分享:
分享到微信朋友圈
X
临床研究
三维准连续式动脉自旋标记在诊断鼻咽癌颈部小淋巴结转移瘤中的应用价值
古冬连 刘露 吴敏 金观桥

Cite this article as: GU D L, LIU L, WU M, et al. Application value of three- dimensional pseudocontinuous arterial spin labeling in diagnosing cervical small lymph node metastases in nasopharyngeal carcinoma[J]. Chin J Magn Reson Imaging, 2024, 15(12): 109-115.本文引用格式:古冬连, 刘露, 吴敏, 等. 三维准连续式动脉自旋标记在诊断鼻咽癌颈部小淋巴结转移瘤中的应用价值[J]. 磁共振成像, 2024, 15(12): 109-115. DOI:10.12015/issn.1674-8034.2024.12.016.


[摘要] 目的 探讨三维准连续式动脉自旋标记(three- dimensional pseudocontinuous arterial spin labeling, 3D pCASL)技术在诊断鼻咽癌颈部小淋巴结转移瘤(small metastatic lymph node, SMLN)中的应用价值。材料与方法 回顾性收集63例初诊鼻咽癌病例,治疗前均行常规平扫+增强磁共振成像(magnetic resonance imaging, MRI)、3D pCASL扫描,治疗后行常规平扫+增强MRI。淋巴结的评价依据是我国鼻咽癌临床分期工作委员会制订的淋巴结转移评判标准,并结合随访期间的MRI图像。根据诊断结果将淋巴结分为3组,大淋巴结转移瘤(large metastatic lymph node, LMLN)组(短径>10 mm)、SMLN 组(短径≤10 mm)、良性淋巴结(benign lymph node, BLN)组。比较各组淋巴结最大径、最短径、最短径/最大径、血流量平均值(average blood flow, BFavg)、血流量最小值(minimum blood flow, BFmin)、血流量最大值(maximum blood flow, BFmax)之间的差异。采用受试者工作特征(receiver operating characteristic, ROC)曲线评价各参数对鼻咽癌颈部SMLN的诊断效能。结果 纳入颈部淋巴结323枚,其中SMLN组152枚,LMLN组97枚,BLN组74枚。SMLN、LMLN、BLN组组间最大径、最短径、最短径/最大径、BFavg、BFmax、BFmin差异均有统计学有意义(P<0.001)。LMLN组最大径、最短径、最短径/最大径均大于SMLN、BLN组(P<0.001),SMLN组最短径、最短径/最大径大于BLN组(P=0.010,P<0.001),BLN组BFavg、BFmax、BFmin均小于SMLN、LMLN组(P<0.001)。BLN组与SMLN组最大径、LMLN组与SMLN组BFavg、BFmax、BFmin差异均无统计学意义(P>0.05)。ROC曲线分析显示,最短径、最短径/最大径、BFavg、BFmax、BFmin鉴别鼻咽癌BLN与SMLN的AUC值分别为0.712、0.740、0.952、0.990、0.791,截断值分别为0.55 cm、0.59、39.4 mL/(min·100 g)、58.5 mL/(min·100 g)、25.5 mL/(min·100 g)。结论 3D pCASL可以有效鉴别鼻咽癌颈部SMLN,提高鼻咽癌N分期的准确性。
[Abstract] Objective To investigate the value of three- dimensional pseudocontinuous arterial spin labeling (3D pCASL) in diagnosing cervical small metastatic lymph node (SMLN) in nasopharyngeal carcinoma.Materials and Methods A total of 63 cases of nasopharyngeal carcinoma were collected retrospectively. A combination of routine plain scanning, enhanced magnetic resonance imaging (MRI), and 3D pCASL scanning was conducted prior to treatment. Subsequent follow-ups involved routine plain scanning and enhanced MRI post-treatment. The evaluation of lymph nodes was based on the criteria for assessing lymph node metastasis established by the nasopharyngeal carcinoma Clinical Staging Committee of our country, combined with MRI images during the follow-up period. According to the diagnostic outcomes, the lymph nodes were categorized into three groups: large metastatic lymph node (LMLN) group (short diameter >10 mm), SMLN group (short diameter ≤10 mm), and benign lymph node (BLN) group. The differences among various groups in terms of the maximum diameter, minimum diameter, ratio of minimum diameter to maximum diameter, average blood flow (BFavg), minimum blood flow (BFmin), and maximum blood flow (BFmax) of lymph nodes were compared among all groups. The diagnostic efficacy of various parameters for cervical SMLN in nasopharyngeal carcinoma was evaluated using the receiver operating characteristic (ROC) curve.Results A total of 323 cervical lymph nodes were included, with 152 in the SMLN group, 97 in the LMLN group, and 74 in the BLN group. There were statistically significant differences in maximum diameter, minimum diameter, the ratio of minimum diameter to maximum diameter, BFavg, BFmax, and BFmin among the SMLN, LMLN, and BLN groups (P<0.001). The maximum diameter, minimum diameter, and the ratio of minimum diameter to maximum diameter in the LMLN group were all greater than those in the SMLN and BLN groups (P<0.001). The shortest diameter and the ratio of shortest diameter to maximum diameter in the SMLN group were both greater than those in the BLN group (P=0.010, P<0.001). The BFavg, BFmax, and BFmin in the BLN group were all lower than those in the SMLN and LMLN groups (P<0.001). There were no statistically significant differences in maximum diameter between the BLN and SMLN groups, as well as in BFavg, BFmax, and BFmin between the LMLN and SMLN groups (P>0.05). ROC analysis revealed that the area under the curve (AUC) values for differentiating BLN from SMLN in nasopharyngeal carcinoma using minimum diameter, the ratio of minimum diameter to maximum diameter, BFavg, BFmax, and BFmin were 0.712, 0.740, 0.952, 0.990, and 0.791, respectively. The corresponding cut-off values were 0.55 cm, 0.59, 39.4 mL/(min·100 g), 58.5 mL/(min·100 g), and 25.5 mL/(min·100 g), respectively.Conclusions 3D pCASL can effectively differentiate cervical SMLN in nasopharyngeal carcinoma, enhancing the accuracy of N staging for nasopharyngeal carcinoma.
[关键词] 鼻咽癌;淋巴结转移瘤;三维准连续式动脉自旋标记;磁共振成像;诊断;应用价值
[Keywords] nasopharyngeal carcinoma;metastatic lymph node;three-dimensional pseudocontinuous arterial spin labeling;magnetic resonance imaging;diagnosis;application value

