹立秋 1   钟文新 1   麦晓飞 1   石桥 2*  

1 深圳市南山区人民医院放射科,深圳 518052

2 深圳市宝安区妇幼保健院放射科,深圳 518100

通信作者:石桥,E-mail:docshi@126.com

作者贡献声明:石桥设计本研究的方案,对稿件重要内容进行了修改;张豪起草和撰写稿件,分析本研究的数据,获得深圳市南山区科技计划项目资助;邹立秋参与研究的构思和设计,对稿件重要内容进行了修改,获得国家自然科学基金面上项目资助;麦晓飞负责采集数据,参与资料的分析,起草部分论文内容;钟文新参与设计实验,实施研究,参与数据的分析及论文重要内容的修改;全体作者都同意发表最后的修改稿,同意对本研究的所有方面负责,确保本研究的准确性和诚信。


基金项目: 国家自然科学基金面上项目 81771805 深圳市南山区科技计划项目 NS2024079
收稿日期:2024-03-07
接受日期:2024-07-30
中图分类号:R445.2  R735.7  R-332 
文献标识码:A
DOI: 10.12015/issn.1674-8034.2024.08.026
本文引用格式:张豪, 邹立秋, 钟文新, 等. MR弹性成像与钆塞酸二钠增强T1 mapping定量评估兔肝纤维化的对比研究[J]. 磁共振成像, 2024, 15(8): 172-178. DOI:10.12015/issn.1674-8034.2024.08.026.

0 引言

       肝纤维化(liver fibrosis, LF)是各种慢性肝病(chronic liver disease, CLD)共同病理基础[1],肝细胞发生慢性损伤后,星状细胞等被激活,肌纤维母细胞增殖、大分子物质(如胶原蛋白Ⅲ、透明质酸)在肝细胞外间隙沉积,导致肝脏结构紊乱,发生血液动力学、肝脏物理等特性改变[2, 3],最终形成肝纤维化。每年约200万人死于CLD,早期诊治CLD是全球卫生健康的焦点问题[2]。CLD的进展,导致肝硬化、肝癌、肝性脑病等并发症[4]。LF分期是CLD中与肝脏并发症发生率、总体死亡率及预后相关的唯一且最重要的因素,早期干预使病程逆转,早期行LF分期对CLD的治疗具有重要临床意义[5]

       肝脏穿刺活检目前是LF诊断及分期之金标准[6],但具有创伤性,患者接受度低,潜在并发症风险(约3%),包括疼痛(约84%)、出血(约11%)、死亡(约0.001%-0.200%)等,以及取样误差、评估误差等影响肝脏穿刺活检病理分期的准确性[7, 8]。无创性影像学手段定量评估在CLD诊疗、随访及疗效评估等具有重要意义[9]。其中,磁共振弹性成像(magnetic resonance elastography, MRE)通过物理的方法反映肝脏病变硬度的改变[10, 11],钆塞酸二钠(gadolinium ethoxybenzyl diethyle netriamine pentaacetic acid, Gd-EOB-DTPA)增强T1 mapping成像,可监测肝细胞功能、血液动力学改变[5, 12]。本实验研究旨在对比MRE、Gd-EOB-DTPA增强T1 mapping在评估动物肝纤维化模型LF早期分期中的价值。

1 材料与方法

1.1 实验动物与模型构建

1.1.1 实验动物

       实验研究在深圳市南山区人民医院动物实验伦理委员会批准下执行(批准号:20170218)。总计120只健康雄性新西兰大白兔(由广东省医学实验动物中心提供),6月龄,体质量约2.0~2.5 kg,按数字表法随机分组为正常对照组(n=20)与LF组(n=100)。动物饲养条件:单笼,恒温18 ℃~28 ℃,湿度40%~70%,间断照明10 h∶14 h,限制动物饮食,但自由饮水。

1.1.2 动物模型构建

       根据动物状态,每周(时间点约第1~3周、4~6周、7~10周)LF组于兔颈背部皮下注射50%四氯化碳(carbon tetrachloride, CCL4)油溶液(100%CCL4液与橄榄油为1∶1配比)1次~2次,剂量分别约0.1、0.2、0.3 mL/kg;正常对照组注射等频率及等剂量生理盐水。

