分享:
分享到微信朋友圈
X
技术研究
不同肠道准备方式对直肠癌患者超高b值DWI图像质量的影响
郝勇飞 李婉清 赵婉婷 王虹 张广文 冯岩 李延静 张劲松

Cite this article as: HAO Y F, LI W Q, ZHAO W T, et al. Effect on image quality of ultra-high b-values diffusion weighted imaging in patients with rectal cancer by using different bowel preparation methods[J]. Chin J Magn Reson Imaging, 2024, 15(6): 115-122.本文引用格式:郝勇飞, 李婉清, 赵婉婷, 等. 不同肠道准备方式对直肠癌患者超高b值DWI图像质量的影响[J]. 磁共振成像, 2024, 15(6): 115-122. DOI:10.12015/issn.1674-8034.2024.06.017.


[摘要] 目的 探讨直肠癌患者MRI检查前采用不同肠道准备方式(番泻叶肠道准备和单纯限制饮食)对扩散加权成像(diffusion weighted imaging, DWI)超高b值DWIb2000(b=2000 s/mm2)和常规b值DWIb1000(b=1000 s/mm2)图像质量的影响。材料与方法 回顾性分析番泻叶肠道准备组和单纯限制饮食组共203例直肠癌患者的DWI(DWIb1000和DWIb2000)图像质量及相关影响因素。评估组间肠道准备质量指标(包括直肠扩张程度、直肠内容物性质、直肠肠腔气体量、肿瘤层面肠腔气体量)和DWI图像的磁敏感伪影程度(整体伪影、肿瘤区伪影)。采用Mann-Whitney U检验或t检验比较两组间肠道准备质量评分和DWI图像伪影评分的差异。肿瘤区DWIb2000伪影评分≤3分被定义为肿瘤区DWIb2000临床显性伪影,使用单因素和多因素logistic回归模型分析肿瘤区DWIb2000临床显性伪影的相关因素(肠道准备方式、性别、年龄、肿瘤位置、肿瘤环周占比、直肠后系膜厚度)。结果 两名独立观察者间在图像质量指标评估中具有良好一致性[组内相关系数(intra-class correlation coefficient, ICC)均>0.70]。番泻叶肠道准备组的肠内容物性质、直肠肠腔气体量、肿瘤层面气体定量评分均高于单纯限制饮食组(P均<0.05)。番泻叶肠道准备组的DWIb1000和DWIb2000的整体伪影和肿瘤区伪影程度均显著低于单纯限制饮食组(P均<0.05)。在相同的肠道准备方式中,DWIb2000的整体伪影及肿瘤区伪影程度均显著高于DWIb1000(P均<0.01)。多因素logistic回归分析表明肠道准备方式是肿瘤区DWIb2000临床显性伪影的独立相关因素(OR=3.463,95% CI:1.738~6.901,P<0.001)。结论 与单纯限制饮食相比,番泻叶肠道准备可以改善肠道准备质量,减少DWI的磁敏感伪影水平,提高直肠超高b值DWIb2000和常规DWIb1000图像质量。
[Abstract] Objective To investigate the effect of image quality of ultra-high b-values diffusion weighted imaging (DWI) with b=2000 s/mm2 (DWIb2000) and routine DWI with b=1000 s/mm2 (DWIb1000) by different bowel preparation (drinking folium sennae soup or dietary restrictions) in patients with rectal cancer (RC).Materials and Methods The DWI (DWIb1000 and DWIb2000) image quality and related factors were analyzed in 203 patients with RC between the bowel preparation group with folium sennae soup and the dietary restrictions groups. The bowel preparation metrics (rectal dilatation, rectal contents, amount of gas in rectum and amount of gas around the tumor) and magnetic susceptibility artifacts (overall artifacts and artifacts of the tumor area) were independently assessed in the DWIb1000 and DWIb2000 by two radiologists. Between-group comparisons of bowel preparation metrics and DWI artifacts were calculated using Mann-Whitney U test or t-test. Subsequently, DWIb2000 artifact scores of less than or equal to 3 score in the tumor area was defined as clinical relevant artifacts. Univariate and multivariable logistic regression was used to assess related factors (bowel preparation methods, age, gender, tumor location, tumor circumferential percentage and thickness of posterior rectal mesentery) for DWIb2000 clinical relevant artifacts.Results Inter-reader agreement was good [intra-class correlation coefficient (ICC) >0.70]. The rectal contents, amount of gas in rectum and amount of gas around the tumor in the bowel preparation group with folium sennae soup were higher than those of the dietary restriction groups (P<0.05), respectively. The degree of overall artifacts and artifacts of the tumor area (DWIb1000 and DWIb2000) in the bowel preparation group with drinking folium sennae soup were significantly higher than those of the dietary restriction groups (P<0.05), respectively. In the same bowel preparation group, the degree of overall artifacts and artifacts of tumor area in DWIb2000 were significantly higher than those of DWIb1000 (P<0.01), respectively. Multivariable logistic regression analysis demonstrated that the bowel preparation method was an independent related factor for DWIb2000 clinical relevant artifacts in tumor area (OR=3.463, 95% CI: 1.738-6.901, P<0.001).Conclusions Compared with the dietary restriction group, patients with bowel preparation by drinking folium sennae soup can obtain good bowel preparation quality, reduce the DWI artifacts and improve the image quality of DWIb1000 and DWIb2000.
[关键词] 直肠癌;磁共振成像;扩散加权成像;伪影;泻药
[Keywords] rectal cancer;magnetic resonance imaging;diffusion weighted imaging;artifacts;cathartics

郝勇飞 1, 2   李婉清 2   赵婉婷 1, 2   王虹 2   张广文 2   冯岩 2   李延静 1   张劲松 2*  

1 延安大学医学院,延安 716000

2 空军军医大学西京医院放射诊断科,西安 710032

通信作者:张劲松,E-mail:stspine@163.com

作者贡献声明::张劲松设计了本研究的方案,并对稿件重要内容进行了修改,获得了国家自然科学基金资助;郝勇飞起草和撰写稿件,获取、分析和解释本研究的数据;李婉清、赵婉婷、王虹、张广文、冯岩、李延静帮助获取、分析或解释本研究的数据,并对稿件重要内容进行了修改;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: 国家自然科学基金项目 82371918
收稿日期:2024-03-12
接受日期:2024-06-03
中图分类号:R445.2  R735.37 
文献标识码:A
DOI: 10.12015/issn.1674-8034.2024.06.017
本文引用格式:郝勇飞, 李婉清, 赵婉婷, 等. 不同肠道准备方式对直肠癌患者超高b值DWI图像质量的影响[J]. 磁共振成像, 2024, 15(6): 115-122. DOI:10.12015/issn.1674-8034.2024.06.017.

