【作者】 陈继业；

【导师】 董家鸿；

【作者基本信息】 中国人民解放军军医进修学院 ， 外科学， 2012， 博士

【摘要】 一、背景与目的基于现代综合医学模式倡导的全新肝脏外科理念和技术体系，为进一步实现彻底去除病灶、最大肝脏保护和最小创伤侵袭三者的统一，董家鸿教授提出了精准肝切除这一新的肝脏外科理念。然而，如何精确判定拟切除门脉流域的边界是实施精准肝切除的关键。目前临床上最广为应用的方法是Makuuchi提出的美兰染色法，然而这一方法需要对毫米级肝段或亚肝段血管进行精确穿刺，对穿刺技术要求极高。亦有学者应用分离肝实质探寻肝段或亚肝段血管并进行结扎，从而确定出肝段或亚肝段的缺血区域，但这一方法会进一步增加手术出血量及手术时间。最近，Navarra等人提出了通过超声引导下射频消融对目标肝段的肝蒂进行热毁损从而来确定拟切除门脉流域的缺血边界。相较于前述两种方法，该方法具有三大优势：1)穿刺技术简便；2)不增加手术出血量及手术时间；3)由于在切除前即对肝段供应血管进行热毁损，从而避免术中肿瘤细胞经门脉系统扩散，理论上更符合肿瘤学治疗原则。然而，由于射频消融可能会直接导致邻近肝段肝蒂组织不可逆的热损伤，因此目前对该项技术的临床推广应用仍存有争议。本试验通过进一步改进该项技术方法并建立活体小型猪精准肝切除模型来评估该项技术在精准肝切除中应用的可行性及可重复性，同时探索操作难点并总结经验。二、材料方法1.动物分组及模型建立：选取健康的广西巴马小型猪12头，体重25－35Kg，随机分为两组：A组:肝蒂射频消融同时未阻断肝动脉组；B组:肝蒂射频消融同时阻断肝动脉组。实验动物全身麻醉气管插管后连接呼吸机,颈动脉插管监测心率、血压变化，颈静脉插管建立输液通道。根据CourtF.G.等人报道的猪肝的肝段解剖特征，对每头试验动物均进行肝Ⅲ、Ⅴ、Ⅵ段切除。具体手术步骤如下：1).试验动物采用平卧位，右肋缘下J形切口逐层切开止血入腹；2).为避免射频消融导致胆囊损伤，常规切除胆囊；3).术中超声探查肝脏，仔细辨别供应肝Ⅲ、Ⅴ、Ⅵ段门静脉；4).超声引导下紧贴肝段血管插入冷循环电级，连接射频消融系统进行肝蒂毁损,电级插入部位距肝蒂分叉处的垂直距离至少为2cm，避免损伤邻近肝段供应血管。当拟切除肝段出现缺血边界，立即停止射频消融并记录射频消融时间；5).根据缺血边界采用钳夹法离断肝实质，完整切除肝Ⅲ、Ⅴ、Ⅵ段。对于B组试验动物，我们开腹后分离肝门部结构，游离出肝动脉，在射频消融的过程中阻断肝动脉，当拟切除肝段出现缺血边界停止射频消融后，我们即解除肝动脉阻断。其余手术过程同A组。2.观察指标：1).动物术中情况：术中持续监测动物血压、心率，评估出血量，记录射频消融时间。2)动物术后存活情况：所以动物均精心饲养14d，存活>14d视为长期存活。术后14d所有存活动物均在全麻下行剖腹探查术，探查是否存在手术并发症(如胆瘘、腹水、腹腔脓肿、腹腔出血等)。3).动物肝功能变化：于术前、术后1d、术后14d不同时相点采取静脉血标本测定丙氨酸氨基转移酶(ALT)、天冬氨酸氨基转移酶(AST)、γ-谷氨酰基转移酶(GGT)、碱性磷酸酶(ALP)、白蛋白(ALB)、总胆红素(TBIL)、直接胆红素(DBIL)。4).术中超声检测指标：术中探查肝脏，超声下测量各肝段门脉直径；于射频前、射频后不同时相点通过术中超声测定切除肝段(Ⅲ、Ⅴ、Ⅵ段)肝蒂门脉及动脉的血流速度；于射频前、射频后、术后14d测定保留肝段(Ⅱ、Ⅳ、Ⅶ、Ⅷ段)肝蒂门脉及动脉的血流速度；射频消融后超声下测定低回声区直径范围。5)病理情况：对切除肝段及术后14d切取保留肝段的肝组织均垂直于肝蒂血管长轴方向进行连续切片取材，每块组织厚度为0.5cm，所有组织进行HE染色，光镜下连续观察肝段胆管、门脉以及动脉的损伤情况3.统计学处理：计量资料以平均值±SD表示，采用SPSS16.0统计软件包进行单因素方差分析、独立样本t检验统计。以P<0.05为统计学差异。三、结果1.可行性分析：在超声引导下，冷循环电级均成功精确的穿刺到目标部位。本试验共选取了36个肝段肝蒂进行射频消融毁损，其中A组15个肝段缺血范围明显，3个肝段(1ofⅢ段，2ofⅤ段)缺血范围不明显，B组所有肝段缺血范围明显，效果明显好于A组(P<0.05)。所有肝蒂进行射频消融的时间为124.0±36.2S，其中A组射频消融时间为133.0±34.7S，B组射频消融时间为117±36.6S，两组间比较无明显差异（P>0.05)。2.动物手术、术后恢复及肝功能变化情况：所有动物手术过程顺利，术中无死亡。两组动物术后均健康存活14天，存活率100%。术后14天进行剖腹探查手术，腹腔探查未发现胆瘘、腹水、腹腔脓肿、腹腔出血等手术并发症。两组动物术后1dALT、AST、TBIL、DBIL水平显著升高，至术后14d血清学指标均恢复正常。3.术中超声检测指标：1)超声下测量各肝段门脉分支的平均直径为2.48±0.59mm；单因素方差分析组间比较无显著性差异（P>0.05）。2）射频消融后所有目标肝段门静脉的血流速度为0cm/s，动脉的血流速度较射频前显著降低（P<0.05），但大部分动脉于射频后仍可检测到血流存在。所有保留肝段的门脉及动脉于射频前、射频后、术后14d检测的血流速度无显著变化（P>0.05）。4.病理情况：对切除肝段标本连续切片取材HE染色后显微镜连续观察显示肝段门静脉的平均毁损长度为1.67±0.43cm，其中A组为1.64±0.38cm，B组为1.69±0.49cm，两组间无显著差异（P>0.05)。肝段动脉的平均毁损长度为1.33±0.40cm，其中A组为1.25±0.35cm，B组为1.42±0.43cm，两组间无显著差异（P>0.05)。肝段胆管的平均毁损长度为1.65±0.35cm，其中A组为1.67±0.34cm，B组为1.64±0.38cm，两组间无显著差异（P>0.05)。对于术后14天切除保留肝段标本，通过HE染色后显微镜连续观察发现其门脉、动脉及胆管组织均正常，未见损伤。进一步观察肝组织见肝细胞无浊肿，肝小叶结构完整，肝组织形态结构正常。四、结论1.本试验结合CourtF.G.报道的猪肝分段及脉管供应特征，对活体猪肝进行了详细的超声探查，明确了猪肝各肝段供应血管的超声影像特征，并选取了肝Ⅲ、Ⅴ、Ⅵ段肝蒂进行射频消融毁损，成功建立了超声引导下肝蒂射频消融肝段切除的动物模型。通过术中持续监测动物的平均动脉压及心率，统计手术出血量及手术时间，术后对动物存活状况的观察以及检测肝功能各项指标，结果证实了该肝段切除模型具有操作性强、安全性好、稳定性好、重复性高的优点。2.通过对比两组动物肝蒂毁损后肝段缺血边界的显示情况，结果显示在肝蒂射频消融的同时行肝动脉暂时阻断可以明显提高肝段显色的效果，同时亦能保障手术的安全性。3.为了可以通过术中超声定位对靶目标进行精准的穿刺，同时考虑到本技术方法需要对肝蒂血管进行毁损，本试验选用冷循环电极。实验中为充分利用射频消融能量，达到最理想的毁损的模式，我们紧贴目标肝蒂插入冷循环电级进行射频消融。结果证实穿刺成功率为100%，操作简单，对穿刺技术要求较低。通过连续切片取材观察毁损肝蒂血管以及胆管病理变化情况，我们认为为避免损伤邻近重要结构电极插入部位应距肝蒂分叉处的垂直距离至少为1.5cm。4.超声引导下射频消融肝蒂毁损是安全、可靠的，该项技术在精准肝切除领域中具有广阔的应用前景，值得在临床上进一步推广。更多还原