1.2 MRI扫描

       采用3.0 T MR扫描仪(Philips Ingenia, Netherlands),18通道相控阵膝关节线圈。构建动物模型后第4、5、6、15周末(每次随机选取5只对照组与25只LF组兔)进行MRI扫描。24 h兔前扫描禁饮食、禁水,扫描前15 min腹腔注射4 mL/kg水合氯醛溶液(4%)后,呼吸面罩行持续性麻醉[吸入2%~3%异氟醚及100%氧气(3 mL/ min)]并用腹带仰卧位固定实验兔。序列参数:(1)轴位T1WI快速自旋回波(fast spin echo, FSE),FOV 140 mm×140 mm,TR 100 ms,TE 1.6 ms,矩阵252×199,翻转角50°,层厚4 mm,层间距0.4 mm;(2)轴位MRE采用稳态自由进动梯度回波序列(SSFP),FOV 180 mm×180 mm,TR 500 ms,TE 55.2 ms,矩阵64×66,翻转角20°,振动频率60 Hz,气压机械波发生器(圆形、直径约18 cm)固定于大白兔腹壁,层厚4 mm,层间距0.4 mm;(3)T1 mapping采用改良Look-Locker反转恢复序列(MOLLI-T1 mapping),平扫及Gd-EOB-DTPA增强20 min后(肝胆期)成像,TR 3 ms,TE 1 ms,翻转角20°,FOV 150 mm×150 mm,矩阵252×199,经兔耳缘静脉注射Gd-EOB-DTPA(剂量0.1 mL/kg)并用生理盐水(2 mL)冲管后增强扫描。

1.3 图像分析

       双盲条件下,获取定量参数值由具备5年腹部影像诊断经验的2名放射诊断医师(1名副主任医师及1名主治医师)执行。MRE图像将由MRE View软件(Resoundant, Inc, USA)处理,生成波图、幅度图和彩色弹性图(95%置信图覆盖)。T1 mapping图像将由MR Map(IDL8.5 ITT Visual Information Solutions, Boulder CO, USA)软件处理,包含平扫及Gd-EOB-DTPA增强肝胆期图像。选取3层肝脏显示最大层面T1WI图像、同层MRE彩色弹性图、同层T1 mapping平扫及增强肝胆期图像,ROI置于同层同位(5个/层),避开肝脏边缘、胆管、血管、伪影,计算平均测值,包含肝脏弹性硬度(liver stiffness, LS)、平扫T1弛豫时间(T1native)、肝胆期T1弛豫时间(T120min)、T1弛豫时间减少率[ΔT120min=(T1native-T120min)/T1native×100%]及1/T1弛豫时间增加值[ΔR120min=(1/T120min-1/T1native)×100%](图1)。

图1  1A~1E示F0~F4期MRE彩色弹性图,随肝脏硬度增加,颜色从紫色到红色(0~8 kPa),所得LS测量值分别为0.96、1.59、1.72、2.32、2.68 kPa;1F~1J示F0~F4期T1native伪彩图,T1native测量值分别为825.21、870.42、924.51、942.78、1004.45 ms;1K~1O示F0~F4期T120min伪彩图,T120min测量值分别为296.35、330.48、371.98、391.32、449.08 ms。MRE:磁共振弹性成像;LS:肝脏弹性硬度;T1native:平扫T1弛豫时间;T120min:肝胆期T1弛豫时间。
Fig. 1  Elastogram of MRE in F0~F4 stage (1A-1E), with the increase of liver stiffness, the color changes from purple to red (0~8 kPa), and obtained LS values are 0.96 kPa, 1.59 kPa, 1.72 kPa, 2.32 kPa and 2.68 kPa, respectively. T1native map from F0 to F4 stage (1F-1J), and T1native values are 825.21 ms, 870.42 ms, 924.51 ms, 942.78 ms and 1004.45 ms, respectively. T120min map from F0 to F4 stage (1K-1O), and T120min values are 296.35 ms, 330.48 ms, 371.98 ms, 391.32 ms and 449.08 ms, respectively. MRE: magnetic resonance elastography; LS: liver stiffness; T1native and T120min: pre- and post-contrast T1 values of the liver.

1.4 病理分析

       实验动物于扫描后均采用空气栓塞法(经耳缘静脉)处死,肝脏用10%甲醛固定后制成石蜡切片,行HE、Masson三色染色,由两位具备10年病理诊断经验的病理医师(均为副主任医师)在双盲条件下于显微镜下行Scheuer病理分期(共5期)[13],F0:无纤维化;F1:无纤维间隔的局限窦周、小叶内纤维化伴门静脉汇管区扩大;F2:汇管区周围纤维化、少量纤维间隔但小叶结构保留;F3:大量纤维间隔、小叶结构紊乱,无肝硬化;F4:肝硬化。

