Peribiliary gland damage due to liver transplantation involves peribiliary vascular plexus and vascular endothelial growth factor.

Extrahepatic bile ducts are characterized by the presence of peribiliary glands (PBGs), which represent stem cell niches implicated in biliary regeneration. Orthotopic liver transplantation may be complicated by non-anastomotic strictures (NAS) of the bile ducts, which have been associated with ischemic injury of PBGs and occur more frequently in livers obtained from donors after circulatory death than in those from brain-dead donors. The aims of the present study were to investigate the PBG phenotype in bile ducts after transplantation, the integrity of the peribiliary vascular plexus (PVP) around PBGs, and the expression of vascular endothelial growth factor-A (VEGF-A) by PBGs. Transplanted ducts obtained from patients who underwent liver transplantation were studied (N=62). Controls included explanted bile duct samples not used for transplantation (N=10) with normal histology. Samples were processed for histology, immunohistochemistry and immunofluorescence. Surface epithelium is severely injured in transplanted ducts; PBGs are diffusely damaged, particularly in ducts obtained from circulatory-dead compared to brain-dead donors. PVP is reduced in transplanted compared to controls. PBGs in transplanted ducts contain more numerous progenitor and proliferating cells compared to controls, show higher positivity for VEGF-A compared to controls, and express VEGF receptor-2. In conclusion, PBGs and associated PVP are damaged in transplanted extrahepatic bile ducts; however, an activation of the PBG niche takes place and is characterized by proliferation and VEGF-A expression. This response could have a relevant role in reconstituting biliary epithelium and vascular plexus and could be implicated in the genesis of non-anastomotic strictures.

Clinical Applications of Three-Dimensional Visualization Model of Arteries Supplying the Extrahepatic Bile Duct for Patients with Biliary Obstruction.

This study aimed to assess distribution characteristics and digital typing of arteries supplying the extrahepatic bile duct for patients with biliary obstruction, and evaluate the three-dimensional (3D) model in surgical decision-making. Forty-one patients with biliary obstruction were retrospectively evaluated. Clinical data obtained by 64-slice multidetector CT angiography scanning were introduced into Medical Image Three-Dimensional Visualization System; then, 3D model of extrahepatic bile duct and its supplying arteries were reconstructed. Based on the 3D model, the origination and bifurcations of the bile duct artery were observed, and the digital types established. Afterwards, plans for preoperative procedures were formulated. Finally, postoperative observations were performed and the biliary complications recorded in detail. The 3D model clearly displayed the origin, course, and distribution of individualized arteries supplying the extrahepatic bile duct, as well as variations. According to 3D model characteristics, the digital types were established. Blood supply to the superior segment of the extrahepatic bile duct encompassed 6 (14.6%), 17 (41.5%), 12 (29.3%), and 6 (14.6%) cases of Types IA, IB, IC, and II, respectively; meanwhile, blood supply to the inferior segment comprised 13 (31.7%), 13 (31.7%), 4 (9.8%), 7 (17.0%), and 4 (9.8%) cases of Types IA, IB, IC, II, and III, respectively. This classification helped in preoperative surgical planning and corroborated intraoperative findings. No postoperative biliary complications were recorded. The 3D model reconstructed using Medical Image Three-Dimensional Visualization System displayed individualized anatomical structures of the extrahepatic bile duct and associated blood supplying arteries, and could contribute to preoperative surgical planning.

Computed tomography virtual endoscopy with angiographic imaging for the treatment of type IV-A choledochal cyst.

Type IV-A choledochal cysts (CCs) are a congenital biliary anomaly which involve dilatation of the extrahepatic and intrahepatic bile ducts. We present the case of a 30-year-old woman with type IV-A CC, on whom three-dimensional computed tomography (3D CT) and virtual endoscopy were performed. 3D CT revealed partial dilatation in the posterior branch of the intrahepatic bile duct and a relative stricture between it and the extrahepatic bile duct. Virtual endoscopy showed that this stricture was membrane-like and separated from the surrounding blood vessels. Based on these image findings, complete cyst resection, bile duct plasty for the stricture, and hepaticojejunostomy were safely performed. To the best of our knowledge, there are no reports of imaging by virtual endoscopy of the biliary tract which show the surrounding blood vessels running along the bile duct.