古冬连    刘露    吴敏    金观桥 *  

广西医科大学附属肿瘤医院医学影像中心,南宁 530021

通信作者:金观桥,E-mail: jinguanqiao77@gxmu.edu.cn

作者贡献声明:金观桥设计本研究的方案,对稿件的重要内容进行了修改,并获得了广西自然科学基金项目的资助;古冬连起草和撰写稿件,获取并分析本研究的数据;刘露、吴敏获取、分析、解释本研究的数据,对稿件的重要内容进行了修改;古冬连、刘露、吴敏获得广西壮族自治区卫生健康委员会自筹经费科研课题资助。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: 广西自然科学基金项目 2023GXNSFAA026225 广西壮族自治区卫生健康委员会自筹经费科研课题 Z-A20240701
收稿日期:2024-08-29
接受日期:2024-12-10
中图分类号:R445.2  R739.62 
文献标识码:A
DOI: 10.12015/issn.1674-8034.2024.12.016
本文引用格式:古冬连, 刘露, 吴敏, 等. 三维准连续式动脉自旋标记在诊断鼻咽癌颈部小淋巴结转移瘤中的应用价值[J]. 磁共振成像, 2024, 15(12): 109-115. DOI:10.12015/issn.1674-8034.2024.12.016.

0 引言

       鼻咽癌是东亚和东南亚地区一种常见的恶性肿瘤,超过70%的新发病例出现在这一区域[1, 2]。初诊鼻咽癌患者颈部淋巴结转移率高达85%~86.4%[3]。随着调强放疗及综合治疗的应用,鼻咽癌总体生存率及局部控制率进一步提高,但鼻咽癌首程治疗失败率约为24.0%~29.5%,主要失败原因是局部区域复发和远处转移[4, 5, 6]。在放疗计划制订过程中,部分假阴性淋巴结(主要是短径<10 mm的小淋巴结)被划分到中剂量区或低剂量区,这是鼻咽癌放疗后颈部淋巴结区域性复发的主要原因[7]。针吸活检有助于诊断淋巴结转移。然而,鼻咽癌颈部转移淋巴结数目多,并且咽后淋巴结解剖位置较深且手术分离复杂,临床实践中对淋巴结进行活检和手术切除较为困难[8]。因此,影像学方法在诊断淋巴结病变中扮演着重要角色。传统磁共振成像(magnetic resonance imaging, MRI)具有软组织分辨率高的优势,已被广泛用于评估淋巴结的大小,淋巴结大小是区分良性和恶性淋巴结最常使用的标准[9]。然而,经常会出现假阳性或假阴性结果,因为大的淋巴结可能是反应性的而非转移性的,而小的淋巴结也可能包含转移病灶[10]。有文献报道25%短径<10 mm的颈部淋巴结可能存在病理上的癌浸润[11, 12]。美国癌症联合委员会的肿瘤-淋巴结-转移(tumor node metastasis, TNM)分期系统和美国国立综合癌症网络(national comprehensive cancer network, NCCN)发布的国际指南,对淋巴结进行更为详细的MRI评估,包括中心坏死和包膜外侵犯等特征[13, 14]。中心坏死对淋巴结转移的诊断特异性为100%,然而其发生率相对较低(19.0%)[10],且恶性小淋巴结内的小坏死区在常规MRI中通常难以观察[11, 12]。因此,常规MRI在鉴别鼻咽癌颈部小淋巴结转移瘤(small metastatic lymph node, SMLN)与良性淋巴结(benign lymph node, BLN)存在一定局限性。

       三维准连续式动脉自旋标记(three- dimensional pseudocontinuous arterial spin labeling, 3D pCASL)是一种利用射频标记的自体动脉血作为内源性示踪剂获得灌注参数血流量(blood flow, BF)的成像技术[15]。该技术无须使用钆对比剂,可避免肾源性系统性纤维化等不良反应,并且具备高可重复性、高空间分辨率和高信噪比。3D pCASL在反映组织血流量方面表现出色,与动态对比增强MRI和微血管密度有良好的一致性[16, 17]。根据既往文献报道,3D pCASL主要应用于中枢神经系统病变的研究[18, 19, 20]。关于鼻咽癌,3D pCASL的研究主要集中在原发肿瘤灶,并且大多应用于肿瘤的分期和疗效预测等方面[21, 22, 23]。关于鼻咽癌颈部SMLN的诊断鲜见报道,并且目前的研究确实存在一些局限性。YU等[3]仅纳入咽后淋巴结作为研究对象,没有全面评估颈部其他区域的淋巴结,SHAO等[24]利用3D pCASL评估鼻咽癌上颈部淋巴结转移,没有明确区分大淋巴结转移瘤(large metastatic lymph node, LMLN)与SMLN,并且排除短径小于5 mm淋巴结,这种选择性排除可能忽略了一部分具有临床意义的SMLN。鉴于现有研究的局限性,本研究旨在探讨3D pCASL在诊断鼻咽癌颈部SMLN中的应用价值,为诊断鼻咽癌颈部SMLN提供一种新的方法,有助于临床医生更准确地分期和制订治疗方案。