1.5 统计学分析

       数据由MedCalc(version 20.022, Acacialaan, Belgium)与IBM SPSS Statistics(version 22.0, Chicago, IL, USA)软件处理,Kolmogorov-Smirnov检验数据是否正态分布,符合正态分布的数据以(x¯±s)表示。采用Kappa检验评估两位病理医师分期结果的一致性(Kappa值:≤0,无一致性;0.01~0.20,一致性差;0.21~0.40,一致性一般;0.41~0.60,一致性中等;0.61-0.80,一致性较高;0.81~1.00,一致性很高)[14]。组内相关系数(intra-class correlation coefficient, ICC)评价两位放射诊断医师测值结果(包括LS、T1native、T120min)的一致性(ICC值:<0.50,一致性差;0.50~0.74,一致性中等;0.75~0.89,一致性较高;0.90~1.00,一致性很高)[15]。单因素方差分析(one-way analysis of variance, ANOVA)比较LF分期间数据差异,组间两两比较采用Dunnett's T3法。Spearman法获得与LF病理分期间相关性。受试者工作特征(receiver operating characteristic, ROC)曲线分析数据对LF分期的诊断效能,包括鉴别正常肝脏与LF(F0 vs. F1~F4)、诊断早期LF(F0 vs. F1~F2)、诊断进展期LF(F0 vs. F3~F4)、鉴别早期与进展期LF(F1~F2 vs. F3~F4),效能判定指标(最优化Youden指数下)为ROC曲线下面积(area under the curve, AUC)、最佳截断值、敏感度、特异度、P值,DeLong法比较AUC差异。P<0.05为差异具有统计学意义。

2 结果

2.1 动物模型及病理结果

       对照组兔中2只因呼吸伪影造成图像不佳、3只因麻醉意外死亡,LF组兔中11只因不耐受CCL4油溶液死亡、4只因麻醉意外死亡、4只因呼吸伪影造成图像不佳予以排除,共96只兔入组(图2),病理结果:F0期15只、F1期22只、F2期22只、F3期18只、F4期19只。两位病理医师分期结果的一致性很高(Kappa值=0.908,P<0.001)。

图2  F0~ F4期兔肝脏病理染色图(Masson ×100)。2A:F0期,无肝纤维化;2B:F1期,无纤维间隔的局限窦周、小叶内纤维化伴门静脉汇管区扩大;2C:F2期,汇管区周围纤维化、少量纤维间隔但小叶结构保留;2D:F3期,大量纤维间隔、小叶结构紊乱,无肝硬化;2E:F4期,肝硬化。
Fig. 2  Pathological staining of liver of F0-F4 stage (Masson ×100). 2A: Stage F0, without hepatic fibrosis; 2B: Stage F1, limited perinusional and intraclobular fibrosis without fibrous septum with enlarged portal portal area; 2C: Stage F2, portal fibrosis, few fibrous septa but lobular structure remained; 2D: Stage F3, numerous fibrous septa, lobular structure disorders, no cirrhosis; 2E: Stage F4, cirrhosis.

2.2 定量参数分析

       各期LS、T1native、T120min、ΔT120min、ΔR120min组内均呈正态分布(表1)。LS值在两位放射诊断医师之间一致性很高(ICC值=0.992,P<0.001),T1native、T120min在两位放射诊断医师之间一致性较高(ICC值=0.818、0.865,P<0.001)(表2)。

2.2.1 不同分期定量参数值比较

       Levene方差齐性检验得T1native、ΔT120min方差齐(P>0.05),LS、T120min、ΔR120min方差不齐(P<0.05)。ANOVA分析T1native、ΔT120min:(1)T1native在F0与F1、F2、F3、F4组间差异有统计学意义(P<0.05);(2)ΔT120min在F0与F3组间,F1与F3、F4组间,F2与F3组间差异有统计学意义(P<0.05)。Welch方差分析LS、T120min、ΔR120min:(1)LS在F0与F1、F2、F3、F4组间,F1与F2、F3、F4组间,F2与F3、F4组间,F3与F4组间差异均有统计学意义(P<0.05);(2)T120min在F0与F1、F2、F3、F4组间,F1与F2、F3、F4组间,F2与F3、F4组间差异有统计学意义(P<0.05);(3)ΔR120min在F0与F2、F3、F4组间,F1与F2、F3、F4组间,F2与F3组间有统计学意义(P<0.05)。

2.2.2 定量参数与LF分期相关性

       LS、T1native、T120min与LF分期呈正相关(r=0.935、0.559、0.770,P<0.001);ΔT120min、ΔR120min与LF分期呈负相关(r=-0.418、-0.686,P<0.001)(图3)。