Indocyanine green reinjection technique for use in fluorescent angiography concomitant with cholangiography during laparoscopic cholecystectomy.

BACKGROUND: For safe laparoscopic cholecystectomy, surgeons must possess detailed knowledge of the anatomy of the bile duct and arterial system as seen through a laparoscope. STUDY DESIGN: We developed an indocyanine green (ICG) reinjection technique for use in fluorescent angiography. Here, we evaluated the efficacy of the ICG reinjection technique in fluorescent angiography in discriminating the arterial system with the concomitant use of fluorescent cholangiography. RESULTS: Twenty-eight patients were enrolled in the study. All patients underwent laparoscopic cholecystectomy without complication. After reinjection of ICG during surgery, fluorescence of the cystic artery was visualized in 25 patients (89%). CONCLUSIONS: Fluorescent angiography using this ICG reinjection technique might enhance the safety of laparoscopic cholecystectomy.

Development of an experimental model of cholestasis induced by hypoxic/ischemic damage to the bile duct and liver tissues in infantile rats.

BACKGROUND: We aimed to develop experimental models of hypoxia/ischemia-induced cholestasis using neonatal and infantile rats. METHODS: Hypoxia/ischemia was induced in the bile duct (BD) by injecting prostaglandin (PG) at birth and/or by coagulation of the hepatic artery (CHA) at about 3 weeks after birth. The rats were divided into 6 groups: control; PG-injected; sham-operated with or without PG; CHA; and CHA + PG. CHA was also performed in adult rats. Liver specimens and blood samples were obtained at 5 weeks after birth, and immunohistochemical and biochemical examinations were performed. RESULTS: (1) BD proliferation with fibrosis (BDPF) was found in the intrahepatic portal tract in the CHA and CHA + PG groups. Low-grade BDPF was observed in the PG group. (2) Cyst formation in the extrahepatic BD (EBD) was observed in the porta hepatis of some rats in the CHA and CHA + PG groups. In these groups, the number of peribiliary vascular plexuses (PVPs) decreased. BD proliferation and infiltration of inflammatory cells were observed in the EBD wall in the CHA + PG group. (3) Ki-67 was expressed in BD and EBD cells in the CHA + PG group. (4) BDPF was not detected in adult rats with CHA. (5) Serum liver function tests indicated obstructive changes in the EBD in the CHA and CHA + PG groups. CONCLUSION: Reduced blood flow in the EBD during infancy induced BDPF and obstructive changes in the EBD, which may, along with immature PVP and inflammatory changes in the EBD, contribute to hypoxia/ischemia of the EBD.

Anastomosis between the hepatic artery and the extrahepatic collateral or between extrahepatic collaterals: observation on angiography.

Transcatheter arterial chemoembolisation for hepatocellular carcinoma is widely carried out not only through the hepatic artery but also through the extrahepatic collateral pathways. Anatomically, there are many anastomoses between the hepatic artery and the extrahepatic collateral as well as among the extrahepatic collaterals. However, these anastomoses may not be shown on angiography because the anastomosing branches are too small. These anastomoses may not only interfere with effective control of hepatocellular carcinoma by transcatheter arterial chemoembolisation but also cause unexpected procedure-related complications. Therefore, radiologists should have sufficient knowledge of these underlying anastomoses. In this report, we present our angiographic images.

Hilar liver resection in dogs.

