马婉玲,赵娓娓,宦怡,魏梦绮,任静,杨勇,张劲松,潘奇,张广文,董军强
(第四军医大学西京医院放射科,陕西西安710032)
IVIM-DW I在胰腺癌和神经内分泌肿瘤诊断和鉴别中的应用价值
马婉玲,赵娓娓,宦怡,魏梦绮,任静,杨勇,张劲松,潘奇,张广文,董军强
(第四军医大学西京医院放射科,陕西西安710032)
目的:探讨IVIM-DWI诊断和鉴别胰腺癌和胰腺神经内分泌肿瘤的最佳量化参数。材料与方法:应用GE Discovery MR750 3.0T磁共振扫描仪对手术证实的22例胰腺癌和17例胰腺神经内分泌肿瘤患者行胰腺多b值DWI。应用IVIM双指数模型测量胰腺癌和胰腺神经内分泌肿瘤与非瘤区胰腺组织的表观扩散系数(ADC)、纯扩散系数(D)、假扩散系数(D*)和灌注分数(ƒ),并应用单因素方差分析进行统计学分析。结果:胰腺癌的ADC和ƒ低于非癌区胰腺组织(0.803×10-3mm2/s vs 0.974×10-3mm2/s;54.527%vs 64.486%),D高于非癌区胰腺组织(0.659×10-3mm2/s vs 0.535×10-3mm2/s),鉴别胰腺癌和非癌区胰腺组织时D诊断效能最高。胰腺神经内分泌肿瘤的ADC、D和ƒ均高于非瘤区胰腺组织(0.933×10-3mm2/s vs 0.753×10-3mm2/s;0.549×10-3mm2/s vs 0.429×10-3mm2/s;67.275%vs 59.655%),鉴别胰腺神经内分泌肿瘤和非瘤区胰腺组织时ADC诊断效能最高。胰腺癌的ADC、D*和ƒ低于胰腺神经内分泌肿瘤(0.803×10-3mm2/s vs 0.933×10-3mm2/s;4.852×10-3mm2/s vs 11.301×10-3mm2/s;54.527%vs 67.275%),鉴别胰腺癌和神经内分泌肿瘤是D*诊断效能最高。胰腺癌的D*明显低于G2~3期胰腺神经内分泌肿瘤(4.852×10-3mm2/s vs 11.937×10-3mm2/s)。结论:IVIM-DWI相关参数(ADC、D*、D、ƒ)可以有效鉴别胰腺癌和胰腺神经内分泌肿瘤与非瘤区胰腺组织,IVIM-DWI是无创性诊断和鉴别胰腺癌和胰腺神经内分泌肿瘤的理想方法之一。
胰腺肿瘤;癌,神经内分泌;磁共振成像,弥散
胰腺是人体第二大消化腺,分泌多种消化酶和激素,同时具有内外分泌功能,是调节人体机能的重要器官之一。在北美,胰腺癌居肿瘤致死疾病的第4位,据文献报道,至2020年,胰腺癌将成为肿瘤中的“第二大杀手”[1]。由于胰腺位于腹膜后,位置较深,早期胰腺癌多无特异性症状和体征,大多数胰腺癌确诊时已处于进展期[2],错过了手术的最佳时机。胰腺神经内分泌肿瘤是第二常见的胰腺肿瘤[3],其生物学行为多样,可以为良性、交界性、恶性[4-5],手术是唯一有效的治疗方法。胰腺神经内分泌肿瘤预后较胰腺癌好,即使处于进展期,它的总体生存期也较胰腺癌长[6]。因此,胰腺癌和胰腺神经内分泌肿瘤的早期诊断和鉴别尤为重要。本研究旨在比较胰腺癌、胰腺神经内分泌肿瘤和非瘤区胰腺组织的IVIMDWI各参数之间的差异,探讨IVIM-DWI诊断和鉴别胰腺癌和胰腺神经内分泌肿瘤的最佳量化参数。
1.1 研究对象
收集2014年5月—2015年11月经手术证实的22例胰腺癌和17例胰腺神经内分泌肿瘤患者的影像学资料。其中胰腺癌患者女9例,男13例,年龄20~76岁,平均55岁;癌灶位于胰头11例,胰颈3例,胰体5例,胰尾3例;主要临床表现为消瘦、消化不良、腹部或腰背部疼痛,其中14例出现黄疸。胰腺神经内分泌肿瘤患者女10例,男7例,年龄24~76岁,平均47.5岁;其中2例有多发病灶共20个瘤灶,位于胰头9个,胰颈3个,胰体4个,胰尾4个;主要临床表现为突发性低血糖7例,无痛性黄疸1例,上腹胀痛不适4例,血小板减少1例,消化道溃疡并出血1例,无任何不适经查体发现3例。
1.2 检查方法
所有患者检查前一日晚饭后口服番泻叶清洁肠道,检查前禁食、禁水4小时。采用GE Discovery MR 750 3.0T磁共振扫描仪,8通道腹部相控阵线圈,所有患者行胰腺常规MRI T1WI、T2WI和多b值DWI扫描。T1WI采用轴位LAVA-Flex,T2WI采用轴位呼吸触发脂肪抑制快速自旋回波序列,多b值DWI应用呼吸触发EPI序列进行轴位扫描,b值为0,10,20,40,60,80,100,150,200,400,800,1 000,1 200,1 500,2 000,3 000,4 000 s/mm2。扫描参数如表1。
表1 MR序列参数
1.3 数据处理及统计学分析