0 引言

       MRI是直肠癌临床精准诊疗的首选检查方式[1, 2]。扩散加权成像(diffusion weighted imaging, DWI)对于直肠癌的精准分期、风险分层、疗效评价、预后评估具有重要的临床价值[3, 4]。近来,由于超高b值DWI可以更敏感地检出肿瘤并揭示组织微观结构的复杂程度[5],已越来越多地应用于盆腔肿瘤(直肠癌、前列腺癌、膀胱癌等)的检查和诊断[6]。研究表明,与常规b值DWIb1000(b=1000 s/mm2)相比,超高b值DWIb2000(b=2000 s/mm2)可以提高直肠肿瘤病变的对比度并进行定量评估[7, 8]。此外,基于超高b值DWI的衍生扩散模型对直肠癌新辅助治疗评估[9]、肿瘤沉积[10]、微卫星不稳定[11]的预测及直肠癌个体化治疗也具有重要临床意义。本课题组在前期研究中发现,基于超高b值DWI(b=0、1700~3500 s/mm2)评估进展期直肠癌预后分层风险效能显著优于常规DWIb1000(P<0.05)[12]。但是,直肠DWI容易受到肠道准备质量指标的影响(如直肠肠腔气体、肠内容物性质、直肠扩张程度),导致局部磁场不均匀引起直肠壁及其肿瘤区域边界失真、重叠以及磁敏感伪影[13],从而可能影响DWI图像对直肠癌的定量分析。目前,北美腹部放射学会(Society of Abdominal Radiology, SAR)中65%的专家推荐使用灌肠或直肠对比剂填充进行直肠MRI检查前准备;而欧洲胃肠与腹部放射学会(European Society of Gastrointestinal and Abdominal Radiology, ESGAR)中有43%专家推荐使用解痉药,14%专家推荐使用微灌肠或灌肠,不推荐常规使用直肠对比剂填充直肠[14]。中华医学会指南推荐直肠MRI检查前可不做肠道准备,检查前限制饮食并空腹扫描即可[15]。因此,对于直肠MRI检查前肠道准备方案尚未达成一致[16],而不同肠道准备方式对超高b值DWIb2000图像质量的影响亦鲜有报道。出于可行性考虑,本文没有采用灌肠或微灌肠的肠道准备方式,而重点对比番泻叶肠道准备和单纯限制饮食方式对超高b值DWIb2000图像质量的影响,同时与常规b值DWIb1000对照,旨在选择出可以提高直肠DWI图像质量的简便易行的直肠准备方式。

1 材料与方法

1.1 研究对象

       本项回顾性研究遵循《赫尔辛基宣言》,经西京医院伦理委员会批准,免除受试者知情同意,批准文号:KY20183004-1。本研究连续性纳入2019年1月至12月(采用番泻叶肠道准备)和2023年1月至12月(采用单纯限制饮食)期间经西京医院病理科证实为直肠癌的231例患者病例。纳入标准:(1)病理确诊为直肠癌患者;(2)检查前尚未接受任何抗肿瘤治疗,如放射治疗、化学治疗、靶向治疗等;(3)具备多参数高分辨率直肠MRI扫描资料,包括常规DWI(b=0、1000 s/mm2);超高b值DWI(b=0、2000 s/mm2);(4)与肠镜检查间隔时间大于3天。排除标准:(1)直肠肿瘤中心最大层面距离肛门1 cm以下,无法准确评估肿瘤区域直肠扩张程度的患者(21例);(2)未按照要求执行肠道准备(番泻叶组未饮用番泻叶溶液、限制饮食组限制饮食期间进食)、MRI图像质量不合格(大量粪渣或大便所导致MRI图像严重的形变或失真等)或扫描序列不全的患者(共6例);(3)接受子宫或前列腺切除、人工髋关节置入等盆腔手术患者(1例)。最终,经纳排标准筛选后,共计纳入患者203例,其中番泻叶肠道准备组107例,单纯限制饮食组96例。

1.2 直肠MRI检查前准备

       本研究中,番泻叶肠道准备组,在检查前一天晚餐后禁食,晚8点后服用番泻叶溶液(10 g番泻叶+500 mL沸水浸泡30 min后饮用),记录患者排便情况,要求排便次数≥2次。单纯限制饮食组参照中华医学会指南推荐,直肠MRI检查与直肠腔内超声、肠镜检查时间间隔大于3天,目的是防止其他肠道检查后导致直肠激惹而加重DWI的运动伪影;检查前一天晚餐后,晚8点后限制饮食[15]。两组患者均未使用抗肠道痉挛药(口服、静脉或肌注)、直肠超声耦合剂填充、静脉造影等其他肠道准备方案。

