CT、MRI评估原发性纵隔肿瘤血管侵犯的应用研究
|
摘要
目的探讨MRI多参数成像及CT增强扫描在评估原发性纵隔肿瘤血管侵犯中的临床应用价值,初步为临床工作制定阅片规范。
材料与方法收集我院胸外科收治的原发性纵隔肿瘤患者30例,术前行常规胸部MRI多参数扫描(包括T1WI、T2WI、T2WI/fs、MR增强)以及常规胸部CT增强扫描;前瞻性评估肿瘤侵犯血管情况。原发性纵隔肿瘤侵犯血管的影像学诊断标准为: (1)肿瘤与纵隔血管之间的脂肪线消失; (2)肿瘤与纵隔血管接触面﹥管周的1/2; (3)肿瘤与纵隔血管接触局部管腔狭窄; (4)血管壁异常密度/信号; (5)血管腔内瘤栓形成。
符合以上其中一个条件,诊断为纵隔肿瘤侵犯血管。
1、根据血管侵犯的影像学诊断标准,通过单独分析MRI的多参数成像(T1WI、T2WI、T2WI/fs、MR增强)以及CT增强扫描共5种不同影像对比度,分别评估肿瘤的血管侵犯情况,以手术、病理结果为金标准,计算5种不同影像对比度的敏感度、特异度、误诊率、漏诊率、阳性预测值、阴性预测值、诊断符合率,以及与手术、病理的吻合度(κ系数)。
2、通过单独分析影像学诊断标准中的5种征象在MRI图像的显示,评估肿瘤的血管侵犯情况,以手术、病理结果为金标准,分别计算5种影像学征象的敏感度、特异度、误诊率、漏诊率、阳性预测值、阴性预测值、诊断符合率,以及与手术、病理的吻合度(κ系数)。
应用SPSS13.0软件进行统计学分析。统计方法使用计数资料的Cochrans’Q检验、κ检验。
结果
1、5种影像对比度(T1WI、T2WI、T2WI/fs、MR增强、CT增强扫描)分别诊断原发性纵隔肿瘤血管侵犯的敏感度、特异度、阳性预测值、阴性预测值、诊断符合率经Cochrans’Q检验,Q值分别为121.945、10.667、14.764、20.586、106.362,P值均小于0.05。5种影像对比度诊断结果与手术、病理对比的κ系数分别为0.713,0.776,0.296,0.232,0.596。
2、5种影像学征象(肿瘤与纵隔血管之间的脂肪线消失、肿瘤与纵隔血管接触面﹥管周的1/ 2、肿瘤与纵隔血管接触局部管腔狭窄、血管壁异常信号、血管腔内瘤栓形成)在MRI上的显示诊断血管侵犯的影像学结果与手术、病理对比,κ系数分别为0.737,0.386,0.410,0.232,0.238。
结论MRI多参数成像与CT增强扫描诊断原发性纵隔肿瘤血管侵犯差异有统计学意义。通过MRI的T2WI判断纵隔肿瘤侵犯血管与手术、病理结果吻合度最高,其余依次为T1WI、CT增强扫描、T2WI/fs、MR增强。通过MRI上脂肪线消失判断血管侵犯与手术、病理结果吻合度最高;MRI显示血管壁异常信号、血管腔内瘤栓形成诊断纵隔血管侵犯的阳性预测值高。MRI扫描应成为原发性纵隔肿瘤术前检查、制定手术方案的必要手段。
Purpose
To explore the Magnetic Resonance Imaging(MRI) and Contrast Enhancement-CT (CE-CT) on evaluation of vascular invasion in primary mediastinal tumors.
Material and Methods
Thirty patients with primary mediastinal tumors and which were going to have surgical therapy, underwent chest MRI check (including T1WI, T2WI, T2WI/fs, CE-MRI) and CE-CT. The diagnostic criteria of the tumor which has invaded the vessel are: (1)The fat between the tumor and the vessel disappeared. (2)The tumor surrounded more than half of the vessel wall. (3)The vessel become narrowing when surrounded by tumor. (4)The density or signal of the vessel wall become abnormal. (5)Embolus appearance.
Coincidencing with one of the above condition, it is consider that the vessel had been invaded.
1、By means of the criteria above, the data of the CT and MR Imaging were analyzed. The relation between the tumor and the vessel was confirmed by operation and pathology. Finaly, the imaging and pathology result were compared.
2、The Pathology of each MRI manifestation which indicated the vessel had invaded was comfirmed.