将数据导入GE AW4.6工作站,应用自带IVIM分析软件测量胰腺癌、胰腺神经内分泌肿瘤和周围非瘤区胰腺组织的IVIM-DWI各参数。参照轴位T1WI(LAVA-Flex)和FSE-T2WI选取病灶最大层面,距离病灶边缘1mm处手动选取尽可能大的不规则形感兴趣区(ROI),非瘤区胰腺组织选取圆形ROI,尽量避开出血、坏死及胰管,每个病灶及非瘤区胰腺组织的ADC、D、D*和ƒ测量3次,取平均值作为最终测量值。ROI面积:胰腺癌57~1 015mm2,平均396.05mm2;胰腺神经内分泌肿瘤20~1379mm2,平均246.1mm2。
对胰腺癌、胰腺神经内分泌肿瘤和周围非瘤区胰腺组织的IVIM-DWI各参数进行单因素方差分析,P<0.05具有统计学意义。
22例胰腺癌最大瘤灶大小约4.91 cm×3.68 cm,最小瘤灶大小约1.53 cm×1.31 cm。在轴位抑脂T2WI,20例胰腺癌呈较高信号,其中12例信号不均匀;1例胰腺癌呈略高信号;1例呈等信号;1例呈略低信号。在轴位LAVA-Flex纯水像,13例胰腺癌呈较低信号,其中5例信号不均匀;8例胰腺癌呈等信号,其中5例信号不均匀;1例呈不均匀略高信号。在DWI,20例胰腺癌呈高信号,其中15例(75%)在低b值(b≤200 s/mm2)、中b值(200 s/mm2<b≤1 500 s/mm2)、高b值(b>1 500 s/mm2)均呈高信号,5例仅在低b值和中b值呈高信号,高b值时呈低信号;2例胰腺癌呈略低信号。胰腺癌的影像学表现见图1,2。22例胰腺癌及非癌区胰腺组织的IVIMDWI各参数比较结果见表2。ADC、D和ƒ鉴别胰腺癌和非癌区胰腺组织的诊断效能见ROC曲线图(图3)及曲线分析结果(表3)。
表2 胰腺癌及非癌区胰腺组织的IVIM-DWI各参数比较(×10-3mm2/s)
表3 ADC、D和ƒ鉴别胰腺癌和癌周胰腺组织的ROC分析结果
图1 胰头癌,胰腺体尾部萎缩,胰管扩张。图1a:轴位抑脂T2WI,胰腺癌呈不均匀较高信号,边界较清。图1b:轴位LAVA-Flex水相,胰腺癌呈不均匀较低信号,边界欠清。图1c:DWI,b=200 s/mm2,胰腺癌呈不均匀高信号,边界欠清。图1d:DWI,b=1 500 s/mm2,胰腺癌呈不均匀高信号,边界清晰。图1e:DWI,b=2 000 s/mm2,胰腺癌呈不均匀高信号,边界清晰。图1f:距离病灶边缘1mm选取不规则形ROI。图1g:增强扫描动脉期,胰头癌呈不均匀轻度强化。图1h:增强扫描静脉期,胰头癌呈延迟性不均匀强化,强化程度高于动脉期。Figure 1.Pancreatic cancer located in pancreatic head.Pancreatic body and tail showed atrophy.Pancreatic duct showed dilation.Figure 1a:Pancreatic cancer showed heterogeneous moderate hyperintensity and relatively clear boundary on axial T2WI with fat suppression.Figure 1b:Pancreatic cancer showed heterogeneous moderate hypointensity and obscure boundary on axial LAVA Flex water phase image.Figure 1c:Pancreatic cancer demonstrated heterogeneous hyperintensity and obscure boundary on DWI with b value of 200 s/mm2.Figure 1d:Pancreatic cancer demonstrated heterogeneous hyperintensity and clear boundary on DWI with b value of 1 500 s/mm2.Figure 1e:Pancreatic cancer demonstrated heterogeneous hyperintensity and clear boundary on DWI with b value of 2 000 s/mm2.Figure 1f:The irregular ROI was manually drawn along the edge 1mm away from the tumor margin.Figure 1g:Pancreatic cancer showed heterogeneous mild enhancement on arterial phase imaging.Figure 1h:Pancreatic cancer showed heterogeneous enhancement on venous phase imaging.The enhancement degree was higher than that of arterial phase.