1 材料与方法

1.1 研究对象

       回顾性收集2019年10月至2020年10月广西医科大学附属肿瘤医院收治的鼻咽癌初诊病例。纳入标准:(1)经鼻咽镜活检病理确诊为鼻咽癌;(2)治疗前均行鼻咽+颈部MRI平扫+增强、3D pCASL扫描;(3)治疗后均每三个月随访并行鼻咽+颈部MRI平扫+增强扫描至治疗后一年以上。排除标准:(1)MRI检查前已经接受任何形式的治疗;(2)合并其他恶性肿瘤;(3)图像伪影大。本研究遵守《赫尔辛基宣言》,经广西壮族自治区肿瘤防治研究所伦理委员会批准,免除受试者知情同意,批准文号:KY2024037。

       所有患者的治疗方案符合中国临床肿瘤学会指南,治疗方案为诱导化疗+同步放化疗或同步放化疗或单纯放疗,采用适形调强放疗。根据TNM分期,N0期咽后区放疗剂量60 Gy,Ⅱ、Ⅲ、Ⅳa区放疗剂量54~56 Gy。N 1~3期临床诊断阳性淋巴结放疗剂量70 Gy,其余颈部(包括咽后、Ⅱ~Ⅴb区)放疗剂量62~66 Gy。Ⅰ区除了有阳性淋巴结及有预防照射指征外,不照射。采用33次的分割方式,1次/日,5次/周。淋巴结的评价依据是我国鼻咽癌临床分期工作委员会制订的淋巴结转移评判标准,并结合随访期间的MRI图像。若首次随访MRI显示,淋巴结与治疗前相比显著缩小(缩小幅度≥30%),或在放疗后大小保持稳定,但随后的随访MRI显示有进展,则诊断为转移淋巴结,若淋巴结在治疗前及治疗后的随访期间保持相同大小,则诊断为BLN[10, 25]。两名具有5年以上诊断经验的放射科医师(住院医师,主治医师)(意见不一致时,通过协商达成一致)根据淋巴结是否发生转移及转移淋巴结短径大小,分为LMLN组(短径>10 mm)、SMLN组(短径≤10 mm)和BLN组。

1.2 MRI扫描方法

       采用3.0 T(GE Discovery 750W, USA)磁共振扫描仪及8通道头颈联合线圈。治疗前均行鼻咽+颈部MRI平扫+增强、3D pCASL扫描,在进行增强扫描之前,先行3D pCASL扫描。治疗后均行鼻咽+颈部MRI平扫+增强扫描。常规MRI扫描参数如下。

       轴位T1WI:TR 568 ms,TE 11.8 ms,FOV 22 cm×22 cm;轴位T2WI-压脂(fat suppression, FS):TR 3897 ms,TE 95 ms,FOV 22 cm×22 cm;冠状位T2WI-最小二乘估算法迭代水脂分离(iterative dixon water-fat separation with echo asymmetry and least-squares estimation, IDEAL):TR 3000 ms,TE 68 ms,FOV 28 cm×28 cm;T1WI-FS增强扫描包括轴位、冠状位、矢状位,TR 426~823 ms,TE 9 ms,FOV 22 cm×22 cm,层厚5 mm,层间距1 mm。

       3D pCASL扫描参数:采用3D螺旋快速自旋回波采集方式,标记平面放置在FOV下方2 cm处,标记延迟时间(post label delay, PLD)1525 ms,TR 4640 ms,TE 10.7 ms,层厚4 mm,层间距0 mm,FOV 22 cm×22 cm,矩阵512×512,激励次数3次,扫描时间4 min 21 s。

1.3 图像处理及数据测量

       评估淋巴结中央坏死或边缘环形强化、包膜外侵犯特征,并从T2WI-FS图像选取淋巴结最大层面测量淋巴结的最大径与最短径,测量3次,取3次测量结果的平均值作为最终测量结果。所有ASL原始数据导入GE-AW4.6工作站,利用Functool-3DASL软件对BF图进行校正及预处理,将BF图与轴位T2WI-FS融合,获得功能与解剖相融合的BF伪彩图。在融合图像上选择淋巴结最大的层面,并参考T1WI增强扫描图像,避开液化或坏死的区域,手动勾画感兴趣区(region of interest, ROI),ROI至少包括淋巴结面积的3/4。软件自动计算出ROI内的以下参数:BF平均值(average blood flow, BFavg)、BF最小值(minimum blood flow, BFmin)、BF最大值(maximum blood flow, BFmax)。两名具有5年以上诊断经验的放射科医师(住院医师和主治医师)通过共同讨论和实时协作确定参数的测量和ROI的勾画,并在此过程中达成一致意见,以确保测量和勾画过程的一致性和准确性。

1.4 统计学分析

       采用SPSS 27.0和MedCalc 22.0进行统计学分析。采用Kolmogorov-Smirnov检验进行数据正态性检验,符合正态分布的定量资料采用单因素方差分析进行组间比较,方差齐采用Scheffe法进行两两比较,用均数±标准差表示,不符合正态分布定量资料采用多样本的Kruskal-Wallis秩和检验进行组间比较,并采用Mann-Whitney U检验进行两两比较,用中位数(上下四分位数)表示。采用受试者工作特征(receiver operating characteristic, ROC)曲线,以分析有差异的参数诊断鼻咽癌颈部SMLN的效能,并采用DeLong检验比较不同参数的ROC曲线下面积(area under the curve, AUC)。以P<0.05为差异有统计学意义。