Statistical Analysis
The statistical analysis were performed with the statistcal software packgae SPSS for Widow, version 13.0, including Cochrans’Q test, andκtest.
Results
1. The sensitivity, specificity, positive predictive value, negative presitive and accuracy for the five imaging (including T1WI, T2WI, T2WI/fs, CE-MRI and CE-CT), detecting whether the vessel had been invded or not, had been tested by Cochrans’Q test. The Q were 121.945, 10.667, 14.764, 20.586, 106.362 respectively,P<0.05. Theκwere 0.713,0.776,0.296,0.232,0.619 respectively, when compared with each imaging and pathology.
2.Κappa value for MRI each criteria on vascular ivasion were 0.737,0.386,0.410,0.232,0.238 respectively.
Conclusion
T2WI is the optimum imaging for evaluation of whether the vessel had been invded or not , and T1WI, CE-CT, T2WI/fs, CE-MRI. The accuracy is biggest, for the disappearance of the fat between the tumor and the invded vessel. The specificity of MRI is biggest, for detecting the abnormal signal on invded vessel wall and blood vessel embolus. The preoperative MRI examination is necessary for detecting the vessel had been invded or not.
材料与方法收集我院胸外科收治的原发性纵隔肿瘤患者30例,术前行常规胸部MRI多参数扫描(包括T1WI、T2WI、T2WI/fs、MR增强)以及常规胸部CT增强扫描;前瞻性评估肿瘤侵犯血管情况。原发性纵隔肿瘤侵犯血管的影像学诊断标准为: (1)肿瘤与纵隔血管之间的脂肪线消失; (2)肿瘤与纵隔血管接触面﹥管周的1/2; (3)肿瘤与纵隔血管接触局部管腔狭窄; (4)血管壁异常密度/信号; (5)血管腔内瘤栓形成。
符合以上其中一个条件,诊断为纵隔肿瘤侵犯血管。
1、根据血管侵犯的影像学诊断标准,通过单独分析MRI的多参数成像(T1WI、T2WI、T2WI/fs、MR增强)以及CT增强扫描共5种不同影像对比度,分别评估肿瘤的血管侵犯情况,以手术、病理结果为金标准,计算5种不同影像对比度的敏感度、特异度、误诊率、漏诊率、阳性预测值、阴性预测值、诊断符合率,以及与手术、病理的吻合度(κ系数)。
2、通过单独分析影像学诊断标准中的5种征象在MRI图像的显示,评估肿瘤的血管侵犯情况,以手术、病理结果为金标准,分别计算5种影像学征象的敏感度、特异度、误诊率、漏诊率、阳性预测值、阴性预测值、诊断符合率,以及与手术、病理的吻合度(κ系数)。
应用SPSS13.0软件进行统计学分析。统计方法使用计数资料的Cochrans’Q检验、κ检验。
结果
1、5种影像对比度(T1WI、T2WI、T2WI/fs、MR增强、CT增强扫描)分别诊断原发性纵隔肿瘤血管侵犯的敏感度、特异度、阳性预测值、阴性预测值、诊断符合率经Cochrans’Q检验,Q值分别为121.945、10.667、14.764、20.586、106.362,P值均小于0.05。5种影像对比度诊断结果与手术、病理对比的κ系数分别为0.713,0.776,0.296,0.232,0.596。
2、5种影像学征象(肿瘤与纵隔血管之间的脂肪线消失、肿瘤与纵隔血管接触面﹥管周的1/ 2、肿瘤与纵隔血管接触局部管腔狭窄、血管壁异常信号、血管腔内瘤栓形成)在MRI上的显示诊断血管侵犯的影像学结果与手术、病理对比,κ系数分别为0.737,0.386,0.410,0.232,0.238。
结论MRI多参数成像与CT增强扫描诊断原发性纵隔肿瘤血管侵犯差异有统计学意义。通过MRI的T2WI判断纵隔肿瘤侵犯血管与手术、病理结果吻合度最高,其余依次为T1WI、CT增强扫描、T2WI/fs、MR增强。通过MRI上脂肪线消失判断血管侵犯与手术、病理结果吻合度最高;MRI显示血管壁异常信号、血管腔内瘤栓形成诊断纵隔血管侵犯的阳性预测值高。MRI扫描应成为原发性纵隔肿瘤术前检查、制定手术方案的必要手段。
Purpose
To explore the Magnetic Resonance Imaging(MRI) and Contrast Enhancement-CT (CE-CT) on evaluation of vascular invasion in primary mediastinal tumors.