图2 胰头神经内分泌肿瘤(G2期),胰腺体尾部萎缩,胰管扩张。图2a:轴位抑脂T2WI,胰头肿瘤呈不均匀较高信号,边界不清。图2b:轴位LAVA-Flex水相,胰头肿瘤呈不均匀较低信号,边界不清。图2c:DWI,b=200 s/mm2,胰头肿瘤呈不均匀高信号,边界较清。图2d:DWI,b=1 500 s/mm2,胰头肿瘤呈不均匀高信号,边界清晰。图2e:DWI,b=2 000 s/mm2,胰头肿瘤呈不均匀高信号,边界清晰。图2f:距离病灶边缘1mm选取不规则形ROI。图2g:增强扫描动脉期,胰头肿瘤呈不均匀轻度强化。图2h:增强扫描静脉期,胰头肿瘤呈不均匀强化,强化程度高于动脉期。图2a~2e,2g,2h影像学表现与胰腺癌难以鉴别。Figure 2.Pancreatic neuroendocrine tumor located in pancreatic head(Grade 2).Pancreatic body and tail showed atrophy.Pancreatic duct showed dilation.Figure 2a:Pancreatic neuroendocrine tumor showed heterogeneous moderate hyperintensity and obscure boundary on axial T2WI with fat suppression.Figure 2b:Pancreatic neuroendocrine tumor showed heterogeneous moderate hypointensity and obscure boundary on axial LAVA Flex water phase image.Figure 2c:Pancreatic neuroendocrine tumor demonstrated heterogeneous hyperintensity and clear boundary on DWI with b value of 200 s/mm2.Figure 2d:Pancreatic neuroendocrine tumor demonstrated heterogeneous hyperintensity and clear boundary on DWI with b value of 1 500 s/mm2.Figure 2e:Pancreatic neuroendocrine tumor demonstrated heterogeneous hyperintensity and clear boundary on DWI with b value of 2 000 s/mm2.Figure 2f:The irregular ROI was manually drawn along the edge 1mm away from the tumor margin.Figure 2g:Pancreatic neuroendocrine tumor showed heterogeneous mild enhancement on arterial phase imaging.Figure 2h:Pancreatic neuroendocrine tumor showed heterogeneous enhancement on venous phase imaging.The enhancement degree was higher than that of arterial phase.It was difficult to differentiate pancreatic neuroendocrine tumor from pancreatic cancer according to the imaging findings on Figure 2a~2e,2g,2h.
图3 ADC、D和ƒ鉴别胰腺癌和非癌区胰腺组织的ROC曲线。Figure 3.ROC curve of ADC,D andƒfor differentiating pancreatic cancer from non-cancerous pancreatic tissue.