2 结果

2.1 临床基线资料

       最终本研究纳入63例患者,男48例,女15例,年龄14~73(46.00±10.84)岁,TNM分期T1期5例,T2期23例,T3期25例,T4期10例;N0期6例,N1期28例,N2期10例,N3期19例;M0期61例,M1期2例;临床总分期(第8版UICC/AJCC)Ⅰ期1例,Ⅱ期12例,Ⅲ期22例,Ⅳ期28例;病理类型为非角化未分化癌59例,非角化分化型4例。颈部淋巴结共323枚,其中SMLN 152枚,LMLN 97枚,BLN 74枚,其中SMLN组有中央坏死或边缘环形强化26枚,包膜外侵犯10枚,LMLN组有中央坏死或边缘环形强化57枚,包膜外侵犯48枚,BLN组未见中央坏死或边缘环形强化、包膜外侵犯。323枚颈部淋巴结区域分布情况见表1

表1  323枚颈部淋巴结区域分布情况
Tab. 1  The regional distribution of 323 cervical lymph nodes

2.2 各组参数比较

       SMLN组、LMLN组与BLN组组间最大径、最短径、最短径/最大径、BFavg、BFmax、BFmin差异均有统计学有意义(P<0.001)(表2)。LMLN组最大径、最短径、最短径/最大径均大于SMLN、BLN组(P<0.001),SMLN组最短径、最短径/最大径大于BLN组(P=0.010,P<0.001),BLN组BFavg、BFmax、BFmin小于SMLN、LMLN组(P<0.001)。BLN组与SMLN组最大径、LMLN组与SMLN组BFavg、BFmax、BFmin差异均无统计学意义(P>0.05)。LMLN、SMLN、BLN组淋巴结示例图像及测量淋巴结BF值ROI见图1, 2, 3

图1  男,64岁,鼻咽未分化型非角化性癌,Ⅱ期(T2N1M0),左侧颈部Ⅱ区LMLN(红箭)。1A:T1WI;1B:T2WI;1C:T1WI增强,1D:ASL/T2WI-FS 融合图,LMLN的ROI由红圈标示,其中颜色编码显示BF值的相对高低,红色表示高BF值区,黄色表示中等BF值区,绿色表示稍低BF值区,ROI内BFavg、BFmin、BFmax分别为90.138 mL/(min·100 g)、28 mL/(min·100 g)、154 mL/(min·100 g)。LMLN:大淋巴结转移瘤;ASL:动脉自旋标记;ROI:感兴趣区;BF:血流量;BFavg:血流量均值;BFmin:血流量最小值;BFmax:血流量最大值。
Fig. 1  A 64-year-old male with Stage Ⅱ (T2N1M0) undifferentiated non-keratinizing of the nasopharynx carcinoma, exhibiting a left cervical level Ⅱ LMLN as indicated by the red arrow. The sequences include T1WI (1A), T2WI (1B), T1WI with enhancement (1C), and an ASL/T2WI-FS fusion image (1D). The ROI of LMLN is marked by a red circle, with color coding indicating the relative levels of BF values, where red represents high BF value areas, yellow indicates medium BF value areas, and green indicates slightly lower BF value areas. The BFavg, BFmin, and BFmax within the ROI are 90.138 mL/(min·100 g), 28 mL/(min·100 g), and 154 mL/(min·100 g), respectively. LMLN: large metastatic lymph node; ASL: arterial spin labeling; ROI: region of interest; BF: blood flow; BFavg: average blood flow; BFmin: minimum blood flow; BFmax: maximum blood flow.
图2  男,30岁,鼻咽非角化性未分化型癌,Ⅲ期(T3N1M0),左侧颈部Ⅱ区SMLN(红箭)。2A:T1WI;2B:T2WI;2C:T1WI增强,2D:ASL/T2WI-FS 融合图,SMLN的ROI由红圈标示,其中颜色编码显示BF值的相对高低,黄色表示中等BF值区,绿色表示稍低BF值区,ROI内BFavg、BFmin、BFmax分别为55.5 mL/(min·100 g)、28 mL/(min·100 g)、83 mL/(min·100 g)。SMLN:小淋巴结转移瘤;ASL:动脉自旋标记;ROI:感兴趣区;BF:血流量;BFavg:血流量均值;BFmin:血流量最小值;BFmax:血流量最大值。
Fig. 2  A 30-year-old male with stage Ⅲ (T3N1M0) undifferentiated non-keratinizing of the nasopharynx carcinoma, exhibiting a left cervical level Ⅱ SMLN as indicated by the red arrow. The sequences are T1WI (2A), T2WI (2B), T1WI with enhancement (2C), and an ASL/T2WI-FS fusion image (2D). The ROI of SMLN is marked by a red circle, with color coding indicating the relative levels of BF values, where yellow represents medium BF value areas, and green indicates slightly lower BF value areas. The BFavg, BFmin, and BFmax within the ROI are 55.5 mL/(min·100 g), 28 mL/(min·100 g), and 83 mL/(min·100 g), respectively. SMLN: small metastatic lymph node; ASL: arterial spin labeling; ROI: region of interest; BF: blood flow; BFavg: average blood flow; BFmin: minimum blood flow; BFmax: maximum blood flow.
图3  女,41岁,鼻咽非角化性未分化型癌,Ⅲ期(T3N1M0),右侧颈部Ⅱ区BLN(红箭)。3A:T1WI;3B:T2WI;3C:T1WI增强,3D:ASL/T2WI-FS 融合图,BLN的ROI由红圈标示,其中颜色编码显示BF值的相对高低,浅蓝色表示低BF值区,ROI内BFavg、BFmin、BFmax分别为22.8 mL/(min·100 g)、16 mL/(min·100 g)、36 mL/(min·100 g)。BLN:良性淋巴结;ASL:动脉自旋标记;ROI:感兴趣区;BF:血流量;BFavg:血流量均值;BFmin:血流量最小值;BFmax:血流量最大值。
Fig. 3  A 41-year-old female with Stage Ⅲ (T3N1M0) undifferentiated non-keratinizing of the nasopharynx carcinoma, exhibiting a right cervical level Ⅱ BLN as indicated by the red arrow. The sequences are T1WI (3A), T2WI (3B), T1WI with enhancement (3C), and an ASL/T2WI-FS fusion image (3D). The ROI of BLN is marked by a red circle, with color coding indicating the relative levels of BF values, where light blue represents low BF value areas. The BFavg, BFmin, and BFmax within the ROI are 22.8 mL/(min·100 g), 16 mL/(min·100 g), and 36 mL/(min·100 g), respectively. BLN: benign lymph node; ASL: arterial spin labeling; ROI: region of interest; BF: blood flow; BFavg: average blood flow; BFmin: minimum blood flow; BFmax: maximum blood flow.
表2  各组参数比较结果
Tab. 2  Comparison results of various parameter groups