Material and Methods
Thirty patients with primary mediastinal tumors and which were going to have surgical therapy, underwent chest MRI check (including T1WI, T2WI, T2WI/fs, CE-MRI) and CE-CT. The diagnostic criteria of the tumor which has invaded the vessel are: (1)The fat between the tumor and the vessel disappeared. (2)The tumor surrounded more than half of the vessel wall. (3)The vessel become narrowing when surrounded by tumor. (4)The density or signal of the vessel wall become abnormal. (5)Embolus appearance.
Coincidencing with one of the above condition, it is consider that the vessel had been invaded.
1、By means of the criteria above, the data of the CT and MR Imaging were analyzed. The relation between the tumor and the vessel was confirmed by operation and pathology. Finaly, the imaging and pathology result were compared.
2、The Pathology of each MRI manifestation which indicated the vessel had invaded was comfirmed.
Statistical Analysis
The statistical analysis were performed with the statistcal software packgae SPSS for Widow, version 13.0, including Cochrans’Q test, andκtest.
Results
1. The sensitivity, specificity, positive predictive value, negative presitive and accuracy for the five imaging (including T1WI, T2WI, T2WI/fs, CE-MRI and CE-CT), detecting whether the vessel had been invded or not, had been tested by Cochrans’Q test. The Q were 121.945, 10.667, 14.764, 20.586, 106.362 respectively,P<0.05. Theκwere 0.713,0.776,0.296,0.232,0.619 respectively, when compared with each imaging and pathology.
2.Κappa value for MRI each criteria on vascular ivasion were 0.737,0.386,0.410,0.232,0.238 respectively.
Conclusion
T2WI is the optimum imaging for evaluation of whether the vessel had been invded or not , and T1WI, CE-CT, T2WI/fs, CE-MRI. The accuracy is biggest, for the disappearance of the fat between the tumor and the invded vessel. The specificity of MRI is biggest, for detecting the abnormal signal on invded vessel wall and blood vessel embolus. The preoperative MRI examination is necessary for detecting the vessel had been invded or not.
引文
[1]李连弟等.中国试点市、县恶性肿瘤的发病与死亡(1998-2003).北京卫生出版社,2007,210-371.
[2] Cameron RB, Loekrersr PI, Thomastr CR. Neoplasma of the Mediastinum. From Vincent Jr TD, Hellmans, Rosenberg SA. Cancer Priciples and Practice of Oncology 7th Lippincott Williams and Wilkims, 2005, 845-860.
[3] Davis RD, Oldhan HN, Sabiston DC, Primary cysts and neoplasms of the mediastinum: recent changes in clinical presentation, methods of diagnosis, management and results. Ann Thoracic Surg, 1987, 44(3):229-237.
[4]赵福元. 842例原发性纵隔肿瘤的诊断和外科治疗. 2001年度天津医学会肿瘤年会论文汇编,2002, 35-37.
[5] Armstrong P, Wilson AG, Dee P, et al. Imaging of disease of the chest. 3rd ed. London: Mosby, 2000.
[6] Mao ZF, Cheng BC, Xia J, et al. Surgical trentment for giant solid tumors of the mediastinum: a study of 26 cases. Int Surg, 2003, 88(3):164-168.
[7]昌盛,毛志福,程邦昌,等.侵及胸廓上口大血管的纵隔肿瘤的外科治疗.临床外科杂志,2005,139(2), 103-104.
[8] Spaggiari L, Leo F, Veronesi G, et al. Superior vena cava resection for lung and mediastinal malignancies: a single experience with 70 cases. Ann Thorac Surg. 2007, 83(1):223-230; discussion 229-230.
[9] Herman M, Paucek B, Raida L, et al. Comparison of MRI and (67)gallium scintigraphy in the evaluation of posttherapeutic residual mediastinal mass in the patients with Hodgkin’s lymphoma. Eur J Radiol, 2007, 64(3), 432-438.
[10] Saraya T, Shimura C, Mikura S, et al. Huge mediastinal mass with SVC syndrome accompanying numerous chest wall collateral vessels. Intern Med, 2008, 47(19):1719-1722.
[11]李建生,汤日杰,彭俊琴,等.螺旋CT多平面成像在纵隔肿瘤诊断及治疗中的价值.现代肿瘤医学, 2009,17(12): 2323-2325.
[12]林国为,沈福民.现代临床流行病学.上海医科大学出版社,第一版, 2000: 56-58.