图3  LS、T1native、T120min、ΔT120min、ΔR120min与肝纤维化分期相关性曲线。LS(3A)、T1native(3B)、T120min(3C)与肝纤维化分期呈正相关(r=0.935、0.559、0.770,P<0.001);ΔT120min(3D)、ΔR120min(3E)与肝纤维化分期呈负相关(r=-0.418、-0.686,P<0.001)。LS:肝脏弹性硬度;T1native:平扫T1弛豫时间;T120min:肝胆期T1弛豫时间;ΔT120min:T1弛豫时间减少率;ΔR120min:1/T1弛豫时间增加值。
Fig. 3  Correlation curves of LS, T1native, T120min, ΔT120min and ΔR120min in LF stages. LS (3A), T1native (3B) and T120min (3C) are positively correlated with LF stages (r=0.935, 0.559, 0.770, P<0.001), respectively. ΔT120min (3D) and ΔR120min (3E) are negatively correlated with LF stages (r=-0.418, -0.686, P<0.001), respectively. LS: liver stiffness; T1native and T120min: pre- and post-contrast T1 values of the liver; ΔT120min: the reduction rate of T1 relaxation time; ΔR120min: the increase in T1 relaxation rate.

2.2.3 分析比较各定量参数值诊断效能

       LS诊断效能最佳(AUCF0 vs. F1~F4、AUCF0 vs. F1~F2、AUCF0 vs. F3~F4、AUCF1~F2 vs. F3~F4分别为0.988、0.979、1.000、0.995)(表3图4)。T1native的AUCF0 vs. F1~F4(0.932)、AUCF0 vs. F1~F2(0.900)稍高于T120min(0.914、0.852),但是DeLong检验显示两两比较差异无统计学意义(Z<1.96,P>0.05);T120min的AUCF0 vs. F3~F4(0.987)、AUCF1~F2 vs. F3~F4(0.896)高于T1native(0.969、0.665),DeLong检验显示AUCF1~F2 vs. F3~F4两两比较差异具有统计学意义(Z=3.135,P=0.002)。ΔR120min诊断效能(AUC分别为0.837、0.735、0.959、0.857)优于ΔT120min(AUC分别为0.644、0.536、0.773、0.751),DeLong检验显示差异具有统计学意义(Z>1.96,P<0.05),ΔT120min的AUCF0 vs. F1~F2无统计学意义(P=0.685)。

图4  LS、T1native、T120min、ΔT120min、ΔR120min鉴别正常肝脏与肝纤维化(F0 vs. F1~F4;4A)、诊断早期肝纤维化(F0 vs. F1~F2;4B)、诊断进展期肝纤维化(F0 vs. F3~F4;4C)、鉴别早期与进展期肝纤维化(F1~F2 vs. F3~F4;4D)的ROC曲线。LS:肝脏弹性硬度;T1native:平扫T1弛豫时间;T120min:肝胆期T1弛豫时间;ΔT120min:T1弛豫时间减少率;ΔR120min:1/T1弛豫时间增加值;ROC:受试者工作特征。
Fig. 4  ROC curve of LS, T1native, T120min, ΔT120min and ΔR120min in identifying liver fibrosis stages between F0 and F1-F4 (4A), F0 and F1-F2 (4B), F0 and F3-F4 (4C), F1-F2 and F3-F4 (4D), respectively. ROC: receiver operating characteristic; LS: liver stiffness; T1native and T120min: pre- and post-contrast T1 values of the liver; ΔT120min: the reduction rate of T1 relaxation time; ΔR120min: the increase in T1 relaxation rate.
表3  判别LS、T1native、T120min、ΔT120min、ΔR120min在肝纤维化分期的诊断效能
Tab. 3  Diagnostic efficacy of LS, T1native , T120min, ΔT120min and ΔR120min in LF stages
表1  LF各期兔LS、T1native、T120min、ΔT120min、ΔR120min(x¯±s
Tab. 1  Summary values of LS, T1native, T120min, ΔT120min and ΔR120min in different LF stages (x¯±s)
表2  两位放射诊断医师重复测量兔LS、T1native、T120min结果及组内相关系数
Tab. 2  The results of LS, T1native and T120min were measured repeatedly by two radiologists, and the ICC