2.3 最短径、最短径/最大径、3D pCASL各参数诊断效能

       ROC曲线分析显示,BFmax、BFavg诊断鼻咽癌SMLN与BLN时效能表现出色,AUC值分别高达0.990、0.952,以BFmax诊断效能最佳。相比之下,BFmin、最短径和最短径/最大径在诊断鼻咽癌SMLN与BLN时的效能一般,这些参数的AUC值相对较低(表3图4)。DeLong检验显示,BFavg、BFmax、BFmin、最短径以及最短径/最大径参数中,BFavg和BFmax的AUC与其他参数的AUC两两比较,差异均有统计学意义(P均<0.05),BFmin、最短径以及最短径/最大径这三组参数间的AUC两两比较,差异均无统计学意义(P均>0.05)(表4)。

图4  各参数鉴别鼻咽癌颈部BLN与SMLN的ROC曲线。BLN:良性淋巴结;SMLN:小淋巴结转移瘤;ROC:受试者工作特征;AUC:曲线下面积;BFavg:血流量均值;BFmax:血流量最大值;BFmin:血流量最小值。
Fig. 4  ROC curve of each parameter for differentiating nasopharyngeal carcinoma cervical BLN from SMLN. BLN: benign lymph node; SMLN: small metastatic lymph node; ROC: receiver operating characteristic; AUC: area under the curve; BFavg: average blood flow; BFmax: maximum blood flow; BFmin: minimum blood flow.
表3  各参数诊断效能比较
Tab. 3  Comparison of the diagnostic efficacy of various parameters
表4  各参数AUC两两比较
Tab. 4  Pairwise comparison of AUC for various parameters

3 讨论

       本研究在国内首次采用3D pCASL技术评估鼻咽癌颈部SMLN、LMLN以及BLN的灌注信息,并测量了淋巴结的最短径和最大径。研究结果表明,BFmax在鼻咽癌SMLN的诊断中显示出最佳的诊断性能,而最短径、最短径/最大径、BFavg和BFmin也显示出不同程度的价值,这对于提高鼻咽癌颈部SMLN诊断的准确性具有重要意义。

3.1 3D pCASL、最短径、最短径/最大径诊断鼻咽癌颈部SMLN的价值

       PLD是3D pCASL技术中的关键参数,确定最佳的PLD值是3D pCASL在头颈部肿瘤研究中的一个难点。既往张波等[26]采用1525 ms作为PLD,发现3D pCASL在鼻咽癌N分期中的诊断效能非常出色,ROC曲线AUC值0.97,敏感度83%,特异度100%。因此本研究选择的PLD为1525 ms,以期获得最佳的诊断效果。YU等[3]报道鼻咽癌咽后转移淋巴结比非转移淋巴结的BF值更高,对于鼻咽癌患者,当咽后淋巴结短径<5 mm或6 mm,或长径<7 mm时,如果BF>54 mL/(min·100 g),仍然认为淋巴结有转移。SHAO等[24]基于3D pCASL直方图参数鉴别鼻咽癌上颈部转移与非转移淋巴结,发现转移淋巴结BF值的第10百分位数低于非转移淋巴结,两组间BF值的第90百分位数、BFavg和BF值中位数差异无统计学意义。本研究发现SMLN、LMLN在BF图的信号强度与肿瘤原发灶相似,并且高于周围肌肉的信号强度,BLN信号强度与周围肌肉相似或略高,与YU等[3]研究结果一致。本研究结果显示转移淋巴结比非转移淋巴结具有更高的BF值,提示转移淋巴结具有较高的血流灌注量,与YU等[3]、RAZEK等[27]的研究结果一致。推测其原因可能是转移淋巴结细胞增殖速度比BLN快,需要更丰富的血液供应,转移淋巴结与BLN微血管结构存在差异,转移淋巴结含有大量新生的毛细血管,且血管扩大、扭曲、混乱,没有形成小动脉、毛细血管和静脉的有序结构,导致血管通透性及血流速度增大,从而BF值增大[28]。本研究结果与SHAO等[24]研究结果不一致,推测原因可能是获得参数方法不同,SHAO等[24]采用3D pCASL直方图获得BF值,而本研究采用BF图与轴位T2WI-FS融合的BF伪彩图获得BF值。在鼻咽癌颈部的SMLN与LMLN之间,BFavg、BFmax、BFmin的差异均无统计学意义,其原因可能是SMLN、LMLN具有相似的生物学特征。