OBJECTIVE: To describe hepatic vasculobiliary anatomy important to hilar liver lobe resection in the dog. STUDY DESIGN: Experimental study. ANIMALS: Canine cadavers (n=7). METHODS: The vasculobiliary system of 7 fresh canine livers was injected with a polymer. The parenchyma was dissected at the level of the hilus to determine the vascular and biliary supply to each liver lobe, and then macerated with a corrosion preparation. The information gathered was used to describe a surgical approach for hilar liver lobe resection. RESULTS: Each liver lobe had a single hepatic artery and biliary duct. The location of these structures was consistent, although minor variations existed (dorsal versus ventral to the lobar portal vein) in the left lateral lobe and papillary process in 2 specimens. Most liver lobes (34/49) were supplied by 1 lobar portal vein and drained by 1 lobar hepatic vein (39/49). The location of the portal and hepatic veins was consistent among specimens. CONCLUSIONS: The left division is the most mobile of the liver lobes and each lobe can be removed separately or en bloc. Because of the location of the hepatic veins, the central division is best removed as a single unit. The right lateral lobe can be removed individually or together with the caudate process. The papillary process is removed by itself. CLINICAL RELEVANCE: A hilar liver lobectomy technique can provide an alternative approach to conventional procedures for tumors that encroach upon the hilus of the liver.

Evaluation of VEGF A expression and microvascular density as prognostic factors in extrahepatic cholangiocarcinoma.

OBJECTIVES: Angiogenesis is essential for tumor growth and metastasis. An association between microvessel density, a measure of tumor angiogenesis, and conventional prognostic variables has been shown for many different tumor entities. In extrahepatic cholangiocarcinoma, the VEGF expression and microvessel density have rarely been investigated. METHODS: Paraffin-embedded specimens from 51 resected adenocarcinomas of the extrahepatic bile duct were immunostained for vascular endothelial growth factor A (VEGF A) and CD 34 to evaluate the microvessel density (MVD). VEGF A staining was evaluated by combining intensity and percentage of positive tumor cells, as low (expression equal or below the median), or high (above the median). Microvessel density was assessed using a method published by Weidner et al. RESULTS: Median disease free survival (DFS) of the study group was 12.5 months (range, 1-66.3 months). DFS was calculated in the 39 patients with complete resection. It was significantly better in patients with low microvessel density than DFS in patients with high microvessel density (33 months (range, 3-66.3 months) vs. 21.8 months (range, 1.6-31.6 months); p=0.022). In contrast, VEGF A expression did not correlate with survival. There was a trend toward a higher VEGF A expression in highly vascularized tumors (p=0.08), but failed to reach statistic significance. CONCLUSIONS: The present study indicates, that vascularisation has an important impact on survival of extrahepatic cholangiocarcinoma patients. Other molecules than VEGF A are probably involved in neovascularization in extrahepatic cholangiocarcinoma.

Reconsideration of the histologic definitions used in the pathologic staging of extrahepatic bile duct carcinoma.

The histologic boundaries of the extrahepatic bile ducts (EBDs) are not well defined, despite the fact that pathologic staging of carcinomas arising in these structures requires the determination of extent of tumor invasion in this area. Perhaps in part, because the smooth muscle band in the EBD wall is not well formed throughout the length of these structures, a previous definition of the outer portion of the bile duct wall included "loose, richly vascularized connective tissue, interlaced with large nerve fibers." We have experienced difficulty in the application of these criteria in staging EBD carcinomas, which requires the histologic determination of the extent of the EBD wall. To systematically study the histologic features of EBD tissue boundaries, 34 EBD specimens obtained from autopsy were analyzed with attention to the distribution of blood vessels and nerve fibers along the length of this system. The EBD specimens were divided into lower, middle, and upper portions, and the locations of blood vessels and nerve fibers were then analyzed separately at each location. We defined the fibromuscular wall as the dense concentric arrangement of collagen and smooth muscle fibers surrounding the EBD mucosa. The location of blood vessels and nerve fibers was categorized as either (1) within, (2) junctional to, or (3) outside of the fibromuscular wall. Blood vessels and nerve fibers are located predominantly outside of the fibromuscular wall and are usually surrounded by adipose tissue throughout the entire EBD, however, their distribution in this location is not consistent. Because of these histologic features, we propose that the bile duct wall is more precisely defined as occurring between the mucosa and the outermost boundary of dense fibromuscular tissue, without consideration of the presence or absence of blood vessels and nerve fibers.