【Abstract】 BACKGROUNDWith people’s deeper understanding of liver anatomy, physiology and pathology as well as the further improvement of surgical technique, the concept of precision liver resection has been promoted recently. It pursues the result of the complete removal of the lesions, maximum preservation of functional liver, minimum blood loss and trauma invasiveness. For the hepatic lesions distributed by segmental-base, such as HCC and hepatolithiasis, anatomic segmentectomy is the most optimum surgical option to achieve the objectives of precision liver resection. The technique of segmentectomy was originally introduced by Makuuchi et al. It has been achieved equivalent long term survival to more radical resection techniques. Control of the inflow vessels and delineation of the relevant segment prior to resection has been achieved by several techniques. The most widely used technique is the puncture technique proposed by Makuuchi et al in1981. Indigo-carmine is injected into the vessel under IOUS-guidance, and the demarcated area for resection becomes evident on the liver surface. Other methods proposed include balloon catheters inserted transhepatically to occlude the feeding portal branch, or, more recently, through the mesenteric vein. The main limitation of these techniques is the high standard of skill required to puncture and occlude small vessels. Mazziotti et al proposed an intraparenchymal approach and ligation of the segmental Glissonian pedicle.This technique may be associated with intra operative blood loss and lengthy operation time. Navarra et al have recently suggested ablation of the segmental Glissonian pedicle by applying radiofrequency under ultrasound guidance. This was performed using the Habib sealer, a rectangular instrument using two rows of RFA probes. This technique however has certain limitations. It does not allow precise location of the target vessels and may be associated with inadvertent damage to other structures. The technique described is a modification of the Navarre procedure. It uses a single cooled-tip electrode, allowing precise targeting of the segmental vessels under IOUS guidance. It has several potential advantages:it is easy to operate and able to be performed by most HPB surgeons or radiologists familiar with intra operative ultrasound and liver anatomy; it minimizes intra operative blood loss and operative times; it may theoretically avoid tumour dissemination through the portal venous system by occlusion of the segmental vessels prior to parenchymal resection. This article reports an assessment of this technique in a preclinical study using a porcine animal model.MATERIALS AND METHODS1. Animals and Groups:The study was carried out on12Bama miniature pigs, with an average weight of25-35kg. The animals were divided into2groups. Group A: precise ablation of segmental glissonian pedicle by ultrasound-guided radiofrequency (6pigs); Group B:precise ablation of segmental glissonian pedicle by ultrasound-guided radiofrequency while the hepatic artery was clamped (6pigs).2. Surgical Procedure and Intraoperative Monitoring:According to the segmental nature of the porcine liver as delineated by Court F. G. et al, resections of segment Ⅲ, V and VI were performed.1)A J-shaped right subcostal incision was performed.2)Cholecystectomy was done routinely to eliminate the potential of radiofrequency induced injury.3)Intraoperative ultrasound (IOUS) was used to scan the intrahepatic vascular anatomy and identify the target liver segmental arterial and portal branches. Coagulative destruction of the feeding vessels was induced with the application of a "cooled-tip" RF with a500kHz-RF Generator. The target radiofrequency point in the segmental portal branch was chosen to be at least2cm distance from the segmental bifurcation. The radiofrequency was ceased when ischemic demarcation of the liver segment was observed. We took down the radiofrequency time and then measured the diameter of the