       本研究中BLN组与SMLN组最大径中位数相同,但差异无统计学意义。SMLN组最短径、最短径/最大径比值大于BLN组,原因可能是良性淋巴结一般呈扁圆形、椭圆形、蚕豆状,而转移淋巴结大多数膨胀性生长,趋向于呈圆形[29]。GUO等[30]利用VX2细胞和蛋黄乳液分别诱导新西兰兔的转移性淋巴结和炎性淋巴结,发现转移性淋巴结最短径/最大径比值大于炎性淋巴结,SONG等[31]发现宫颈癌盆腔转移淋巴结最短径/最大径比值大于炎性淋巴结,研究结果均与本研究结果一致。本研究结果表明最短径、最短径/最大径对鉴别BLN、SMLN具有一定的应用价值。

       TNM分期系统和我国鼻咽癌临床分期工作委员会制订的淋巴结转移评判标准中均包括中心坏死和包膜外侵犯,但本研究中发现SMLN组152枚仅有26枚出现中央坏死或边缘环形强化,10枚出现包膜外侵犯,SMLN的中央坏死和包膜外侵犯发生率较低。因此,除了传统的大小和形态学标准外,还应通过3D pCASL技术评估这些小淋巴结的灌注信息,以提高鼻咽癌SMLN诊断的准确性。

3.2 各参数诊断效能比较

       本研究通过 ROC曲线比较最短径、最短径/最大径、3D pCASL各参数鉴别鼻咽癌颈部BLN与SMLN的诊断效能,发现BFmax、BFavg鉴别鼻咽癌颈部BLN与SMLN具有较高的准确性,BFmin、最短径、最短径/最大径鉴别鼻咽癌颈部BLN与SMLN有一定的准确性。RAZEK等[27]研究发现以BF>39.5或40.5 mL/(min·100 g)(两位测量者测量结果)为截断值区分头颈部鳞状细胞癌转移性淋巴结与反应性淋巴结,AUC值分别为0.94、0.945,敏感度分别为94.1%、74.4%,特异度分别为83.3%、96.8%,本研究中BFavg参数的截断值、诊断效能与RAZEK等[27]研究结果相似。SHAO等[24]用淋巴结最短径鉴别鼻咽癌上颈部转移与BLN,AUC值为0.685,敏感度80.60%,特异度50.49%,本研究中最短径ROC曲线AUC值、敏感度、特异度与SHAO等[24]等研究结果相似。本研究及SHAO等[24]研究结果说明单纯依靠淋巴结大小鉴别BLN与SMLN效能较低,还应综合考虑BF等功能参数,以提高诊断的准确性和可靠性。

3.3 本研究的局限性

       本研究存在一些局限性:(1)BLN与SMLN均来自鼻咽癌患者,采用我国鼻咽癌临床分期工作委员会制订的淋巴结转移评判标准并结合随访期间的MRI图像判断淋巴结是否转移,未经过病理证实,可能会导致假阳性或假阴性的结果;(2)本项回顾性研究的样本量不足,尤其是BLN样本量较小,结果可能会产生偏倚,后续仍需要扩大样本量进一步探讨;(3)未纳入融合成团块状、边界不清的LMLN,未来的研究可以考虑使用更先进的成像技术和分析方法,以包括这些边界不清晰的淋巴结,从而提高研究的全面性和准确性;(4)未能纳入淋巴结的表观弥散系数值作为评估参数,可能会降低对淋巴结良恶性的鉴别能力。

4 结论

       综上所述,3D pCASL技术能评估淋巴结血流灌注信息,有效区分鼻咽癌颈部SMLN与BLN。该技术补充了常规影像学的不足,提高了N分期的准确性,并有望作为一种无创方法,提升鼻咽癌SMLN靶区勾画的精确度。