3 讨论

       本动物实验采用兔CCL4肝纤维化模型,其优势是肝实质内纤维化的分布均匀,能够准确进行病理学分期。Gd-EOB-DTPA动态增强成像是利用双输入-双室Extended Tofts药物动力学模型,反映肝细胞功能及微循环灌注[16]。T1 mapping序列是客观反映质子纵向弛豫时间(T1值),并且T1值与Gd-EOB-DTPA浓度在肝实质内分布呈高度线性相关[17];MRE通过机械波在肝实质中传播速度的平方与肝脏硬度呈正相关[7]。当肝纤维化时肝细胞损伤、修复,大分子物质过度沉积,导致肝脏结构、功能、灌注、硬度等发生改变[5]。采用Gd-EOB-DTPA增强T1 mapping及MRE评估LF分期,对比两者早期的诊断效能。

3.1 MRE诊断效能

       MRE是LF分期目前最准确的无创性影像诊断手段,它不受MR机器、场强及扫描参数等因素影响,能够客观、定量评估肝脏物理特征[18],重复性高,被称为“肝脏影像触诊”[7]。机械波发生器将剪切波(采用频率60 Hz)经软管传播至肝脏,利用肝组织硬度和机械波传播速度之间的关系来评估肝脏硬度,肝纤维化组织因细胞外基质沉积导致肝脏弹性硬度的增加,采用运动编码梯度获得肝脏质点空间位移,获得相位图,显示剪切波波长不同,获得波图,拟合反演算法生成伪彩弹性图(95%置信图覆盖)[19, 20]。文献报道MRE能早期诊断LF、病情监测、随访复查,还能纵向评估药物疗效等[21, 22, 23]。目前多采用2D MRE,包括梯度回波序列(gradient recalled echo, GRE)、自旋平面回波序列(spin-echo echo-planar imaging, seEPI)[21, 22]。也有研究显示[24, 25],3D MRE(seEPI)更客观,不易受肝脏铁沉积、腹水、波间干扰伪影的影响,成像范围覆盖全肝,评估更全面,成像失败率(1.7%)远低于2D GRE(7.01%)、2D seEPI(5.2%)。BHUIYAN等[26]研究认为3D MRE并没有显示额外的诊断价值,2D MRE扫描速度更快,伪影较少。本实验ANOVA显示LS均能鉴别各期LF,Spearman分析显示LS与病理学分期呈明显正相关(r=0.935,P<0.001),诊断效能AUCF0 vs. F1~F4、AUCF0 vs. F1~F2、AUCF0 vs. F3~F4、AUCF1~F2 vs. F3~F4分别为0.988、0.979、1.000、0.995,与前期研究结果一致[27],具有较高的敏感度及特异度,早期LF敏感度(86.36%)、特异度(100.00%),有助于治疗方案的制订。

3.2 Gd-EOB-DTPA增强T1 mapping诊断效能

       Gd-EOB-DTPA为肝细胞特异性MR对比剂,注射10~20 min通过肝细胞表面的OATP B1/B3蛋白受体摄取并通过MRP2蛋白受体转运至胆管[28, 29],广泛应用于临床提高肝脏局灶性及弥漫性病变的检出[28]。肝损伤时肝细胞破坏,OATP、MRP2等转运蛋白受体减少,细胞外大分子物质沉积,导致肝细胞功能及肝组织微循环灌注改变[16],Gd-EOB-DTPA增强成像后LF不同分期的T1弛豫时间存在差异[17]。T1 mapping序列最初应用于心肌病变的诊断,T1 mapping序列包括MOLLI序列、sh MOLLI序列(shortened modified look-locker inversion recovery)、SASHA序列(saturation recovery single-shot acquisition)等,最常用的是MOLLI序列,CHILD等[30]发现MOLLI序列诊断心肌纤维化效能最高。MOLLI是基于look-locker序列的T1 mapping,对MR场强不均匀性及扫描仪影响不敏感,图像采集与分析的一致性高,与既往研究相比,Gd-EOB-DTPA增强前后T1 mapping成像获得T1值更能客观、定量反映肝脏病理变化[31]。多数研究[23, 32, 33]认为,影响T1值因素较多,如肝细胞水肿、炎症、脂肪沉积、细胞外基质沉积、肝脏硬度增高等增加T1值,而铁含量的增加会降低T1值。本研究示T120min能鉴别F0与F1、F2、F3、F4组间,F1与F2、F3、F4组间,F2与F3、F4组间差异,并且与LF分期相关性较高(r=0.770,P<0.001),虽然AUCF0 vs. F1~F4(0.914)、AUCF0 vs. F1~F2(0.852)稍低于T1native(0.932、0.900),但是DeLong检验显示两两比较差异并不显著(Z<1.96,P>0.05);并且AUCF0 vs. F3~F4(0.987)、AUCF1~F2 vs. F3~F4(0.896)高于T1native(0.969、0.665),AUCF1~F2 vs. F3~F4间比较差异性显著。EOB基团对显示肝细胞功能学及血流动力学改变具有较高的特异性,有助于鉴别早期与进展期LF,有助于临床治疗方案选择。计算所得ΔR120min更能客观反映肝胆期前后T1弛豫率的改变,诊断效能(AUC分别为0.837、0.735、0.959、0.857)优于ΔT120min诊断效能(0.644、0.536、0.773、0.751),差异性显著(Z>1.96,P<0.05)。ΔR120min(r=-0.686)与LF分期相关性优于ΔT120min(r=-0.418),由于ΔR120min不易受T1native的影响,并且ΔT120min的AUCF0 vs. F1~F2无统计学意义(P=0.685),推测LF早期肝细胞转运受体蛋白减少、肝窦毛细血管化或未影响EOB基团转运,而且肝脏急性炎症、脂肪变性等也干扰T1值,但不影响肝胆期T1值。