1.3 直肠MRI扫描设备及其参数

       本研究所有患者直肠MRI扫描均采用GE Discovery MR 750 3.0 T磁共振仪,8通道腹部相控阵线圈。患者仰卧位,定位中心点为双侧髂前上棘连线中心与耻骨联合连线的中点。两组患者扫描序列及参数均一致,采用轴位T1WI平扫,矢状位及冠状位T2WI,斜轴位FRFSE Small FOV T2WI,常规b值DWI(b=0、1000 s/mm2),超高b值DWI(b=0,2000 s/mm2),具体扫描参数如下:T1WI序列,TR 791.0 ms,TE 7.8 ms,矩阵416×128,层厚4.0 mm,层间距0.4 mm;斜轴位FRFSE Small FOV T2WI序列,TR 5964 ms,TE 130 ms,矩阵352×128,FOV 220 mm×220 mm,层厚3.0 mm,层间距0.3 mm;DWIb1000序列,TR 4607 ms,TE 58.4 ms,矩阵128×128,FOV 400 mm×320 mm,层厚5.0 mm,层间距0.5 mm,激励次数为4;DWIb2000序列,TR 4607 ms,TE 78.4 ms,矩阵128×128,FOV 400 mm×320 mm,层厚5.0 mm,层间距0.5 mm,激励次数为6。

1.4 肠道准备质量指标及其直肠DWI图像伪影的评分准则

       本研究分别由高年资(主治医师,13年腹部影像诊断经验)和低年资(住院医师,2年腹部影像诊断经验)影像诊断医师盲法独立评估直肠MRI检查肠道准备质量指标(直肠肠腔气体量、直肠肿瘤层面气体、肠内容物性质、直肠扩张程度)和不同b值DWI(DWIb1000,DWIb2000)图像质量。当存在分歧时,与具有20年影像诊断经验的主任医师共同商议决定。为使图像质量评分标准化,另外选择20例患者图像由两位医师共同阅读并达成一致。参照既往文献,纳入5个评估指标[17, 18, 19],其中4个指标采用5分likert量表评估:

       (1)直肠肠腔气体定量值[18, 19],在T2WI图像中,气体所在直肠层数/整体直肠层数:1分≥4/5,2分<4/5,3分<3/5,4分<2/5,5分<1/5。

       (2)直肠肿瘤层面气体定量值[18, 19],在T2WI图像中,气体所在直肠肿瘤层面数/整体肿瘤层数:1分≥4/5,2分<4/5,3分<3/5,4分<2/5,5分<1/5。

       (3)肠内容物性质[18, 19],在T2WI图像中,1分=(充满型气体>80%,图1A),2分=(团状气体,图1B),3分=(点状气体,图1C),4分=(气体+液体混合物,图1D),5分=(单一液体,图1E)。

       (4)直肠扩张程度,在T2WI中,直肠肿瘤层面最大面积/直肠坐骨结节层面的面积。

       (5)将T2WI和ADC图(b=0、1000 s/mm2)作为参照,评估直肠整体和直肠肿瘤DWIb1000和DWIb2000磁敏感伪影程度[17],具体评分可见图2:1分=广泛伪影[≥3/4直肠壁或肿瘤扭曲失真,肿瘤边界和细节特征不能识别,其他解剖区域(膀胱、前列腺、精囊腺、子宫)广泛失真和伪影],2分=重度伪影(1/2~3/4直肠壁或肿瘤扭曲失真,肿瘤边界和细节大部分不能识别,其他解剖区域大量失真和伪影),3分=中度伪影(1/4~1/2直肠壁或肿瘤扭曲失真,肿瘤边界和细节少部分不能识别,其他解剖区域中等程度失真和伪影),4分=轻微伪影(≤1/4直肠壁或肿瘤区失真和伪影,肿瘤边界和细节特征基本显示良好,其他解剖区域少许失真和伪影),5分=无伪影(直肠壁或肿瘤区无失真和伪影,清晰显示肿瘤边界及其细节特征,其他解剖区域无失真和伪影)。