[1]
JIANG Y T, LIANG Z G, CHEN K H, et al. A dynamic nomogram combining tumor stage and magnetic resonance imaging features to predict the response to induction chemotherapy in locally advanced nasopharyngeal carcinoma[J]. Eur Radiol, 2023, 33(3): 2171-2184. DOI: 10.1007/s00330-022-09201-8.
[2]
ZHANG Y, HU Y X, ZHAO S, et al. The utility of 18F-FDG-PET-CT metabolic parameters in evaluating the primary tumor aggressiveness and lymph node metastasis of nasopharyngeal carcinoma[J/OL]. Clin Med Insights Oncol, 2024, 18: 11795549231225419 [2024-08-28]. https://pubmed.ncbi.nlm.nih.gov/38322667/. DOI: 10.1177/11795549231225419.
[3]
YU X D, YANG F, LIU X, et al. Arterial spin labeling and diffusion-weighted imaging for identification of retropharyngeal lymph nodes in patients with nasopharyngeal carcinoma[J/OL]. Cancer Imaging, 2022, 22(1): 40 [2024-08-28]. https://pubmed.ncbi.nlm.nih.gov/35978445/. DOI: 10.1186/s40644-022-00480-4.
[4]
CHEN Y P, CHAN A T C, LE Q T, et al. Nasopharyngeal carcinoma[J]. Lancet, 2019, 394(10192): 64-80. DOI: 10.1016/S0140-6736(19)30956-0.
[5]
LI W Z, LV X, HU D, et al. Effect of induction chemotherapy with paclitaxel, cisplatin, and capecitabine vs cisplatin and fluorouracil on failure-free survival for patients with stage IVA to IVB nasopharyngeal carcinoma: a multicenter phase 3 randomized clinical trial[J]. JAMA Oncol, 2022, 8(5): 706-714. DOI: 10.1001/jamaoncol.2022.0122.
[6]
ZHAO Q, DONG A N, CUI C Y, et al. MRI-based metastatic nodal number and associated nomogram improve stratification of nasopharyngeal carcinoma patients: potential indications for individual induction chemotherapy[J]. J Magn Reson Imaging, 2023, 57(6): 1790-1802. DOI: 10.1002/jmri.28435.
[7]
刘芬, 方向军, 刘书林. 体素内不相干运动扩散加权成像对鼻咽癌颈部转移性小淋巴结的诊断价值[J]. 中国医学影像学杂志, 2020, 28(10): 766-769, 778. DOI: 10.3969/j.issn.1005-5185.2020.10.011.
LIU F, FANG X J, LIU S L. Intravoxel incoherent motion diffusion-weighted imaging for the diagnosis of small metastatic cervical lymph nodes of nasopharyngeal carcinoma[J]. Chin J Med Imag, 2020, 28(10): 766-769, 778. DOI: 10.3969/j.issn.1005-5185.2020.10.011.
[8]
WANG P, HU S D, WANG X Y, et al. Synthetic MRI in differentiating benign from metastatic retropharyngeal lymph node: combination with diffusion-weighted imaging[J]. Eur Radiol, 2023, 33(1): 152-161. DOI: 10.1007/s00330-022-09027-4.
[9]
CHEN J, LUO J W, HE X, et al. Evaluation of contrast-enhanced computed tomography (CT) and magnetic resonance imaging (MRI) in the detection of retropharyngeal lymph node metastases in nasopharyngeal carcinoma patients[J/OL]. Cancer Manag Res, 2020, 12: 1733-1739 [2024-08-28]. https://pubmed.ncbi.nlm.nih.gov/32210614/. DOI: 10.2147/CMAR.S244034.
[10]
ZHANG G Y, LIU L Z, WEI W H, et al. Radiologic criteria of retropharyngeal lymph node metastasis in nasopharyngeal carcinoma treated with radiation therapy[J]. Radiology, 2010, 255(2): 605-612. DOI: 10.1148/radiol.10090289.
[11]
MACHADO N O, MACHADO L S. Laparoscopic cholecystectomy in the third trimester of pregnancy: report of 3 cases[J]. Surg Laparosc Endosc Percutan Tech, 2009, 19(6): 439-441. DOI: 10.1097/SLE.0b013e3181c30fed.
[12]
唐爽, 林灿洁, 潘志格, 等. MRI表观扩散系数在诊断鼻咽癌颈部小淋巴结转移瘤及勾画放疗靶区中的应用效果[J]. 广西医学, 2018, 40(17): 1949-1951, 1954. DOI: 10.11675/j.issn.0253-4304.2018.17.09.
TANG S, LIN C J, PAN Z G, et al. Effects of MRI apparent diffusion coefficient applied to diagnosing small cervical lymph node metastases of nasopharyngeal carcinoma and delineating radiotherapy target volume[J]. Guangxi Med J, 2018, 40(17): 1949-1951, 1954. DOI: 10.11675/j.issn.0253-4304.2018.17.09.
[13]
XIE H, HUANG W J, LI S L, et al. Radiomics-based lymph nodes prognostic models from three MRI regions in nasopharyngeal carcinoma[J/OL]. Heliyon, 2024, 10(10): e31557 [2024-08-28]. https://pubmed.ncbi.nlm.nih.gov/38803981/. DOI: 10.1016/j.heliyon.2024.e31557.
[14]
AMIN M B, GREENE F L, EDGE S B, et al. The Eighth Edition AJCC Cancer Staging Manual: continuing to build a bridge from a population-based to a more "personalized" approach to cancer staging[J]. CA Cancer J Clin, 2017, 67(2): 93-99. DOI: 10.3322/caac.21388.
[15]
徐慧敏, 刘颖, 霍然, 等. 不同延迟时间动脉自旋标记与CT灌注成像在颈动脉内膜切除术前后的应用对比[J]. 实用放射学杂志, 2023, 39(10): 1694-1697, 1716. DOI: 10.3969/j.issn.1002-1671.2023.10.030.
XU H M, LIU Y , HUO R, et al. A comparative study between arterial spin labeling with different post labeling delay and CT perfusion in pre-and post-operative carotid endarterectomy[J]. J Pract Radiol, 2023, 39(10): 1694-1697, 1716. DOI: 10.3969/j.issn.1002-1671.2023.10.030.
[16]
DANGOULOFF-ROS V, DEROULERS C, FOISSAC F, et al. Arterial spin labeling to predict brain tumor grading in children: correlations between histopathologic vascular density and perfusion MR imaging[J]. Radiology, 2016, 281(2): 553-566. DOI: 10.1148/radiol.2016152228.
[17]
