【作者】 周晓俊；

【导师】 钱海鑫；

【作者基本信息】 苏州大学 ， 外科学， 2010， 博士

【摘要】 第一部分64-MSCT三维重建对肝脏血管的解剖性研究目的1.通过64-MSCT肝脏三期扫描,应用三维重建技术对肝动脉、肝静脉和门静脉血管成像,分析三者在肝内的分布结构和不同变异类型;2.探讨肝内血管解剖结构及其变异对肝脏分段、肝脏肿瘤切除术、肝脏移植术和介入手术指导意义。材料与方法按一定纳入标准收集了苏州大学附属第一医院2008年1月~2008年6月期间行上腹部64-MSCT增强扫描的60例检查者作为研究对象,其中男37例,女23例,平均年龄49岁。全部检查者都应用64-MSCT(Somatom Sensation 64,Siemens)行平扫及三期增强扫描,将获得的扫描影像数据以1mm薄层重建后传至后期工作站。由2名影像科医生应用工作站软件对图像作以下处理:最大密度投影法(maximum intensity projection, MIP)、容积重建法(volume rendering, VR)、表面遮盖显示法(surface shadow display, SSD)及多轴向投影容积再现(multi-projection volumetric rendering, MPVR),将肝脏及内部血管三维重建。再有1名影像科医生和1名肝脏外科医生对三维重建的图像进行分析,根据肝内血管解剖分型的标准对肝动脉、肝静脉和门静脉进行解剖分型并记录所占比例。结果1. 60例检查者中肝动脉MichelsⅠ型(正常型)有47例(78.3%),其他13例(21.7%)存在肝动脉变异。其中MichelsⅡ型1例(1.7 %),Ⅲ型4例(6.7 %),Ⅴ型2例(3.3%),Ⅵ型3例(5.0%),Ⅷ型1例(1.7%),Ⅸ型1例(1.7 %),未发现Ⅳ型、Ⅶ型和Ⅹ型;另外还发现一例未被Michels分型包括在内的变异类型,无肝固有动脉,占1.7%。2.门静脉分型中,Ⅰ型(正常型)有49例(81.7%),Ⅱ型7例(11.7%),Ⅲ型3例(5.0%),Ⅳ型1例(1.7%)。未发现特殊变异类型。3.肝静脉分型中,绝大多数情况下(52/60,86.7%)肝右静脉单独汇入下腔静脉(Ⅰ型),而肝中、左静脉合干后汇入下腔静脉;少数(8/60,1.33 %)肝左、中、右静脉分别汇入下腔静脉(Ⅱ型)。并且,本组60例检查者中,共有14例(23.3%)显示肝右后下静脉(inferior right hepatic vein, IRHV)存在。根据Nakamura分型,肝静脉解剖结构变异Ⅰ型46例,占76.7%;Ⅱ型10例,占16.7%;Ⅲ型4例,占6.7%。结论1. 64-MSCT出现不但加快了数据采集的速度以及扫描完成后对获得的数据进行薄层重建,并可利用MIP、VR等后期软件处理技术可以对肝内血管系统在工作站进行三维重建成像,使高分辨率的CT血管成像(CTA)成为可能。相对于经典的DSA血管造影技术而言,这项技术不仅具有无创、操作简单等优点,而且可以从任意角度旋转来观察肝内血管及病变的情况。2. 64-MSCT可以对肝内血管系统进行三维重建,从而对肝动脉、门静脉和肝静脉的解剖结构及变异进行研究,本组研究发现三者变异的发生率分别为21.7%、18.3%和23.3%。了解肝内血管系统的解剖结构及变异对肝段的重新定位及划分,对肝脏占位性病变的部分切除术、对肝脏肿瘤的介入手术、对肝脏移植术,尤其是活体肝移植术的术前判断及评估、手术方案的制定、手术的预后及并发症的早期发现和处理都有重要的意义。第二部分肝脏体积测量在肝脏外科中的临床应用价值目的研究不同方法测量肝脏体积的可靠性和准确性,探讨64-MSCT测量肝脏体积对肝脏储备功能的评估及对预后判断的临床应用价值。材料与方法按一定纳入标准收集了苏州大学附属第一医院2008年1月~2009年12月期间50例行肝脏大部分切除术和6例行原位移植术的患者(男性42例,女性14例,平均年龄57岁)作为研究对象。56例患者术前都应用64-MSCT(Somatom Sensation 64,Siemens)行平扫及三期增强扫描,将获得的扫描影像数据1mm薄层重建后传至后期工作站。由1名影像科医生和1名肝脏外科医生分别应用64-MSCT自带后期处理软件(Leonardo, Siemens Medical Solutions, Malvern, Pennsylvania)、eFilm Workstation TM 2.1(Merge Healthcare)和Myrian○R XP-Liver医学影像分析系统(Intrasense○R , France)等三种不同方法测量肝脏体积,包括:全肝体积、肿瘤体积、功能肝体积、模拟切除肝体积和剩余肝体积,获得的结果分别与术后标本用排水法测得的实际体积进行比较,分析其相关性。记录患者的术前一般资料:年龄、性别、身高、体重、HbsAg和C-P分级等;术中资料:手术时间、术中失血量、肝门阻断时间、手术方式、剩余肝体积和残肝分数(%RLV)等;以及术后资料:肝功能、并发症情况等,应用单因素分析和Logistic多因素回归模型分析判断各种临床风险因素与术后发生肝功能障碍的影响,并应用受试者工作特性(ROC)曲线分析残肝分数对术后发生重度肝功能障碍的预测能力。结果1.三种方法(64-MSCT层切法、eFilm法和Myrian法)测量的全肝平均体积分别为:1355.9±806.8cm3,1438.9±344.7cm3和1450.4±354.2 cm3;模拟切除后预切除标本的平均体积分别为:754.9±326.8cm3,740.7±322.1cm3和753.6±330.1 cm3;而实际切除标本的平均质量为:718.30±327.15g,用排水法测量的平均体积为:729.3±332.4 cm3。前三者与后者体积的相关性比较,r=0.948,0.928和0.951(P<0.05);与后者质量的相关性比较,r=0.948, 0.927和0.945(<0.05);三种方法相互间相关性比较,r=0.995,0.986和0.980(<0.05)。说明三种方法测量的肝脏体积与实际体积呈高度正相关。2.单因素分析显示术中失血量、剩余肝体积和残肝分数与术后肝功能损害程度相关,而年龄、性别、C-P分级、肿瘤大小、肝硬化程度、手术时间、肝门阻断时间等因素对术后发生肝功能损害的影响无统计学意义。Logistic多因素回归模型分析表明残肝分数(%RLV)是术后发生严重肝功能损害的独立预测因素。应用ROC曲线分析残肝分数预测术后发生严重肝功能损害的临界点是0.32(P<0.05)。结论1.三种工作站软件用于肝脏体积的测量,各有优劣。相比较而言, Myrian○R XP-Liver医学影像分析系统操作简单,高效,操作耗时短,获得的图像直观立体,测量数据准确,并且可在个人电脑上安装,带入手术室操作,方便手术医生。2.经不同工作站软件测量获得的肝脏体积与实际肝脏体积之间虽然存在着一定的差异,但是都与实际体积存在着固定的线性关系,呈高度正相关。因此可应用64-MSCT比较准确地测量肝脏体积从而对肝脏储备功能提供全面的评价。3.肝脏大部分切除术后发生严重肝功能障碍与术中失血量、剩余肝体积和残肝分数等因素相关;残肝分数<0.32,术后发生严重肝功能障碍的危险性增加。因此,应用64-MSCT及后期工作站软件对肝脏体积的测量,可以有效预测术后肝功能损害发生的风险。第三部分虚拟肝脏手术与辅助风险分析对肝脏大部分切除术手术方案与术中策略的影响目的探讨应用肝脏三维重建模型来进行虚拟肝脏手术,对实际手术方案的制定及术中策略的影响。材料与方法按一定纳入标准前瞻性地选择了苏州大学附属第一医院2009年1月~2009年12月期间行较复杂的肝大部分切除术患者22例,其中男17例,女5例,平均年龄47岁。所有患者术前都行64-MSCT三期薄层扫描,将获得的影像数据导入Myrian○R XP-Liver医学影像分析系统(Intrasense○R , France),选用门静脉后期的图像三维重建来进行虚拟肝脏手术切除。步骤一:图像分析及三维重建;步骤二:虚拟手术切除,测量计算各部分的体积;步骤三:预后风险分析,调整和制定手术方案。将虚拟手术制定的方案与实际手术操作和术后情况进行比较。结果1.