图1  在斜轴位T2WI中对直肠内容物性质评分。1A~1E:肠内容物性质分别为充满型气体(1分;1A)、团状气体(2分;1B)、点状气体(3分;1C)、气体+液体混合物(4分;1D)、液体混合物(5分;1E)。
Fig. 1  The score of rectal contents is evaluated in T2WI. 1A-1E: The rectal contents are filled gas (1 score; 1A), mass gas (2 score; 1B), dotted gas (3 score; 1C), gas and liquid mixture (4 score; 1D), and liquid mixture (5 score; 1E), respectively.
图2  DWIb1000和DWIb2000图像伪影评分。2A~2D:男,67岁,直肠癌,分期T2N1a,采用番泻叶肠道准备方式。2A:T2WI显示肠内容物性质以单一液体为主,直肠扩张良好;2B~2D:DWI显示直肠壁清晰,肠壁形态规整,其他解剖区未见明显伪影,整体DWIb1000和DWIb2000伪影评分分别为4分和5分。2E~2H:男,71岁,直肠癌,分期T3N1b,采用单纯限制饮食方式。2E:T2WI显示肠内容以团状气体为主,直肠扩张欠佳;2F~2H:整体DWIb1000显示直肠壁模糊,直肠壁前壁与前列腺以及直肠后系膜交界区可见明显线状高信号伪影(箭),且DWIb2000直肠壁明显变形,整体DWIb1000和DWIb2000伪影评分分别为2分和1分。2I~2L:女,65岁,直肠癌,分期T2N0,采用番泻叶肠道准备方式。2I:T2WI显示肠内容以单一液体为主,直肠扩张程度良好;2J:ADC图显示肿瘤区呈不规则稍低信号,边界清楚(绿色ROI区);2K~2L:DWIb1000显示肿瘤为稍高信号,肿瘤形态、边界和细节清晰,肿瘤区DWIb1000未见伪影,DWIb2000显示图像背景信号抑制加重,图像颗粒度增加,肿瘤区对比度增加,外缘有轻微信号缺失(箭),肿瘤区DWIb1000和DWIb2000伪影评分分别为5分和4分。2M~2P:女,59岁,直肠癌,分期T2N0,采用单纯限制饮食方式。2M:T2WI显示肠内容物性质以点状气体与混合液体为主,直肠扩张良好;2N:ADC图显示肿瘤为稍低信号,并清晰勾勒肿瘤的边界(绿色ROI区);2O~2P:DWI显示有团状气体位于肿瘤交界区,与T2WI和ADC对比,DWIb1000显示直肠肿瘤区点状高信号伪影,而DWIb2000显示肿瘤区扭曲,形态失真严重(箭),不能勾勒完整肿瘤边界,肿瘤区DWIb1000和DWIb2000伪影评分分别为3分和2分。DWI为扩散加权成像;ADC:表观扩散系数;ROI:感兴趣区;DWIb1000为扩散加权成像b值=1000 s/mm2;DWIb2000为扩散加权成像b值=2000 s/mm2
Fig. 2  The artifact scores are evaluated in DWIb1000 and DWIb2000. 2A-2D: Male, 67 years old, rectal cancer(T2N1a), received bowel preparation by drinking folium sennae soup. 2A: T2WI shows good rectal dilatation and fluid-filled contents in the rectum; 2B-2D: DWI shows a clear rectal wall with regular shape and no obvious artifacts in other anatomical areas; the overall artifact scores are rated 4 and 5 in DWIb1000 and DWIb2000, respectively. 2E-2H: Male, 71 years old, rectal cancer(T3N1b), used dietary restrictions. 2E: T2WI shows that there is poor rectal dilatation and mass gas in the rectum; 2F-2H: DWI1000 shows the blurred rectal wall and the linear high signal artifact (arrows), and DWIb2000 shows the twisted and distorted rectal wall (arrows) in the prostate and posterior mesorectal area; the overall artifact scores are rated 2 and 1 in DWIb1000 and DWIb2000, respectively. 2I-2L: Female, 65 years old, rectal cancer(T2N0), received bowel preparation by drinking folium sennae soup. 2I: T2WI shows that there is good rectal dilatation and fluid-filled contents in the rectum; 2J: ADC maps shows a slightly low signal in the tumor with clear boundaries (the green ROI area); 2K-2L: DWIb1000 shows the clear features of morphology, boundaries and details in the tumor without artifact; the image background suppression, lesion-to-background contrast and image granularity are increased in DWIb2000, but there is a slight signal loss at tumor outer edge in DWIb2000, artifact scores of the tumor area are rated 5 and 4 in DWIb1000 and DWIb2000, respectively. 2M-2P: Female, 59 years old, rectal cancer(T2N0), used dietary restrictions. 2M: T2WI shows that there is good rectal dilatation and the mixture of gas and liquid in rectum; 2N: ADC maps shows a slightly low signal with clear boundaries (the green ROI area) in tumor; 2O-2P: Compared with T2WI and ADC, DWIb1000 shows the high signal artifact, while DWIb2000 shows a severe distortion (arrow) in tumor, and could not outline the complete tumor boundary, artifact scores of tumor area are rated 3 and 2 in DWIb1000 and DWIb2000, respectively.
图3  番泻叶肠道准备组和单纯限制饮食组肠道准备质量与DWI伪影评分百分位图。DWI为扩散加权成像;DWIb1000为DWI b值=1000 s/mm2;DWIb2000为DWI b值=2000 s/mm2
Fig. 3  Comparison of the percentile distribution chart of bowel preparation quality metrics and DWI artifacts in bowel preparation by drinking folium sennae soup and dietary restriction groups. DWI: diffusion weighted imaging; DWIb1000 is b value = 1000 s/mm2; DWIb2000 is b value = 2000 s/mm2.

1.5 统计学方法

       采用SPSS 26.0软件和Medcalc 20软件对数据进行统计学分析,当P<0.05(双侧)认为差异具有统计学意义。当单因素二元logistic回归分析P值小于0.1的相关因素将进行多因素二元logistic回归分析。采用组内相关系数(intra-class correlation coefficient, ICC)评价图像主观定量评分观察者间的一致性(0.00~0.20为一致性极差;0.21~0.40为一致性较差;0.41~0.60为一致性中等;0.61~0.80为一致性较好;0.81~1.00为一致性好)。对计量资料进行正态性检验(Shapiro-Wilk test检验)和方差齐性检验(Levene检验),符合正态分布则采用均数±标准差(x¯±s)表示,不符合正态分布的定量资料以中位数(上、下四分位数)MQ1,Q3)表示。使用独立样本t检验或Mann-Whitney U进行差异性分析。分类资料采用例数和百分比表示,采用χ2检验处理两组分类变量。对DWIb1000与DWIb2000在相同直肠准备方式下的伪影程度的差值进行正态性检验,随后使用配对Mann-Whitney U比较不同b值间DWI图像质量评分的差异。将高年资医师评估的肿瘤区DWIb2000的伪影评分进行二分类作为主要研究结局,图像伪影评分≤3分定义为肿瘤区DWIb2000临床显性伪影(可能影响医师判读直肠癌定量分析的伪影);进一步采用多因素二元logistic回归(进入法)来评估肿瘤区DWIb2000临床显性伪影相关因素。

2 结果

2.1 患者基线特征

       番泻叶肠道准备组纳入直肠癌患者病例107例,男69例,女38例,年龄(60.88±10.92)岁。单纯限制饮食组纳入96例直肠癌患者,男62例,女34例,年龄(63.33±9.80)岁,两组患者组间年龄、性别、肿瘤位置、肿瘤环周比、直肠后系膜厚度、肿瘤病理T分期等临床特征差异均无统计学意义(P均>0.05),详见表1

表1  患者组间基线特征比较
Tab. 1  Comparison of patient baseline characteristics between bowel preparation group by drinking folium sennae soup and dietary restrictions group

2.2 影像诊断医师观察者间参数评分一致性

       ICC评估采用双向混合模型、绝对一致性、平均度量的方法对各项肠道准备指标评分和磁敏感伪影评分进行分析,2名影像诊断医师在各项参数评分中观察者评价间均具有良好的一致性(ICC值均>0.70),详见表2