3.3 比较MRE、Gd-EOB-DTPA增强T1 mapping诊断效能

       对比LF不同分期的LS值与T1native、T120min、ΔT120min、ΔR120min值,显示LS对LF分期诊断效能较高,LS值在诊断早期LF(F0 vs. F1~F2)敏感度(86.36%)、特异度(100.00%)高于其他定量值。ΔT120min诊断效能AUCF0 vs. F3~F4(0.773)、AUCF1~F2 vs. F3~F4(0.751)亦相对较佳,与YANG等[34]基于高脂饮食诱导的非酒精性脂肪性肝病家兔模型和LI等[31]评估慢性乙型肝炎患者获得的ΔT120min值的结果一致。ΔT120min受细胞内、外对比剂的双重影响,并且与T1native密切相关,未来仍需进一步探究ΔT120min的影响因素及诊断效能。

3.4 本研究的不足

       本实验研究仍有局限性:(1)2D-GRE-MRE不能覆盖全肝,本研究采用ROI测量法,而不是全肝评估,具有一定的实验误差;(2)肝纤维化晚期导致的铁沉积影响图像的质量;(3)LF组间病理改变存在重叠,水肿、炎症、脂肪、铁沉积等影响;(4)MRE需外置机械波发生器,价格较昂贵,未应用于临床。

4 结论

       本实验对比MRE、Gd-EOB-DTPA增强T1 mapping成像在LF分期早期定量评估的诊断价值中,两者均显示出较好的诊断价值,MRE优于Gd-EOB-DTPA增强T1 mapping。