17例胰腺神经内分泌肿瘤共20个瘤灶,最大瘤灶大小约5.58cm×3.34cm,最小瘤灶大小约0.61cm×0.32 cm。在轴位抑脂T2WI,9个胰腺神经内分泌肿瘤呈较高信号,边界清晰,其中2个信号不均匀;6个胰腺神经内分泌肿瘤呈略高信号,其中1个信号不均匀,4个边界不清;5个呈略低信号,其中1个边界不清。在轴位LAVA-Flex水像,18个胰腺神经内分泌肿瘤呈较低信号,其中3例边界不清;2例胰腺神经内分泌肿瘤呈等信号。在DWI,18个胰腺神经内分泌肿瘤呈高信号,其中9个(50%)在低、中、高b值均呈高信号,9个仅在低b值和中b值呈高信号,高b值时呈低信号;1个胰腺神经内分泌肿瘤仅在低b值呈略高信号;1个胰腺神经内分泌肿瘤呈低信号。胰腺神经内分泌肿瘤的影像学表现见图3。胰腺神经内分泌肿瘤和瘤周胰腺组织的IVIM-DWI各参数比较结果见表4。ADC、D和ƒ鉴别胰腺神经内分泌肿瘤和瘤周胰腺组织的诊断效能见ROC曲线图(图4)及曲线分析结果(表5)。G2~3期胰腺神经内分泌肿瘤和瘤周胰腺组织的IVIM-DWI各参数比较结果见表6。
胰腺癌和胰腺神经内分泌肿瘤的IVIM-DWI各参数比较结果见表7。ADC、D*和ƒ鉴别胰腺癌和胰腺神经内分泌肿瘤的诊断效能见ROC曲线图(图5)及曲线分析结果(表8)。
胰腺癌和G2~3期胰腺神经内分泌肿瘤的IVIM-DWI各参数比较结果见表9。
表4 胰腺神经内分泌肿瘤和瘤周胰腺组织的IVIM-DWI各参数比较(×10-3mm2/s)
表5 ADC、D和ƒ鉴别胰腺神经内分泌肿瘤和瘤周胰腺组织的ROC分析结果
图4 ADC、D和ƒ鉴别胰腺神经内分泌肿瘤和非瘤区胰腺组织的ROC曲线。Figure 4.ROC curve of ADC,D andƒfor differentiating pancreatic neuroendocrine tumors from non-tumorous pancreatic tissue.
表6 G2~3期胰腺神经内分泌肿瘤和瘤周胰腺组织的IVIM-DWI各参数比较(×10-3mm2/s)
表7 胰腺癌和胰腺神经内分泌肿瘤的IVIM-DWI各参数比较(×10-3mm2/s)
图5 ADC、D*和ƒ鉴别胰腺癌和神经内分泌肿瘤的ROC曲线。Figure 5.ROC curve of ADC,D*andƒfor differentiating pancreatic cancer from neuroendocrine tumors.
胰腺癌在所有癌症中预后最差,手术切除是最有效的根治方法,但仅不足20%的患者有手术机会,并且术后平均5年生存率不到5%[7]。胰腺癌早期症状不典型,目前尚缺乏高敏感性的早期诊断方法,多数胰腺癌进展期才确诊,贻误了最佳治疗时机。血清肿瘤标志物CA19-9被认为是目前诊断胰腺癌的最好指标[8],但非肿瘤梗阻性胰腺炎患者CA19-9也会升高。细针穿刺病理检查是最接近诊断金标准的检查方法[9],但对于病灶较小的早期胰腺癌患者效果不佳,有创且有一定风险。CT和MRI是目前临床上确诊胰腺癌最常用的方法[10],但常规CT和MRI平扫及增强扫描均在出现解剖学改变才能检出胰腺癌。
表8 ADC、D*和ƒ鉴别胰腺癌和胰腺神经内分泌肿瘤的ROC分析结果
表9 胰腺癌和G2~3期胰腺神经内分泌肿瘤的IVIM-DWI各参数比较(×10-3mm2/s)
胰腺神经内分泌肿瘤是源于胰腺多能神经内分泌干细胞的一类肿瘤,一般分为功能性与无功能性两类。功能性胰腺神经内分泌肿瘤通常较小,以胰岛细胞瘤最常见,可出现低血糖等典型表现,手术是唯一有效的治疗方法。无功能性胰腺神经内分泌肿瘤通常较大,直径常超过2.0 cm,而且病灶越大,其恶性可能性越大[11-12]。有研究[13]表明无功能性胰腺神经内分泌肿瘤生长缓慢、症状不典型,90%的病人就诊时已进展为恶性,部分已经发生转移及周围组织侵犯,延误了最佳手术时机,因此早期诊断胰腺神经内分泌肿瘤尤为重要。