hypoechoic region and the flow rate of the segmental arterial and portal branches which we had ablated and the adjacent liver segmental arterial and portal branches by ultrasound.4)The liver parenchyma is resected by the clamp-crush method without any form of hepatic inflow occlusion.For group B, the hepatic artery was identified and clamped temporarily during application of the RFA.3. Postoperative Monitoring:The animals were kept in boxes on-site14days after operation. Plasma samples were withdrawn from a central venous catheter on the first and14th postoperative days. Hepatocellular injury was assessed by standard laboratory assays of aspartate aminotransferase (AST) and alanine aminotransferase (ALT). Other recorded serum parameters were y-glutamyltransferase (y-GT), alkaline phosphatase(ALP), bilirubin, albumin.In the14th postoperative day, Laparotomy was done to explore any procedure related complications such as biliary fistula, ascites, hepatic abscess, peritoneal haemorrhage. The flow rate of the arterial and portal branches of the remnant segments was documented. A total hepatectomy was then done to observe the histological change of the remnant segmental arterial, portal and biliary tract branches.4. Histology:all of the segments including the target segments and the remnant segments which were resected in the14th postoperative day were sliced into thin sections of5-mm thick perpendicular to the glissonian pedicle long axis. Hematoxylin and eosin (H&E) staining was done to observe the pathologic changes of segmental arterial, portal and biliary tract branches.5. Data Analysis:The One-Way ANOVA test and the χ2test were used for the continuous and the categorical data, respectively. The significance level for all tests was fixed at P≤0.05. The statistical analyses were performed using spss16.0.2software.RESULTS1. Feasibility:In the12pigs, the precise puncture of the electrode was successfully applied without any technically-related inconvenience. A total of36hepatic segments were resected (18of group A,18of group B). In group A,15areas of marked discoloration on the surface of liver were obvious,3areas (1of segment Ⅲ,2of segment V) were not obvious (Fig.7). In group B, all of the areas of marked discoloration were obvious, the effect of demarcation was significantly better than that in group A (P＝0.000). Radiofrequency time was set at300seconds for the3 unobvious areas. Except these3areas, the average radiofrequency time was124.0±36.2seconds. The radiofrequency time was similar between the2groups (133.0±34.7versus117±36.6seconds; P>0.05).2. General Clinical and Biological Tolerance:Tolerance of the procedure was excellent and recovery was uneventful. No septic or cardiovascular, respiratory complications. Postoperative mortality was nil. In the14th postoperative day, laparotomy was performed, with no evidence of complications such as biliary fistula, ascites, hepatic abscess and peritoneal haemorrhage. Immediately after resection, all serum parameters of hepatocellular injury were elevated. Peak values were observed in all groups on the first postoperative day, and all parameters had normalized by day14.3. Blood Flow of the Target and Remnant Segmental Vessels:The flow rate of the target segmental portal branches was0cm/s after radiofrequency. The flow rate of the arterial branches was significantly reduced compared to before radiofrequency. However, most of the arterial branches blood flow could still be detected after radiofrequency. There was only1segmental arterial branch (segment