[1]
ZERUNIAN M, PUCCIARELLI F, MASCI B, et al. Updates on quantitative MRI of diffuse liver disease: a narrative review[J/OL]. Biomed Res Int, 2022, 2022: 1147111 [2023-10-23]. https://pubmed.ncbi.nlm.nih.gov/36619303/. DOI: 10.1155/2022/1147111.
[2]
WELLE C L, OLSON M C, REEDER S B, et al. Magnetic resonance imaging of liver fibrosis, fat, and iron[J]. Radiol Clin North Am, 2022, 60(5): 705-716. DOI: 10.1016/j.rcl.2022.04.003.
[3]
FOWLER K J, VENKATESH S K, OBUCHOWSKI N, et al. Repeatability of MRI biomarkers in nonalcoholic fatty liver disease: the NIMBLE consortium[J/OL]. Radiology, 2023, 309(1): e231092 [2023-10-26]. https://pubmed.ncbi.nlm.nih.gov/37815451/. DOI: 10.1148/radiol.231092.
[4]
NOGAMI A, YONEDA M, IWAKI M, et al. Non-invasive imaging biomarkers for liver steatosis in non-alcoholic fatty liver disease: present and future[J]. Clin Mol Hepatol, 2023, 29(Suppl): S123-S135. DOI: 10.3350/cmh.2022.0357.
[5]
VENKATESH S K, TORBENSON M S. Liver fibrosis quantification[J]. Abdom Radiol (NY), 2022, 47(3): 1032-1052. DOI: 10.1007/s00261-021-03396-y.
[6]
ZHANG C Y, LIU S, YANG M. Treatment of liver fibrosis: past, current, and future[J]. World J Hepatol, 2023, 15(6): 755-774. DOI: 10.4254/wjh.v15.i6.755.
[7]
OZTURK A, OLSON M C, SAMIR A E, et al. Liver fibrosis assessment: MR and US elastography[J]. Abdom Radiol (NY), 2022, 47(9): 3037-3050. DOI: 10.1007/s00261-021-03269-4.
[8]
YAMADA A, KAMAGATA K, HIRATA K, et al. Clinical applications of artificial intelligence in liver imaging[J]. Radiol Med, 2023, 128(6): 655-667. DOI: 10.1007/s11547-023-01638-1.
[9]
TIPIRNENI-SAJJA A, BRASHER S, SHRESTHA U, et al. Quantitative MRI of diffuse liver diseases: techniques and tissue-mimicking phantoms[J]. MAGMA, 2023, 36(4): 529-551. DOI: 10.1007/s10334-022-01053-z.
[10]
ICHIKAWA T, OBA H, OWADA M, et al. Evaluation of the effects of pemafibrate on metabolic dysfunction-associated steatotic liver disease with hypertriglyceridemia using magnetic resonance elastography combined with fibrosis-4 index and the magnetic resonance imaging-aspartate aminotransferase score[J]. JGH Open, 2023, 7(12): 959-965. DOI: 10.1002/jgh3.13012.
[11]
SANYAL A J, CASTERA L, WONG V W S. Noninvasive assessment of liver fibrosis in NAFLD[J]. Clin Gastroenterol Hepatol, 2023, 21(8): 2026-2039. DOI: 10.1016/j.cgh.2023.03.042.
[12]
BREIT H C, VOSSHENRICH J, HEYE T, et al. Assessment of hepatic function employing hepatocyte specific contrast agent concentrations to multifactorially evaluate fibrotic remodeling[J]. Quant Imaging Med Surg, 2023, 13(7): 4284-4294. DOI: 10.21037/qims-22-884.
[13]
SCHEUER P J, MAGGI G. Hepatic fibrosis and collapse: histological distinction by orecin staining[J]. Histopathology, 1980, 4(5): 487-490. DOI: 10.1111/j.1365-2559.1980.tb02943.x.
[14]
MCHUGH M L. Interrater reliability: the kappa statistic[J]. Biochem Med, 2012, 22(3): 276-282.
[15]
KOO T K, LI M Y. A guideline of selecting and reporting intraclass correlation coefficients for reliability research[J]. J Chiropr Med, 2016, 15(2): 155-163. DOI: 10.1016/j.jcm.2016.02.012.
[16]
江锦赵, 邹立秋, 张豪, 等. Gd-EOB-DTPA动态增强MRI定量评估兔肝纤维化分期的实验研究[J]. 磁共振成像, 2021, 12(6): 66-71. DOI: 10.12015/issn.1674-8034.2021.06.013.
JIANG J Z, ZOU L Q, ZHANG H, et al. Quantitative experimental study in a rabbit model of liver fibrosis by DCE-MRI with Gd-EOB-DTPA[J]. Chin J Magn Reson Imag, 2021, 12(6): 66-71. DOI: 10.12015/issn.1674-8034.2021.06.013.
[17]
LI J M, CAO B L, BI X J, et al. Evaluation of liver function in patients with chronic hepatitis B using Gd-EOB-DTPA-enhanced T1 mapping at different acquisition time points: a feasibility study[J]. Radiol Med, 2021, 126(9): 1149-1158. DOI: 10.1007/s11547-021-01382-4.
[18]
ALHARBI F M, BESBES F R, ALMUTAIRI B S, et al. Fusion of magnetic resonance elastography images with computed tomography and magnetic resonance imaging using the human visual system[J/OL]. Cureus, 2023, 15(9): e45109 [2023-12-22]. https://pubmed.ncbi.nlm.nih.gov/37842423/. DOI: 10.7759/cureus.45109.
[19]
EHMAN R L. Magnetic resonance elastography: from invention to standard of care[J]. Abdom Radiol (NY), 2022, 47(9): 3028-3036. DOI: 10.1007/s00261-022-03597-z.
[20]