表2  两名医师评分一致性
Tab. 2  The score agreement between two physicians

2.3 两组间肠道准备质量指标及常规b值DWIb1000和超高b值DWIb2000图像伪影评分的比较

       在肠道准备质量指标(肠内容物性质、直肠肠腔气体量、直肠肿瘤层面气体量)方面,番泻叶肠道准备组显著优于单纯限制饮食组(P均<0.05;表3图3);番泻叶肠道准备组的直肠扩张程度显著低于限制饮食组(P<0.05)。在整体和肿瘤区DWIb1000和DWIb2000伪影评估中,番泻叶肠道准备的评分均明显优于单纯限制饮食(P均<0.01;表3图3)。番泻叶肠道准备组中肿瘤区DWIb2000临床显性伪影占比显著低于限制饮食组(14.0% vs. 36.4%,χ2值=13.658,P<0.001)。在相同肠道准备方式中,整体及肿瘤区的DWIb2000伪影评分程度均显著高于DWIb1000(P<0.01;表4图3)。

       单因素logistic分析表明,肠道准备方式和直肠后系膜厚度(P<0.1)可能是肿瘤区DWIb2000临床显性伪影的相关因素,其余指标差异均无统计学差异。进一步采用多因素logistic回归(进入法)分析表明,仅有肠道准备方式是肿瘤区DWIb2000临床显性伪影的独立相关因素(OR=3.463,95% CI:1.738~6.901,P<0.001),详见表5

表3  组间肠道准备质量以及DWI伪影评分比较
Tab. 3  Comparison of bowel preparation quality metrics and DWI artifacts between bowel preparation by drinking folium sennae soup and dietary restrictions group
表4  常规b值DWIb1000与超高b值DWIb2000伪影评分比较
Tab. 4  Comparison of artifact scores between DWIb1000 and DWIb2000
表5  肿瘤区DWIb2000图像临床显性伪影的logistic回归模型结果
Tab. 5  Results of logistic regression model based on clinical relevant artifacts of DWIb2000 in the tumor area

3 讨论

       本研究旨在分析番泻叶肠道准备和单纯限制饮食对肠道准备质量指标以及直肠癌DWI(DWIb1000、DWIb2000)图像质量的影响,并探索了肿瘤区DWIb2000产生临床显性伪影的相关危险因素。结果显示,与单纯限制饮食相比,番泻叶肠道准备的直肠内容物主要以液体为主,可显著减少重度直肠扩张、直肠气体量及DWIb1000和DWIb2000图像的磁敏感伪影程度(P均<0.05)。多因素logistic回归表明肠道准备方式是肿瘤区DWIb2000临床显性伪影的相关独立因素(P<0.001),提示检查前采用番泻叶肠道准备可以显著降低DWI图像整体及肿瘤区的伪影程度,提高超高b值DWIb2000的图像质量。

3.1 肠道准备方式对肠道准备质量及常规b值DWIb1000图像质量的影响

       目前,优化直肠DWI图像的两大关键因素是改善成像技术和检查前肠道准备方式[16]。其中,优化直肠DWI技术已有广泛研究[20, 21],而探索适合的直肠MRI检查前肠道准备方案对改善直肠DWI质量更具有经济性和实用性[22]。目前,直肠MRI检查前准备方案的选择依然存在争论,专业协会ESGAR与SAR也尚未达成共识[23]。考虑到患者的依从性和安全性,本研究未采用其他肠道准备方式如灌肠[24]或直肠耦合剂填充[25]。另外,直肠耦合剂填充可能导致过度充盈的直肠壁,会降低初次直肠癌T分期的准确率;特别是上段直肠癌评估需要足量超声耦合剂,可能虚假地将肿瘤推至直肠更高处,从而影响直肠肿瘤距离肛门的位置[25, 26]。所以本研究仅分析了番泻叶肠道准备和单纯限制饮食两种肠道准备方式对肠道准备质量指标以及DWI图像质量的影响。结果表明,番泻叶肠道准备组中66.35%的患者肠腔均为单一液体,而单纯限制饮食组中仅有9.37%的患者肠腔是单一液体(P<0.001)。同时,与单纯限制饮食组相比,番泻叶肠道准备组可显著改善肠道准备质量并减少肿瘤区DWIb1000伪影水平(P均<0.05)。其原因可能是番泻叶肠道准备组的肠道准备质量较高(肠腔内容物以液体为主,气体和粪渣少),减少了肠腔气体与肿瘤的交界面,可改善肿瘤局部区域磁场均匀度并减少组织间磁化率差别,进而减少了DWI图像伪影。既往研究表明,与单纯限制饮食组相比,在直肠MRI检查前进行微灌肠肠道准备,新辅助治疗后的直肠DWIb1000磁敏感伪影水平显著降低,但对于治疗前基线扫描DWI图像磁敏感伪影降低差异无统计学意义[17]。其原因可能是基线扫描时肿瘤环周比和肿瘤体积较大,通过微灌肠不足以改变直肠肠腔气体量,因此,微灌肠尚不能减少基线直肠癌DWI伪影水平。这也表明在基线直肠癌DWI检查前可能需要更为充分的肠道准备,检查前30分钟内通过直肠给药的微灌肠方式可能受限于患者个体使用方法及其体位的影响,而导致肠道准备质量的差异性;而番泻叶肠道准备通过标准口服药物,方便易行,肠道准备质量较为稳定,可以较好改善直肠癌DWI伪影水平。

       此外,本研究表明番泻叶肠道准备的整体DWIb1000伪影显著低于单纯限制饮食组(P<0.001),进一步证明番泻叶肠道准备可以改善直肠周围解剖分辨率(膀胱、前列腺、子宫等),改善整体盆腔DWIb1000图像伪影程度。PURYSKO等[18]也发现前列腺MRI检查前通过灌肠的肠道准备方式可以显著改善肠道准备质量指标和DWI图像伪影程度。同时,肠腔内大量气体和直肠重度扩张所引起的DWI伪影会显著影响前列腺外周区肿瘤定量评估[19, 27],且肠腔气体量与DWI伪影程度呈正相关[18]。所以,稳定且有效的肠道准备,对于改善盆腔常规DWIb1000成像质量同样具有重要临床价值。