[13]颜虹.主编.医学统计学第一版.北京:人民卫生出版社,2005:121-207
[14] Loyer EM, Davad CL, Dubrow RA, et al. Vascular involvement in pancreatic adenocarcinoma: reassessment by thin-section CT. Abdom Imaging, 1996, 21(3): 202-206.
[15] Lu DS, Reber HA, Krasny RM, et al. Local staging of pancreatic cancer: criteria for unresectability of major vessels as revealed by pancreatic-phase, thin-section helical CT. AJR Am J Roentgenol, 1997,168(4):1439-1443.
[16] Valls C, Andia E, Sanchez A, et al. Dual-phase helical CT of pancreatic adenocarcinoma: assessment of resetalility before surgery. AJR Am J Roentgenol, 2002,178 (4): 821-826.
[17] Brügel M, Rummeny EJ, Dobritz M. Vascular invasion in pancreatic cancer:value of multislice helical CT. Abdom Imaging, 2004, 29(2):239-245.
[18] Fletcher JG, Wiersema MJ, Farrell MA, et al. Pancteatic malignancy: value of arterial, pancreatic, and hepatic phase imaging with multi-detector row CT. Radiology, 2003, 229(1):81-90.
[19] Nakayama Y, Yamashita Y, Kadota M, et al. Vsscular encasement by pancreatic cancer: correlation of CT findings with surgical and pathologic results. J comput Assist Tomogr, 2001, 25(3): 337-342.
[20] Choi BI, Chung MJ, Han JK, et al. Detection of pancreatic adenocarcinoma: relative value of arterial and late phases of spiral CT. Abdom Imaging, 1997, 22(2):199-203
[21] Bluemke DA, Cameron HL, Hruban RH, et al. Potentially respectable pancreatic adenocacinoma: spiral CT assessment with surgical and pathologic correlation. Radiology 1995, 197(2):381-385.
[22] Freeny PC, Traverso LW, Ryan JA, et al. Diagnosis and staging of pancteatic adenocarcinoma with dynamic computed tomography. Am J Surg, 1993, 165(5):600-606.
[23] Kingenbeck-Regn K, Schaller T, Flohr T, et al. Subsecond multi-slice computed tomography: basics and applications. Eur J Radiol, 1999,31(2):110-124.
[24] Krestin GP, Glazre GM. Advances in CT IV. Berlin: Springer 1998, 49-58.
[25] Marincek B, Ros PR, Reiser M, et al. Multislice CT: a practical guide. Berlin: Springer 2001, 79-134.
[26] Spring BI, Schiebler ML. Normal anatomy of the thoracic inlet as seen on the transaxialMR imaging. 1999, 157(4), 707-710.
[27] Grenacher L, Klauss M, Dukic L, et al. Diagnosis and staging of pancreatic carcinoma: MRI versus multislice-CT—a prospective study. Rofo, 2004,176(11):1624-1633
[28] Murakami K, Nawano S, Moriyama N, et al. Usefulness of magnetic resonance imaging with dynamic contrast enhancement and fat suppression in detecting a pancreatic tumor. Jpn J Clin Oncol, 1998, 28(2):107-111.
[29] Soto JA, Barish MA, Yucel KE, et al. Pancreatic duct: MR cholangiopancreatography with a three-dimentional fast spin-echo technique. Radiology 1995, 196(2):459-464.
[2] Cameron RB, Loekrersr PI, Thomastr CR. Neoplasma of the Mediastinum. From Vincent Jr TD, Hellmans, Rosenberg SA. Cancer Priciples and Practice of Oncology 7th Lippincott Williams and Wilkims, 2005, 845-860.
[3] Davis RD, Oldhan HN, Sabiston DC, Primary cysts and neoplasms of the mediastinum: recent changes in clinical presentation, methods of diagnosis, management and results. Ann Thoracic Surg, 1987, 44(3):229-237.
[4]赵福元. 842例原发性纵隔肿瘤的诊断和外科治疗. 2001年度天津医学会肿瘤年会论文汇编,2002, 35-37.
[5] Armstrong P, Wilson AG, Dee P, et al. Imaging of disease of the chest. 3rd ed. London: Mosby, 2000.
[6] Mao ZF, Cheng BC, Xia J, et al. Surgical trentment for giant solid tumors of the mediastinum: a study of 26 cases. Int Surg, 2003, 88(3):164-168.