磁共振扩散加权成像(DWI)通过检测人体组织中水分子扩散运动受限的方向和程度,间接反映组织微观结构的变化,从细胞及分子水平研究疾病的病理生理状态,可用于良恶性肿瘤的定性及鉴别诊断[14]。ADC作为一种可以客观评价的量化参数,被广泛应用于良恶性病变的鉴别诊断、肿瘤预后判断及疗效评估等临床研究中。ADC值并不能单纯反映活体组织内水分子扩散,它同时也受毛细血管网血流灌注效应的影响,而后者代表了体素内血管内水分子的宏观运动[15-18],因此,活体组织测得ADC值往往高于扩散系数D。Le Bihan等[15]提出应用IVIM双指数模型对多b值DWI进行分析,可以同时得到灌注相关参数(ƒ,D*)和纯扩散参数(D),可用于量化DWI图像中的两种运动成分。
有研究显示胰腺癌ADC值明显低于正常胰腺[19],而Yoshikawa等[20]的研究表明胰腺癌ADC值高于正常胰腺。上述研究应用单指数模型DWI所得胰腺癌ADC值结论不一,而Lemke等学者[21]应用IVIMDWI研究胰腺癌的结果表明正常胰腺组织和胰腺癌的扩散系数D无统计学差异,胰腺癌和正常胰腺组织ADC值之间的差异是基于灌注分数ƒ的差异,应用IVIM-DWI鉴别胰腺癌和正常胰腺组织时灌注分数ƒ是最佳参数。我们的研究结果表明,胰腺癌和非癌区胰腺组织的ADC、纯扩散系数D、灌注分数ƒ均有统计学差异,IVIM-DWI可以有效地鉴别胰腺癌和非癌区胰腺组织,与Lemke等学者[21]的研究结果一致。
胰腺神经内分泌肿瘤通常表现为富血供,增强后动脉期显著强化[22]。本研究显示胰腺神经内分泌肿瘤的灌注分数ƒ显著高于非瘤区胰腺组织。而胰腺癌富含纤维组织等间质成分,血管分布稀疏,增强后早期强化明显低于正常胰腺[23],通常与富血供胰腺神经内分泌肿瘤容易鉴别。但分化差、级别较高的胰腺神经内分泌肿瘤更具有侵袭性,肿块一般较大且边界不清,动脉期强化低于正常胰腺,常规影像学表现有时与胰腺癌难以鉴别[24]。即使处于进展期的胰腺神经内分泌肿瘤,也比胰腺癌生长缓慢,切除率更高,对化疗更敏感,预后更好[6,25-26],因此,术前准确地鉴别恶性胰腺神经内分泌肿瘤和胰腺癌对评估病人的预后和术后治疗方案的制定提供一定的帮助。本研究显示胰腺癌的假扩散系数D*明显低于G2~3期胰腺神经内分泌肿瘤;以5.6×10-3mm2/s为阈值,鉴别两者的敏感性为90.91%,特异性为62.50%,准确性为83.33%。虽然无统计学差异,但胰腺癌的ADC、ƒ低于G2~3期胰腺神经内分泌肿瘤。与非瘤区胰腺组织相比,胰腺癌的ADC和ƒ明显减低,而G2~3期胰腺神经内分泌肿瘤的ADC、ƒ轻微增高。IVIM-DWI所得参数ADC、D*和ƒ可以有效鉴别恶性胰腺神经内分泌肿瘤和胰腺癌,为临床制定治疗方案提供影像学依据。
综上所述,应用双指数模型分析多b值DWI所得参数ADC、D、D*、ƒ可以鉴别胰腺癌和胰腺神经内分泌肿瘤以及瘤周胰腺组织。IVIM-DWI无创且简单易行,无需注射对比剂就可以同时得到灌注相关参数和扩散参数,对于难以鉴别的胰腺癌和神经内分泌肿瘤,尤其是肾功能较差的患者提供有效的影像学评估方法。
[1]Waghraya M,Yalamanchilia M,Magliano MP,et al.Deciphering the role of stroma in pancreatic cancer[J].Curr Opin Gastroenterol,2013,29(5):537-543.