应用Myrian○R XP-Liver医学影像分析系统获得的肝脏三维重建模型形象逼真,立体感强,肝脏形态、肿瘤的大小、位置和与周围血管的空间毗邻关系一目了然,并且可从多个角度进行观察分析。虚拟的图像与手术中实际所见匹配良好,达到理想的模拟效果。2. 22例患者都施行了肝脏大部分切除术,其中20例完整切除肿瘤,达到R0切除,安全切缘距肿瘤边缘最小距离4mm~4cm,平均2.4cm;2例为R1切除,1例原发性肝癌切缘最小距离肿瘤<4mm,切缘阳性;另1例肝门部胆管癌,术后病理镜下证实胆管断端阳性。本组病例无手术死亡和住院期间死亡,出现严重肝功能障碍2例,均保守治疗后好转出院,其中1例术后三月由于上消化道出血死亡。3.本组病例中有6例患者通过预后风险分析,计算剩余肝脏体积和残肝分数,分析剩余肝组织可能出现缺血/淤血的范围,认为术后肝功能障碍发生危险性增加,于是调整了切除范围,最大限度保留功能性肝组织,从而制定了合理的手术方案,影响了术中策略,取得良好的手术效果。结论应用Myrian○R XP-Liver医学影像分析系统进行肝脏三维重建及虚拟肝脏切除手术,可以为精确地安全地施行复杂的肝脏大部分切除术,最大限度减少术后肝功能障碍的发生,制定合理的手术方案和采取正确的术中策略提供重要的术前参考。更多还原

【Abstract】 Part I Studies of anatomy of liver vascular with 64-MSCT three-dimensional reconstructionObjectiveTo analyze the inner anatomical structure and variation of hepatic artery, vein and portal vein through hepatic angiography using three-dimensional reconstruction technique. To investigate the instructive significance for segmentation, liver tumor resection, liver transplantation and intervention for different vascular anatomical structure and variation.Materials and MethodsTotal 60 patients, who were performed upper abdominal examination using triple-phase contrast-enhancement 64-MSCT in the first affiliated hospital of Soochow university, were collected into this study in a period of 6 months, from Jan to Jun, 2008. All of the data sets were transfered to workstation after being reconstructed in 1-mm section and 1-mm interval. Two radiologists took the charge of dealing with the data sets for maximum intensity projection (MIP), volume rendering (VR), surface shadow display (SSD), multi-projection intensity rendering (MPVR) and angiography. Furthermore, analyzed by 1 radiologist and 1 surgeon majoring in liver surgery, the anatomical structure of hepatic artery, vein and portal vein were classified according to certain standards and the rates of variation were recorded. Results1. Among the 60 patients, there were 47 patients (78.3%) belonging to Michels type I (normal hepatic artery anatomy), and other 13 patients (21.7%) showing the existence of variation. They were typeⅡin 1 (1.7%), typeⅢin 4 (6.7%), typeⅤin 2 (3.3%), typeⅥin 3 (5.0%), both typeⅧand typeⅨin 1 (1.7%), and there was no typeⅩ,Ⅳ, andⅦ. However, a special variation type without proper hepatic artery was presented, which could not be included in the Michels Classification.2. Normal portal vein anatomy (type I) was shown in 49 cases (81.7%), typeⅡin 7 (11.7%), typeⅢin 3 (5.0%) and typeⅣin 1 (1,7%). There was no special variation founded.3. Most cases (52/60, 86.7%) of RHV drained into IVC alone (belonging to type I) , and MHC combined with LHV into IVC. Less cases (8/60, 1.33%) of these three hepatic vein drained into IVC respectively (typeⅡ). There were 14 cases (23.3%) showing the existence of inferior right hepatic vein (IRHV). Following to Nakamura Classification, 46 cases (76.7%) were presented in normal type (type I), typeⅡin 10 (16.7%), and typeⅢin 4 (6.7%).Conclusion1. The application of triple-phase scanning with 64-MSCT make it possible to acquire hepatic angiography with a higher resolution, depending on some functional software, such as MIP or VR. Compared with the conventional examination DSA, 64-MSCT three-dimensional reconstruction is a non-invasive and convenient vascular examination, which can provide complete and comprehensive information to observe the inner structure of vascular and the lesion from different angles.2. 