3.2 肠道准备方式对直肠超高b值DWIb2000图像质量的影响

       研究表明,相比直肠常规DWIb1000,超高b值DWI组织背景信号抑制明显,可以减少T2穿透效应,增加直肠肿瘤的对比度[7, 8, 28, 29, 30],提高直肠肿瘤病理分化程度的预测效能[31]。另外,超高b值DWI是构建衍生扩散模型,如拉伸指数模型(stretched exponential model, SEM)[12, 32]、扩散峰度成像(diffusion kurtosis imaging, DKI)[9, 33]的基础序列图像,并且衍生扩散模型对于直肠癌的风险分层、新辅助治疗疗效评估也具有重要价值。由于磁敏感伪影常发生在空气—组织交界面[13],直肠中团状气体与肿瘤界面易导致肿瘤形态及边界失真、模糊、肿瘤细节特征丢失(图2P),但目前直肠超高b值DWI的磁敏感伪影表现与肠道准备间的关系尚未见报道。同样地,本研究亦发现番泻叶肠道准备组整体及肿瘤区DWIb2000伪影程度显著低于单纯限制饮食组(P<0.01);且番泻叶肠道准备组中肿瘤区DWIb2000临床显性伪影占比显著低于单纯限制饮食组(14.0% vs. 36.4%;P<0.001)。其原因可能是与上述番泻叶肠道准备组的肠道准备质量较高有关。此外,本研究发现在相同的直肠MRI检查前准备方案中,与常规肿瘤区DWIb1000相比,肿瘤区DWIb2000图像平均伪影程度会加重(P均<0.01),这表明DWIb2000随着b值的增加,扩散权重加大,信噪比减低,扫描时间加长,局部磁场均匀度可能进一步失衡,导致DWIb2000肿瘤区域伪影加重(图2P)。因此,超高b值DWI图像的扫描参数有待进一步优化,以提高信噪比、减少图像伪影程度、减少扫描时间等。

       多因素logistic回归分析表明,仅有肠道准备方式是影响肿瘤区DWIb2000临床显性伪影的独立相关因素(P<0.001),且单纯限制饮食患者出现肿瘤区DWIb2000临床显性伪影的可能性是番泻叶肠道准备患者的3.463倍(OR=3.463)。因此,肠道准备方案是直肠DWIb2000肿瘤区图像质量的重要保障,特别是目前基于超高b值DWI及其扩散衍生参数(α、DDC、MD、MK)的多模态影像组学和深度学习的预测模型十分依赖精准的肿瘤感兴趣区,而超高b值DWI临床显性伪影会显著影响医师对肿瘤感兴趣区的辨识和精准分割;肿瘤区超高b值DWI临床显性伪影的改善可以保证肿瘤区定量参数和高阶特征的可靠性和重现性[34, 35]。故本研究推荐直肠超高b值DWI检查前常规进行番泻叶肠道准备可以减少肿瘤区DWI伪影程度,这有利于超高b值DWI及其衍生扩散模型(SEM、DKI)对直肠癌预后风险分层、疗效评估的精确评估。

3.3 研究的局限性

       本研究尚有不足之处:(1)由于肠道准备质量指标和DWI伪影评估属于等级资料,且目前缺乏评价图像伪影的定量参数,不可避免存在主观判断。但本研究开始时对两位诊断医师进行联合培训,并随后进行盲法独立评估,两位诊断医师观察者间一致性良好(ICC值均>0.70),可保证该研究的可重复性。(2)本研究未评估DWI图像运动伪影,主要原因是直肠运动伪影相对较少,另外运动伪影的影响因素复杂,受限于个体生理运动(肠蠕动、呼吸)以及自主运动(吞咽、体位移动)的影响[36]

4 结论

       综上所述,直肠MRI检查前采用番泻叶肠道准备可以改善肠道准备质量,减少直肠DWI的磁敏感伪影,可为临床常规b值DWIb1000及超高b值DWIb2000及其衍生模型定量评估直肠癌提供良好的图像质量。与其他的肠道准备方案相比较(直肠耦合剂填充、灌肠、直肠导管抽吸等方案),番泻叶肠道准备方法简单易行,肠道准备质量较为稳定,可以推荐为直肠常规b值或超高b值DWI检查前常规肠道准备方法。