[7]昌盛,毛志福,程邦昌,等.侵及胸廓上口大血管的纵隔肿瘤的外科治疗.临床外科杂志,2005,139(2), 103-104.
[8] Spaggiari L, Leo F, Veronesi G, et al. Superior vena cava resection for lung and mediastinal malignancies: a single experience with 70 cases. Ann Thorac Surg. 2007, 83(1):223-230; discussion 229-230.
[9] Herman M, Paucek B, Raida L, et al. Comparison of MRI and (67)gallium scintigraphy in the evaluation of posttherapeutic residual mediastinal mass in the patients with Hodgkin’s lymphoma. Eur J Radiol, 2007, 64(3), 432-438.
[10] Saraya T, Shimura C, Mikura S, et al. Huge mediastinal mass with SVC syndrome accompanying numerous chest wall collateral vessels. Intern Med, 2008, 47(19):1719-1722.
[11]李建生,汤日杰,彭俊琴,等.螺旋CT多平面成像在纵隔肿瘤诊断及治疗中的价值.现代肿瘤医学, 2009,17(12): 2323-2325.
[12]林国为,沈福民.现代临床流行病学.上海医科大学出版社,第一版, 2000: 56-58.
[13]颜虹.主编.医学统计学第一版.北京:人民卫生出版社,2005:121-207
[14] Loyer EM, Davad CL, Dubrow RA, et al. Vascular involvement in pancreatic adenocarcinoma: reassessment by thin-section CT. Abdom Imaging, 1996, 21(3): 202-206.
[15] Lu DS, Reber HA, Krasny RM, et al. Local staging of pancreatic cancer: criteria for unresectability of major vessels as revealed by pancreatic-phase, thin-section helical CT. AJR Am J Roentgenol, 1997,168(4):1439-1443.
[16] Valls C, Andia E, Sanchez A, et al. Dual-phase helical CT of pancreatic adenocarcinoma: assessment of resetalility before surgery. AJR Am J Roentgenol, 2002,178 (4): 821-826.
[17] Brügel M, Rummeny EJ, Dobritz M. Vascular invasion in pancreatic cancer:value of multislice helical CT. Abdom Imaging, 2004, 29(2):239-245.
[18] Fletcher JG, Wiersema MJ, Farrell MA, et al. Pancteatic malignancy: value of arterial, pancreatic, and hepatic phase imaging with multi-detector row CT. Radiology, 2003, 229(1):81-90.
[19] Nakayama Y, Yamashita Y, Kadota M, et al. Vsscular encasement by pancreatic cancer: correlation of CT findings with surgical and pathologic results. J comput Assist Tomogr, 2001, 25(3): 337-342.
[20] Choi BI, Chung MJ, Han JK, et al. Detection of pancreatic adenocarcinoma: relative value of arterial and late phases of spiral CT. Abdom Imaging, 1997, 22(2):199-203
[21] Bluemke DA, Cameron HL, Hruban RH, et al. Potentially respectable pancreatic adenocacinoma: spiral CT assessment with surgical and pathologic correlation. Radiology 1995, 197(2):381-385.
[22] Freeny PC, Traverso LW, Ryan JA, et al. Diagnosis and staging of pancteatic adenocarcinoma with dynamic computed tomography. Am J Surg, 1993, 165(5):600-606.
[23] Kingenbeck-Regn K, Schaller T, Flohr T, et al. Subsecond multi-slice computed tomography: basics and applications. Eur J Radiol, 1999,31(2):110-124.
[24] Krestin GP, Glazre GM. Advances in CT IV. Berlin: Springer 1998, 49-58.
[25] Marincek B, Ros PR, Reiser M, et al. Multislice CT: a practical guide. Berlin: Springer 2001, 79-134.
[26] Spring BI, Schiebler ML. Normal anatomy of the thoracic inlet as seen on the transaxialMR imaging. 1999, 157(4), 707-710.
[27] Grenacher L, Klauss M, Dukic L, et al. Diagnosis and staging of pancreatic carcinoma: MRI versus multislice-CT—a prospective study. Rofo, 2004,176(11):1624-1633
[28] Murakami K, Nawano S, Moriyama N, et al. Usefulness of magnetic resonance imaging with dynamic contrast enhancement and fat suppression in detecting a pancreatic tumor. Jpn J Clin Oncol, 1998, 28(2):107-111.
[29] Soto JA, Barish MA, Yucel KE, et al. Pancreatic duct: MR cholangiopancreatography with a three-dimentional fast spin-echo technique. Radiology 1995, 196(2):459-464.