FEUILLE C, KARI S, PATEL R, et al. Utility and impact of magnetic resonance elastography in the clinical course and management of chronic liver disease[J/OL]. Sci Rep, 2024, 14(1): 1765 [2023-12-22]. https://pubmed.ncbi.nlm.nih.gov/38242967/. DOI: 10.1038/s41598-024-51295-1.
[21]
PEPIN K M, WELLE C L, GUGLIELMO F F, et al. Magnetic resonance elastography of the liver: everything you need to know to get started[J]. Abdom Radiol, 2022, 47(1): 94-114. DOI: 10.1007/s00261-021-03324-0.
[22]
HAN X J, YANG D W, XU H, et al. Superiority of spleen stiffness on two-dimensional magnetic resonance elastography over liver stiffness and serum tests in assessing portal hypertension in chronic liver disease[J]. Quant Imaging Med Surg, 2024, 14(2): 1429-1440. DOI: 10.21037/qims-22-1415.
[23]
ERDEN A, KURU ÖZ D, ADıGÜZEL M, et al. Quantitative liver magnetic resonance imaging: correlation between conventional magnetic resonance imaging, laboratory values, and prognostic indices in Budd-Chiari syndrome[J]. Diagn Interv Radiol, 2023, 29(3): 428-436. DOI: 10.4274/dir.2022.221462.
[24]
CATANIA R, LOPES VENDRAMI C, BOLSTER B D, et al. Intra-patient comparison of 3D and 2D magnetic resonance elastography techniques for assessment of liver stiffness[J]. Abdom Radiol, 2022, 47(3): 998-1008. DOI: 10.1007/s00261-021-03355-7.
[25]
BRADLEY C R, COX E F, PALANIYAPPAN N, et al. Variability of noninvasive MRI and biological markers in compensated cirrhosis: insights for assessing disease progression[J/OL]. Eur Radiol Exp, 2022, 6(1): 52 [2023-12-22]. https://pubmed.ncbi.nlm.nih.gov/36274113/. DOI: 10.1186/s41747-022-00303-y.
[26]
BHUIYAN E H, OZKAYA E, KENNEDY P, et al. Magnetic resonance elastography fornoninvasive detection of liver fibrosis: is there an added value of 3D acquisition?[J]. Abdom Radiol, 2023, 48(11): 3420-3429. DOI: 10.1007/s00261-023-04036-3.
[27]
邹立秋, 张豪, 钟文新, 等. MR弹性成像与钆塞酸二钠动态增强MRI定量评估肝纤维化的实验研究[J]. 中华放射学杂志, 2021, 55(11): 1202-1208. DOI: 10.3760/cma.j.cn112149-20210408-00340.
ZOU L Q, ZHANG H, ZHONG W X, et al. An experimental study of MR elastography and Gd-EOB-DTPA dynamic contrast-enhanced MRI in quantitative evaluation of liver fibrosis[J]. Chin J Radiol, 2021, 55(11): 1202-1208. DOI: 10.3760/cma.j.cn112149-20210408-00340.
[28]
WANG C, SUN W R, WU N, et al. A convenient and reproducible protocol for acquisition of the hepatocyte phase for liver function-impaired patients in gadoxetic acid disodium-enhanced magnetic resonance imaging[J]. Quant Imaging Med Surg, 2024, 14(2): 1904-1915. DOI: 10.21037/qims-23-1147.
[29]
WANG C, YUAN X D, WU N, et al. Optimization of hepatobiliary phase imaging in gadoxetic acid-enhanced magnetic resonance imaging: a narrative review[J]. Quant Imaging Med Surg, 2023, 13(3): 1972-1982. DOI: 10.21037/qims-22-916.
[30]
CHILD N, SUNA G, DABIR D, et al. Comparison of MOLLI, shMOLLLI, and SASHA in discrimination between health and disease and relationship with histologically derived collagen volume fraction[J]. Eur Heart J Cardiovasc Imaging, 2018, 19(7): 768-776. DOI: 10.1093/ehjci/jex309.
[31]
LI Q T, CHEN T Y, SHI N N, et al. Quantitative evaluation of hepatic fibrosis by fibro Scan and Gd-EOB-DTPA-enhanced T1 mapping magnetic resonance imaging in chronic hepatitis B[J]. Abdom Radiol, 2022, 47(2): 684-692. DOI: 10.1007/s00261-021-03300-8.
[32]
AHN J H, YU J S, PARK K S, et al. Effect of hepatic steatosis on native T1 mapping of 3T magnetic resonance imaging in the assessment of T1 values for patients with non-alcoholic fatty liver disease[J]. Magn Reson Imaging, 2021, 80: 1-8. DOI: 10.1016/j.mri.2021.03.015.
[33]
THOMAIDES-BREARS H B, LEPE R, BANERJEE R, et al. Multiparametric MR mapping in clinical decision-making for diffuse liver disease[J]. Abdom Radiol, 2020, 45(11): 3507-3522. DOI: 10.1007/s00261-020-02684-3.
[34]
YANG R, CHEN Z S, PAN J, et al. Non-contrast T1ρ dispersion versus Gd-EOB-DTPA-enhanced T1mapping for the risk stratification of non-alcoholic fatty liver disease in rabbit models[J]. Magn Reson Imaging, 2024, 107: 130-137. DOI: 10.1016/j.mri.2024.01.013.

上一篇 人工智能辅助压缩感知与并行采集技术在肩关节MRI中的对比研究
下一篇 胰腺黏液纤维肉瘤MRI表现一例并文献复习
  
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