[1]
BENSON A B, VENOOK A P, AL-HAWARY M M, et al. Rectal cancer, version 2.2022, NCCN clinical practice guidelines in oncology[J]. J Natl Compr Canc Netw, 2022, 20(10): 1139-1167. DOI: 10.6004/jnccn.2022.0051.
[2]
国家卫生健康委员会医政司, 中华医学会肿瘤学分会. 中国结直肠癌诊疗规范(2023版)[J]. 协和医学杂志, 2023, 14(4): 706-733. DOI: 10.12290/xhyxzz.2023-0315.
Department of Medical Administration of the National Health Commission of the People's Republic of China. Diagnostic and therapeutic criteria for colorectal cancer in China (2023 edition)[J]. Med J Peking Union Med Coll Hosp, 2023, 14(4): 706-733. DOI: 10.12290/xhyxzz.2023-0315.
[3]
HORVAT N, CARLOS TAVARES ROCHA C, CLEMENTE OLIVEIRA B, et al. MRI of rectal cancer: tumor staging, imaging techniques, and management[J]. Radiographics, 2019, 39(2): 367-387. DOI: 10.1148/rg.2019180114.
[4]
杨宏宇, 高旭, 沈秀芝, 等. 常规MRI检查及扩散加权成像在直肠癌术前评估中的应用价值[J]. 磁共振成像, 2021, 12(9): 65-68, 87. DOI: 10.12015/issn.1674-8034.2021.09.015.
YANG H Y, GAO X, SHEN X Z, et al. Application value of conventional MRI examination and diffusion weighted imaging in preoperative evaluation of rectal cancer[J]. Chin J Magn Reson Imag, 2021, 12(9): 65-68, 87. DOI: 10.12015/issn.1674-8034.2021.09.015.
[5]
TANG L, ZHOU X J. Diffusion MRI of cancer: from low to high b-values[J]. J Magn Reson Imaging, 2019, 49(1): 23-40. DOI: 10.1002/jmri.26293.
[6]
DREHER C, KUDER T A, KÖNIG F, et al. Advanced diffusion-weighted abdominal imaging: qualitative and quantitative comparison of high and ultra-high b-values for lesion detection and image quality[J]. Invest Radiol, 2020, 55(5): 285-292. DOI: 10.1097/RLI.0000000000000639.
[7]
DELLI PIZZI A, CAPOSIENA D, MASTRODICASA D, et al. Tumor detectability and conspicuity comparison of standard b1000 and ultrahigh b2000 diffusion-weighted imaging in rectal cancer[J]. Abdom Radiol, 2019, 44(11): 3595-3605. DOI: 10.1007/s00261-019-02177-y.
[8]
HAUSMANN D, LIU J, BUDJAN J, et al. Image Quality Assessment of 2D versus 3D T2WI and Evaluation of Ultra-high b-Value (b=2, 000 mm/s2) DWI for Response Assessment in Rectal Cancer[J]. Anticancer Res, 2018, 38(2): 969-978. DOI: 10.21873/anticanres.12311.
[9]
ZHANG X Y, WANG L, ZHU H T, et al. Predicting rectal cancer response to neoadjuvant chemoradiotherapy using deep learning of diffusion kurtosis MRI[J]. Radiology, 2020, 296(1): 56-64. DOI: 10.1148/radiol.2020190936.
[10]
HONG Y, SONG G S, JIA Y P, et al. Predicting tumor deposits in patients with rectal cancer: using the models of multiple mathematical parameters derived from diffusion-weighted imaging[J/OL]. Eur J Radiol, 2022, 157: 110573 [2024-06-01]. https://pubmed.ncbi.nlm.nih.gov/36347167/. DOI: 10.1016/j.ejrad.2022.110573.
[11]
王莉莉, 雷建凯, 李生虎, 等. 基于磁共振扩散峰度成像的直肠癌微卫星不稳定状态研究[J]. 磁共振成像, 2023, 14(8): 73-78. DOI: 10.12015/issn.1674-8034.2023.08.012.
WANG L L, LEI J K, LI S H, et al. Microsatellite instability of rectal cancer based on magnetic resonance diffusion kurtosis imaging[J]. Chin J Magn Reson Imag, 2023, 14(8): 73-78. DOI: 10.12015/issn.1674-8034.2023.08.012.
[12]
ZHANG G W, XU Z L, ZHENG J Y, et al. Ultra-high b-Value DWI in predicting progression risk of locally advanced rectal cancer: a comparative study with routine DWI[J/OL]. Cancer Imaging, 2023, 23(1): 59 [2024-06-01]. https://pubmed.ncbi.nlm.nih.gov/37308941/. DOI: 10.1186/s40644-023-00582-7.
[13]
BIHAN D L, POUPON C, AMADON A, et al. Artifacts and pitfalls in diffusion MRI[J]. J Magn Reson Imaging, 2006, 24(3): 478-488. DOI: 10.1002/jmri.20683.
[14]
KRDZALIC J, MAAS M, GOLLUB M J, et al. Guidelines for MR imaging in rectal cancer: Europe versus United States[J]. Abdom Radiol, 2019, 44(11): 3498-3507. DOI: 10.1007/s00261-019-02251-5.
[15]
中国医师协会结直肠肿瘤专业委员会诊疗技术专委会, 中华医学会放射学分会腹部学组. 直肠癌MR扫描及结构式报告规范专家共识[J]. 中华放射学杂志, 2021, 55(11): 1121-1127. DOI: 10.3760/cma.j.cn112149-20210518-00490.
Colorectal Cancer Diagnosis and Treatment Technology Professional Committee of Chinese Medical Doctor Association, Abdomen Group Chinese Society of Radiology Chinese Medical Association. Expert consensus on MR examination and structured report of rectal cancer[J]. Chin J Radiol, 2021, 55(11): 1121-1127. DOI: 10.3760/cma.j.cn112149-20210518-00490.
[16]
FRAUM T J, MA J F, JHAVERI K, et al. The optimized rectal cancer MRI protocol: choosing the right sequences, sequence parameters, and preparatory strategies[J]. Abdom Radiol, 2023, 48(9): 2771-2791. DOI: 10.1007/s00261-023-03850-z.
[17]
JAYAPRAKASAM V S, JAVED-TAYYAB S, GANGAI N, et al. Does microenema administration improve the quality of DWI sequences in rectal MRI?[J]. Abdom Radiol, 2021, 46(3): 858-866. DOI: 10.1007/s00261-020-02718-w.
[18]
PURYSKO A S, MIELKE N, BULLEN J, et al. Influence of Enema and dietary restrictions on prostate MR image quality: a multireader study[J]. Acad Radiol, 2022, 29(1): 4-14. DOI: 10.1016/j.acra.2020.10.019.
[19]
REISCHAUER C, CANCELLI T, MALEKZADEH S, et al. How to improve image quality of DWI of the prostate-enema or catheter preparation?[J]. Eur Radiol, 2021, 31(9): 6708-6716. DOI: 10.1007/s00330-021-07842-9.
[20]