V) with a flow rate of Ocm/s in group A versus6segmental arterial branches (3of segment Ⅲ,1of segment V and2of segment VI) in group B after radiofrequency. The flow rate of the arterial branches after radiofrequency in group B was lower than that in group A. In all groups, there was no significant difference in the flow rate of the segmental arterial and portal branches (remnant segments) among pre-radiofrequency, post-radiofrequency and14days after operation.4. Ultrasound Imaging: All of ablation zones in the liver were seen as a hypoechoic region with a central hyperechoic zone. The mean diameter of the hypoechoic regions measured by ultrasound was1.52±0.17cm (1.49±0.19versus1.55±0.14cm;P>0.05).5. Histology:Using standard hematoxylin and eosin staining, histologic evaluation of formalin fixed paraffin-embedded sections of the target segments revealed evidence of damage to the segmental glissonian pedicle. There was prominent heat injury in the tunica media and tunica intima of the portal veins. The injured portal veins measured1.67±0.43cm (group A:1.64±0.38versus group B:1.69±0.49cm;P>0.05). There was prominent heat injury in the tunica adventitia and the outer portion of tunica media of the hepatic artery, and there was mild injury in the endothelium. The damaged arteries measured1.33±0.40cm (group A:1.25±0.35versus group B:1.42±0.43cm; P>0.05). Severe injury was identified in the out portion of the tunica media and the epithelium of the bile duct; and less severe injury was identified in the inner portion of the tunica media. The injured bile ducts measure1.65±0.35cm (group A:1.67±0.34versus group B:1.64±0.38cm;P>0.05). Histologic evaluation of sections of the remnant segments which were resected on the14th postoperative day revealed that there was no injury to other adjacent arterial, portal and biliary tract branches.CONCLUSIONS1. The technique described in this article was evaluated on a porcine model. The feasibility and safety of the technique was confirmed. No mortality or morbidity occurred, Liver function tests of hepatocellular injury recovered by day14in both groups. There was no histological damage nor any significant change to the flow rate of the non targeted segmental arterial, portal or biliary duct structures observed to day14.2. Temporary occlusion of the hepatic artery adds further benefit in the effect of demarcation and there was no significant difference in the volume of ablation zones or hepatocellular damage between the two groups.3. Careful attention to technical details is essential. Considering mechanism of radiofrequency ablation, precise location of the tip using IOUS adjacent to the target vessel to make maximal use of the radiofrequency energy is essential and avoids the technically demanding intra luminal access as described by Makuuchi et al. Histologic evaluation of5-mm thick continuous sections revealed the mean ablated length of segmental arterial, portal and biliary tract branches was less than2cm. The probe in fact needs to be at least1.5cm away from vessels feeding segments not to be resected or structures that need to be left intact to avoid thermal injuries that can lead to postoperative vascular or biliary complications.4. The herein proposed technique of precise ablation of the segmental glissonian pedicle by ultrasound-guided radiofrequency has been used with success in experimental animals so far, confirming the technique to be quick, safe and effective. We believe that the technique could have a wide implication and potential in liver surgery.更多还原