[2]Warshaw AL,Fernandez-del Castillo C.Pancreatic carcinoma[J].N Engl JMed,1992,326(7):455-465.
[3]Klimstra DS,Pitman MB,Hruban RN.An algorithmic approach to the diagnosis of pancreatic neoplasms[J].Arch Pathol Lab Med,2009,133(3):454-464.
[4]Burns WR,Edil BH.Neuroendocrine pancreatic tumors:guidelines for management and update[J].Curr Treat Options Oncol,2012,13(1):24-34.
[5]王佳,钟燕,王海屹,等.胰腺神经内分泌癌患者的磁共振影像诊断特征分析[J].中华医学杂志,2012,92(7):483-486.
[6]Philips S,Shah SN,Vikram R,et al.Pancreatic endocrine neoplasms:a current update on genetics and imaging[J].Br J Radiol,2012,85(1014):682-696.
[7]Conlon KC,Klimstra DS,Brennan MF.Long-term survival after curative resection for pancreatic ductal adenocarcinoma.Clinicopathologic analysis of 5-year survivors[J].Ann Surg,1996,223(3):273-279.
[8]Goonetilleke KS,Siriwardena AK.Systematic review of carbohydrate antigen(CA 19-9)as a biochemical marker in the diagnosis of pancreatic cancer[J].Eur J Surg Oncol,2007,33(3):266-270.
[9]王洪江,邵本深,赵作伟.术中Tm-cut型针活检对胰腺肿块的诊断价值[J].中国普通外科杂志,2005,14(5):399-400.
[10]Kinney TP,Freeman ML.Recent advances and novel methods in pancreatic imaging[J].Minerva Gastroenterol Dietol,2008,54(1):85-95.
[11]Brenner R,Metens T,Bali M,et al.Pancreatic neuroendocrine tumor:added value of fusion of T2-weighted imaging and high b-value diffusion-weighted imaging for tumor detection[J].Eur J Radiol,2012,81(5):746-749.
[12]Jang KM,Kim SH,Lee SJ,et al.The value of gadoxetic acidenhanced and diffusion-weighted MRI for prediction of grading of pancreatic neuroendocrine tumors[J].Acta Radiol,2014,55(2):140-148.
[13]Schmid-Tannwald C,Schmid-Tannwald CM,Morelli JN,et al.Comparison of abdominal MRI with diffusion-weighted imaging to68Ga-DOTATATE PET/CT in detection of neuroendocrine tumors of the pancreas[J].Eur J Nucl Med Mol Imaging,2013,40(6):897-907.
[14]Koh DM,Collins DJ.Diffusion-weighted MRI in the body:applications and challenges in oncology[J].AJR,2007,188(6):1622-1635.
[15]Le Bihan D,Breton E,Lallemand D,et al.MR imaging of intravoxel incoherent motions:application to diffusion and perfusion in neurologic disorders[J].Radiology,1986,161(2):401-407.
[16]Le Bihan D,Breton E,Lallemand D,et al.Separation of diffusion and perfusion in intravoxel incoherent motion MR imaging[J].Radiology,1988,168(2):497-505.
[17]Le Bihan D,Turner R,Moonen CT,et al.Imaging of diffusion and microcirculation with gradient sensitization:design,strategy,and significance[J].JMagn Reson Imaging,1991,1(1):7-28.
[18]Le Bihan D.IVIM method measures diffusion and perfusion[J].Diagn Imaging(San Franc),1990,12(6):133-136.
[19]Muraoka N,Uematsu H,Kimura H,et al.Apparent diffusion coefficient in pancreatic cancer:characterization and histopathological correlations[J].J Magn Reson Imaging,2008,27(6):1302-1308.
[20]Yoshikawa T,Kawamitsu H,Mitchell DG,et al.ADC measurement of abdominal organs and lesions using parallel imaging technique[J].AJR,2006,187(6):1521-1530.
[21]Lemke A,Laun FB,Klauss M,et al.Differentiation of pancreas carcinoma from healthy pancreatic tissue using multiple b-values:comparison of apparent diffusion coefficient and intravoxel incoherent motion derived parameters[J].Invest Radiol,2009,44(12):769-775.
[22]Semelka RC,Custodio CM,Cem Balci N,et al.Neuroendocrine tumors of the pancreas:spectrum of appearances on MRI[J].J Magn Reson Imaging,2000,11(2):141-148.