64-MSCT three-dimensional reconstruction can make it more clearly to understand the anatomical structure and the variation of inner hepatic vascular system. It has significantly valuation in segmentation, planning for tumor resection, intervention treatment of liver malignance and liver transplantation, especially in living-donor liver transplantation. Key Words: multi-slice computed tomography, liver, three-dimensional reconstruction, angiography, variationPartⅡClinical application of liver volumetry in hepatic surgeryObjectiveTo evaluate the accuracy and reliability of liver volumetry using different methods and also to investigate the clinical value of liver volumetry in assessing liver functional reserve and as a parameter to predict the prognosis after major hepatectomy.Materials and MethodsThere were 50 patients, who were performed major hepatectomy, and 6 patients, who were performed orthotopic liver transplantation, in the first affiliated hospital of Soochow university from January 2008 to December 2009, were collected into this study and analyzed retrospectively. All of the patients were examined by triple-phase contrast-enhancement 64-MSCT and were applied for three-dimensional reconstruction. Three different kinds of workstation software were used to measure the volume of liver, such as Leonardo (Seimens Medical Solution), bundled with 64-MSCT, eFilm Workstation TM 2.1 (Merge Healthcare) and Myrian○R XP-Liver (Intrasense○R , France). Total liver volume (TLV), tumor volume (TuV) were measured by 64-MSCT. Functional liver volume (FLV), virtual resected liver volume (vRLV), remnant liver volume (RLV) and relative residual liver volume (%RLV) were calculated. All the results were compared with the actual volume, which was measured by the method of water displacement after tumor resectionAll the pre-, peri- and post-operative information were recorded, such as sex, age, weight, height, C-P classification, operating time, blood loss and the remnant liver volume, complications and so on. Univariate analysis and multivariate logistic regression analysis were performed to delineate perioperative predictors of severe hepatic dysfunction. Receiver operator characteristic curve analysis was undertaken to determine the critical %RLV predicting severe hepatic dysfunction. Results1. The mean TLV of 56 patients measured by three different methods was 1355.9±806.8cm3, 1438.9±344.7cm3 and 1450.4±354.2 cm3. The mean volume of virtual reseced specimen volume (vRSV) was 754.9±326.8cm3, 740.7±322.1cm3 and 753.6±330.1 cm3. Compared with the mean volume of actual resected specimen volume (aRSV) using water displacement was 729.3±332.4 cm3 and the mean weight was 718.30±327.15g. Comparison by means of linear regression analysis between volume measurement on the software and the actual resected volume and weight showed a nearly ideal correlation coefficient (r=0.948, 0.927and 0.945, P<0.05; r=0.995, 0.986 and 0.980, P<0.05).2. Univariate analysis revealed that intraoperative blood loss, RLV and %RLV were significant prognostic indication for severe hepatic dysfunction. The incidence of severe hepatic dysfunction following hepatic resection increased significantly with smaller %RLV. A critical %RLV of 0.32 was identified as associated with severe hepatic dysfunction (P<0.05).Conclusion1. Three methods used for measuring the volume of liver have different characteristics. Myrian○R XP-Liver was recommended for its convenience and efficiency.2. Although the results measured by these three methods were