INOUE A, TANABE M, IHARA K, et al. Evaluation of diffusion-weighted magnetic resonance imaging of the rectal cancers: comparison between modified reduced field-of-view single-shot echo-planar imaging with tilted two-dimensional radiofrequency excitation pulses and conventional full field-of-view readout-segmented echo-planar imaging[J]. Radiol Med, 2023, 128(10): 1192-1198. DOI: 10.1007/s11547-023-01699-2.
[21]
ZHOU M, GONG T, CHEN M N, et al. High-resolution integrated dynamic shimming diffusion-weighted imaging (DWI) in the assessment of rectal cancer[J]. Eur Radiol, 2023, 33(8): 5769-5778. DOI: 10.1007/s00330-023-09494-3.
[22]
李兆祥, 薛华丹, 秦明伟, 等. 开塞露在直肠癌MRI术前T1和T2分期中的应用[J]. 磁共振成像, 2014, 5(5): 352-357. DOI: 10.3969/j.issn.1674-8034.2014.05.008.
LI Z X, XUE H D, QIN M W, et al. The value of Enema Glycerine applying in preoperative MRI T1 staging and T2 staging of rectal cancer[J]. Chin J Magn Reson Imag, 2014, 5(5): 352-357. DOI: 10.3969/j.issn.1674-8034.2014.05.008.
[23]
GOLLUB M J, LALL C, LALWANI N, et al. Current controversy, confusion, and imprecision in the use and interpretation of rectal MRI[J]. Abdom Radiol, 2019, 44(11): 3549-3558. DOI: 10.1007/s00261-019-01996-3.
[24]
VAN GRIETHUYSEN J J M, BUS E M, HAUPTMANN M, et al. Gas-induced susceptibility artefacts on diffusion-weighted MRI of the rectum at 1.5 T - Effect of applying a micro-enema to improve image quality[J]. Eur J Radiol, 2018, 99: 131-137. DOI: 10.1016/j.ejrad.2017.12.020.
[25]
STIJNS R C, SCHEENEN T W, DE WILT J H, et al. The influence of endorectal filling on rectal cancer staging with MRI[J/OL]. Br J Radiol, 2018, 91(1089): 20180205 [2024-06-01]. https://pubmed.ncbi.nlm.nih.gov/29888969/. DOI: 10.1259/bjr.20180205.
[26]
PALMUCCI S, PICCOLI M, PIANA S, et al. Diffusion MRI for rectal cancer staging: ADC measurements before and after ultrasonographic gel lumen distension[J]. Eur J Radiol, 2017, 86: 119-126. DOI: 10.1016/j.ejrad.2016.11.017.
[27]
SCHMIDT C, HÖTKER A M, MUEHLEMATTER U J, et al. Value of bowel preparation techniques for prostate MRI: a preliminary study[J]. Abdom Radiol, 2021, 46(8): 4002-4013. DOI: 10.1007/s00261-021-03046-3.
[28]
陆志华, 蒋恒, 翁晓燕, 等. 优化选择高和超高b值直肠癌弥散加权成像[J]. 中国医学影像技术, 2020, 36(12): 1858-1862. DOI: 10.13929/j.issn.1003-3289.2020.12.021.
LU Z H, JIANG H, WENG X Y, et al. Optimal selection of high and ultrahigh b value for diffusion weighted imaging in rectal cancer[J]. Chin J Med Imag Technol, 2020, 36(12): 1858-1862. DOI: 10.13929/j.issn.1003-3289.2020.12.021.
[29]
BATES D D B, GOLIA PERNICKA J S, FUQUA J L, et al. Diagnostic accuracy of b800 and b1500 DWI-MRI of the pelvis to detect residual rectal adenocarcinoma: a multi-reader study[J]. Abdom Radiol, 2020, 45(2): 293-300. DOI: 10.1007/s00261-019-02283-x.
[30]
LU Z H, XIA K J, JIANG H, et al. Improved effects of the b-value for 2000sec/mm2 DWI on an accurate qualitative and quantitative assessment of rectal cancer[J]. Arab J Gastroenterol, 2023, 24(4): 230-237. DOI: 10.1016/j.ajg.2023.09.001.
[31]
QIU J X, LIU J, BI Z X, et al. Integrated slice-specific dynamic shimming diffusion weighted imaging (DWI) for rectal Cancer detection and characterization[J/OL]. Cancer Imaging, 2021, 21(1): 32 [2024-06-01]. https://pubmed.ncbi.nlm.nih.gov/33827704/. DOI: 10.1186/s40644-021-00403-9.
[32]
ZHAO L, LIANG M, YANG Y, et al. Value of multiple models of diffusion-weighted imaging for improving the nodal staging of preoperatively node-negative rectal cancer[J]. Abdom Radiol, 2021, 46(10): 4548-4555. DOI: 10.1007/s00261-021-03125-5.
[33]
CHEN W C, WEI Q R, HUANG W K, et al. Combining diffusion kurtosis imaging and clinical data for predicting the extramural venous invasion of rectal adenocarcinoma[J/OL]. Eur J Radiol, 2022, 148: 110155 [2024-06-01]. https://pubmed.ncbi.nlm.nih.gov/35045353/. DOI: 10.1016/j.ejrad.2022.110155.
[34]
KOCAK B, AKINCI D'ANTONOLI T, ATES KUS E, et al. Self-reported checklists and quality scoring tools in radiomics: a meta-research[J]. Eur Radiol, 2024 [2024-02-01]. https://link.springer.com/article/10.1007/. DOI: 10.1007/s00330-023-10487-5.
[35]
中华医学会放射学分会医学影像大数据与人工智能工作委员会, 中国研究型医院学会医学影像与人工智能专业委员会, 中国医师协会临床精准医疗专业委员会, 等. 影像组学多模态MRI量化分析用于直肠癌新辅助治疗后疗效及预后评估的方法学专家共识[J]. 中华放射学杂志, 2022, 56(6): 608-615. DOI: 10.3760/cma.j.cn112149-20211220-01122.
Image Big Data Artificial Intelligence Working Committee of Chinese Society of Radiology Chinese Medical Association, Medical Imaging and Artificial Intelligence Branch of Chinese Research Hospital Association, Clinical Precision Medicine Society of Chinese Medical Doctor Association, et al. Methodological expert consensus on radiomics using quantitative analysis of multimodal MRI for response and prognosis evaluation of neoadjuvant therapy for rectal cancer[J]. Chin J Radiol, 2022, 56(6): 608-615. DOI: 10.3760/cma.j.cn112149-20211220-01122.
[36]
KROMREY M L, TAMADA D, JOHNO H, et al. Reduction of respiratory motion artifacts in gadoxetate-enhanced MR with a deep learning-based filter using convolutional neural network[J]. Eur Radiol, 2020, 30(11): 5923-5932. DOI: 10.1007/s00330-020-07006-1.

上一篇 电针与干细胞治疗大鼠急性周围神经损伤的磁共振扩散张量成像评估
下一篇 自身免疫性胰腺炎患者少见的影像表现
  
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