[23]Gabata T,Matsui O,Kadoya M,et al.Small pancreatic adenocarcinomas:efficacy of MR imaging with fat suppression and gadolinium enhancement[J].Radiology,1994,193(3):683-688.
[24]Worhunsky DJ,Krampitz GW,Poullos PD,et al.Pancreatic neuroendocrine tumours:hypoenhancement on arterial phase computed tomography predicts biological aggressiveness[J].HPB(Oxford),2014,16(4):304-311.
[25]Kloppel G,Heitz PU.Pancreatic endocrine tumors[J].Pathol Res Pract,1988,183(2):155-168.
[26]Sutliff VE,Doppman JL,Gibril F,et al.Growth of newly diagnosed,untreated metastatic gastrinomas and predictors of growth patterns[J].J Clin Oncol,1997,15(6):2420-2431.
The application value of IVIM-DW I for diagnosing and differentiating pancreatic cancer and neuroendocrine tumor
MA Wan-ling,ZHAO Wei-wei,HUAN Yi,WEIMeng-qi,REN Jing,YANG Yong,ZHANG Jin-song,PAN Qi,ZHANG Guang-wen,DONG Jun-qiang
(Department of Radiology,Xijing Hospital,Fourth Military Medical University,Xi’an 710032,China)
Objective:To explore the optimal quantitative parameters obtained from IVIM-DWI for diagnosing and differentiating pancreatic cancer and neuroendocrine tumor.M ethods:Subjects comprised 22 patients with pancreatic cancer and 17 patients with pancreatic neuroendocrine tumors,All the patients were confirmed by surgery.Pancreas multiple b value DWI was performed using GE Discovery MR750 3.0T scanner.Apparent diffusion coefficient(ADC),pure diffusion constant(D),pseudodiffusion coefficient(D*)and perfusion fraction(ƒ)were calculated by using IVIM model.Parameters obtained from IVIM-DWI were tested by One-Way ANOVA.Results:ADC andƒvalues of pancreatic cancer were lower than non-cancerous pancreatic tissue(0.803×10-3mm2/s vs 0.974×10-3mm2/s;54.527%vs 64.486%,respectively).D of pancreatic cancer was higher than noncancerous pancreatic tissue(0.659×10-3mm2/s vs 0.535×10-3mm2/s).D is superior to ADC andƒfor the differentiation between pancreatic cancer and non-cancerous pancreas.ADC,D andƒvalues of pancreatic neuroendocrine tumor were higher than non-tumorous pancreatic tissue(0.933×10-3mm2/s vs 0.753×10-3mm2/s;0.549×10-3mm2/s vs 0.429×10-3mm2/s;67.275%vs 59.655%,respectively).ADC is superior to D andƒfor the differentiation between pancreatic neuroendocrine tumor and nontumorous pancreas.ADC,D*andƒvalues of pancreatic cancer were lower than pancreatic neuroendocrine tumor(0.803×10-3mm2/s vs 0.933×10-3mm2/s;4.852×10-3mm2/s vs 11.301×10-3mm2/s;54.527%vs 67.275%,respectively).D*is superior to ADC andƒ for the differentiation between pancreatic cancer and neuroendocrine tumor.D*value of pancreatic cancer was significantly lower than G2~3 grade pancreatic neuroendocrine tumor(4.852×10-3mm2/s vs 11.937×10-3mm2/s).Conclusion:Quantitative parameters ADC,D,D*andƒobtained from IVIM-DWI can diagnose and differentiate pancreatic cancer,neuroendocrine tumor and non-cancerous pancreatic tissue.IVIM-DWImay be a promising and non-invasive tool for diagnosing and differentiating pancreatic carcinoma and neuroendocrine tumor.
Pancreatic neoplasms;Carcinoma,neuroendocrine;Diffusion magnetic resonance imaging
R735.9;R445.2
A
1008-1062(2017)01-0049-06
2016-04-25;
2016-05-16
马婉玲(1974-),女,陕西渭南人,副主任医师。E-mail:marynee@163.com
宦怡,第四军医大学西京医院放射科,710032。E-mail:huanyi3000@163.com
国家自然科学基金重大国际(地区)合作与交流项目(81220108011);国家自然科学基金青年面上连续项目(81370039)。