XIAO B H, WANG P G, ZHAO Y R, et al. Nasopharyngeal carcinoma perfusion MRI: comparison of arterial spin labeling and dynamic contrast-enhanced MRI[J/OL]. Medicine, 2020, 99(22): e20503 [2024-08-28]. https://pubmed.ncbi.nlm.nih.gov/32481470/. DOI: 10.1097/MD.0000000000020503.
[18]
王扬, 苏辉, 王宁, 等. 动脉自旋标记成像观察帕金森病患者纹状体-丘脑-皮层环路的代谢网络连接属性[J]. 磁共振成像, 2024, 15(1): 70-75, 81. DOI: 10.12015/issn.1674-8034.2024.01.011.
WANG Y, SU H, WANG N, et al. The observations on the properties of metabolic network connectivity within striatal-thalamo-cortical circuit in patients with Parkinson's diseases by arterial spin labeling imaging[J]. Chin J Magn Reson Imag, 2024, 15(1): 70-75, 81. DOI: 10.12015/issn.1674-8034.2024.01.011.
[19]
YU X J, HONG W P, YE M T, et al. Atypical primary central nervous system lymphoma and glioblastoma: multiparametric differentiation based on non-enhancing volume, apparent diffusion coefficient, and arterial spin labeling[J]. Eur Radiol, 2023, 33(8): 5357-5367. DOI: 10.1007/s00330-023-09681-2.
[20]
吕瑞瑞, 杨治花, 葛鑫, 等. 集成MRI联合三维动脉自旋标记成像鉴别胶质瘤复发和假性进展的初步研究[J]. 磁共振成像, 2022, 13(8): 19-23, 35. DOI: 10.12015/issn.1674-8034.2022.08.004.
LÜ R R, YANG Z H, GE X, et al. Preliminary study of synthetic MRI combined with three-dimensional arterial spin labeling imaging in differentiating recurrence and pseudoprogression of glioma[J]. Chin J Magn Reson Imag, 2022, 13(8): 19-23, 35. DOI: 10.12015/issn.1674-8034.2022.08.004.
[21]
刘腾, 肖磊, 韦波, 等. 单、双指数模型扩散加权成像及动脉自旋标记预测复发性鼻咽癌近期疗效的应用价值[J]. 磁共振成像, 2023, 14(9): 63-69. DOI: 10.12015/issn.1674-8034.2023.09.011.
LIU T, XIAO L, WEI B, et al. Application value of mono- and bi-exponential model diffusion weighted imaging and arterial spin labeling in predicting short-term curative effect of recurrent nasopharyngeal carcinoma[J]. Chin J Magn Reson Imag, 2023, 14(9): 63-69. DOI: 10.12015/issn.1674-8034.2023.09.011.
[22]
廖丽萍, 廖海, 罗宁斌, 等. 体素内不相干运动扩散加权成像与动脉自旋标记技术在鼻咽癌不同分期中的扩散和灌注特征[J]. 放射学实践, 2022, 37(1): 29-34. DOI: 10.13609/j.cnki.1000-0313.2022.01.006.
LIAO L P, LIAO H, LUO N B, et al. Investigating the diffusion and perfusion characteristics of intravoxel incoherent motion diffusion-weigh-ted imaging and arterial spin labeling in different stages of nasopharyngeal carcinoma[J]. Radiol Pract, 2022, 37(1): 29-34. DOI: 10.13609/j.cnki.1000-0313.2022.01.006.
[23]
SUN Z Q, HU S D, GE Y X, et al. Can arterial spin labeling perfusion imaging be used to differentiate nasopharyngeal carcinoma from nasopharyngeal lymphoma?[J]. J Magn Reson Imaging, 2021, 53(4): 1140-1148. DOI: 10.1002/jmri.27451.
[24]
SHAO L, YANG X, SUN Z, et al. Three-dimensional pseudo-continuous arterial spin-labelled perfusion imaging for diagnosing upper cervical lymph node metastasis in patients with nasopharyngeal carcinoma: a whole-node histogram analysis[J/OL]. Clin Radiol, 2024, 79(5): e736-e743 [2024-08-28]. https://pubmed.ncbi.nlm.nih.gov/38341343/. DOI: 10.1016/j.crad.2024.01.017.
[25]
XIAO X T, WU Y S, CHEN Y P, et al. Patterns and prognosis of regional recurrence in nasopharyngeal carcinoma after intensity-modulated radiotherapy[J]. Cancer Med, 2023, 12(2): 1399-1408. DOI: 10.1002/cam4.5020.
[26]
张波, 查云飞, 陈忠强, 等. IVIM与3D pCASL在鼻咽癌临床分期中的对照研究[J]. 磁共振成像, 2017, 8(9): 647-653. DOI: 10.12015/issn.1674-8034.2017.09.002.
ZHANG B, ZHA Y F, CHEN Z Q, et al. Comparative study of IVIM and 3D pCASL in the clinical staging of nasopharyngeal carcinoma[J]. Chin J Magn Reson Imag, 2017, 8(9): 647-653. DOI: 10.12015/issn.1674-8034.2017.09.002.
[27]
RAZEK A A K A, HELMY E. Multi-parametric arterial spin labeling and diffusion-weighted imaging in differentiation of metastatic from reactive lymph nodes in head and neck squamous cell carcinoma[J]. Eur Arch Otorhinolaryngol, 2021, 278(7): 2529-2535. DOI: 10.1007/s00405-020-06390-0.
[28]
ABDEL RAZEK A A K, TALAAT M, EL-SEROUGY L, et al. Clinical applications of arterial spin labeling in brain tumors[J]. J Comput Assist Tomogr, 2019, 43(4): 525-532. DOI: 10.1097/RCT.0000000000000873.
[29]
SUN Z Q, LI J, WANG T, et al. Predicting perigastric lymph node metastasis in gastric cancer with CT perfusion imaging: a prospective analysis[J/OL]. Eur J Radiol, 2020, 122: 108753 [2024-08-28]. https://pubmed.ncbi.nlm.nih.gov/31794892/. DOI: 10.1016/j.ejrad.2019.108753.
[30]
GUO L J, LIU X M, HUA J, et al. Differential detection of metastatic and inflammatory lymph nodes using inflow-based vascular-space-occupancy (iVASO) MR imaging[J/OL]. Magn Reson Imaging, 2022, 85: 128-132 [2024-08-28]. https://pubmed.ncbi.nlm.nih.gov/34687849/. DOI: 10.1016/j.mri.2021.10.035.
[31]
SONG Q L, YU Y Y, ZHANG X M, et al. Value of MRI and diffusion-weighted imaging in diagnosing normal-sized pelvic lymph nodes metastases in patients with cervical cancer[J/OL]. Br J Radiol, 2022, 95(1138): 20200203 [2024-08-28]. https://pubmed.ncbi.nlm.nih.gov/33095657/. DOI: 10.1259/bjr.20200203.

上一篇 基于HRMR-VWI分析颈动脉斑块的特征与Plaque-RADS评分的临床应用价值
下一篇 基于DCE-MRI和DWI瘤内及瘤周的影像组学预测乳腺癌HER-2状态的价值
  
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