different with actual volume measured by water displacement, all of them had a proper linear correlation with actual results.3. Residual liver volume after virtual hepatectomy could be a predictor of hepatic dysfunction. The likelihood of severe hepatic dysfunction following hepatic resection can be predicted by a small %RLV. Reducing the intraoperative blood loss and increasing the RLV as much as possible can decrease the risk of hepatic post-operative dysfunction. PartⅢImpact of virtual hepatectomy and computed-assisted risk analysis on operation planning and intraoperative strategy in major hepatic resectionObjectiveTo evaluate the impact of virtual hepatectomy and computed-assisted risk analysis on operation planning for major hepatectomies, in particular on extent of resection or need for vascular reconstruction.Materials and MehtodsTwenty-two consecutive patients (17 male, 5 female; median age, 47 years) admitted to the first affiliated hospital of Soochow university from January to December 2009 for complicated major hepatectomy. Underlying tumors were primary hepatocellular carcinoma (n=14, 63.6%), metastases of colorectal cancer (n=2, 9.1%), hemangioma (n=3,13.6%), Klaskin’s tumor (n=2, 9.1%) and gallbladder carcinoma (n=1, 4.5%). The CT scanning was performed with 64-MSCT (Somatom Sensation 64, Germany). All the data of imaging was imported into Myrian○R XP-Liver (Intrasense○R , France) system, and the three-dimensional reconstruction model was performed for virtual hepatectomy. The following steps were taken by 1 radiologist and 1 surgeon majoring in liver surgery: step 1, image analysis and 3-D reconstruction; step 2, virtual resection and liver volumetry; step 3, computer-assisted risk analysis for the hepatic dysfunction, and adjustment of the operation planning.Results1. The three-dimensional reconstruction model of liver by Myrian○R XP-Liver system is visualized and verisimilar. The ideal simulation effect can be achieved compared the virtual image with actual operation.2. Overall, 22 patients underwent complicated major hepatectomy and in 20 of the 22 resections, complete tumor removal was achieved (R0 resection). The achieved safety margins ranged from 4mm to 4cm, with a median of 1.8 cm. There was 2 cases R1 resection. In these 2 cases, there were microscopic tumor at the cutting margin of the right portal vein and bile duct. There was no one died in operation and hospital stay. 2 cases appeared severe hepatic dysfunction after operation and was discharged 2~3 weeks later after conservative treatment. One of them died of severe upper gastrointestinal bleeding three months later.3. In 6 patients, the results of computer-assisted risk analysis made us change operation planning and surgical strategy with regard to extent resection or the need for vascular reconstruction. With the planning, an optimal identification of safety resection margins could be achieved. The volume of remnant liver and the ischemia/congestion territory within the remaining parenchyma can be calculated. The operation planning and intraoperative strategy were adjusted to avoid liver parenchyma over resection and maintain a sufficient amount of liver tissue to sustain hepatic function.ConclusionThe application of hepatic three-dimensional reconstruction and virtual hepatectomy, using Myrian○R XP-Liver system, can provide the important preoperative reference for a valuable planning and intraoperative strategy of complex hepatectomy.更多还原