The finest branches of the biliary tree might induce biliary vascularization necessary for biliary regeneration.

BACKGROUND/AIMS: The finer branches of the biliary tree play an important role in biliary regeneration. They are consistently escorted by microvessels. Defects in the vascularization of these structures could impair bile duct regeneration. Therefore, we investigated the pattern of the escorting microvessels during the development of bile duct loss in the human liver, using chronic rejection as a model. METHODS: The number of interlobular bile ducts, bile ductules and extraportal biliary cells with and without escorting microvessels and the expression of VEGF-A were studied in follow-up biopsies of 12 patients with chronic rejection and 16 control patients with acute rejection without progression to chronic rejection. RESULTS: The controls showed a proliferation of bile ductules at 1-week and 1-month. Proliferation of bile ductules without microvessels preceded proliferation of bile ductules with microvessels. Proliferation of the microvascular compartment followed biliary proliferation. This sequence of events was not observed in the chronic rejection group, in which all biliary structures decreased in time. VEGF-A expression was increased at 1-week and 1-month in both groups. CONCLUSIONS: An immediate proliferative response of the finer branches of the biliary tree followed by proliferation of the microvascular compartment after biliary injury seems to be a prerequisite for bile duct regeneration.

Angioarchitecture of the rabbit extrahepatic bile ducts and gallbladder.

The angioarchitecture of extrahepatic bile ducts and gallbladder of the miniature rabbit was studied by scanning electron microscopy (SEM) of vascular corrosion casts. Light microscopy of Masson-stained, paraffin-embedded transverse tissue sections served to attribute cast vascular structures to defined layers of bile ducts and gallbladder. In all segments of the bile tract, a mucosal and a subserosal vascular network was found. In glandular segments, the mucosal network was composed of a meshwork of subepithelial and circumglandular capillaries, which serve the mucosal functions. Differences in the angioarchitectonic patterns existed only in the subserosal networks as hepatic ducts own one supplying arteriole only, while the common bile duct owns a well-defined rete arteriosum subserosum. A well-developed dense subserosus venous plexus was present throughout the bile tract. Vascular patterns of the gallbladder body resembled those of the bile duct, whereby the dense subserous venous plexus was located close to the mucosal capillary network. The subserosal network in the neck of the gallbladder resembled that of the cystic duct. Spatial changes of the mucosal vascular network during volume changes of the gallbladder were documented. Measurements from tissue sections revealed bile tract diameters of 220-400 microm (extrahepatic ducts), 500-650 microm (cystic duct), and 4-6 mm (common bile duct). Data gained from high-powered SEM micrographs of vascular corrosion casts revealed vessel diameters of 200 microm (cystic artery), 90-110 microm (cystic vein), 30-40 microm (feeding arterioles), and 25-110 microm (subserosal venules). Crypt diameters in the filled gallbladder were 300-1,500 mum; those in the contracted organ were 100-600 microm.

Histological characteristics of tumor cells and stromal cells in vessels and lymph nodes are important prognostic parameters of extrahepatic bile duct carcinoma: a prospective study.

Extrahepatic bile duct carcinomas (EHBDCs) consist of primary tumors, tumors in vessels, and tumors in lymph nodes. The purpose of this study was to prospectively investigate whether the histological characteristics of tumor cells and tumor stromal cells in vessels and lymph nodes were significantly associated with the outcomes of 60 EHBDC patients as compared with the histological characteristics of tumor cells and tumor stromal cells in primary tumors. Multivariate analyses, using the Cox proportional hazard regression model, showed that in EHBDCs without nodal metastasis, blood vessel tumor emboli with an angiomatous stroma significantly increased the hazard ratios (HRs) of tumor recurrence and death ( P < .05). In EHBDCs with nodal metastasis, the presence of tumor necrosis in the nodal tumors significantly increased the HRs of tumor recurrence and initial distant organ metastasis ( P < .05). In EHBDCs located in the distal to middle portion of the extrahepatic bile duct, blood vessel tumor emboli with an angiomatous stroma significantly increased the HRs of tumor recurrence, initial distant organ metastasis, and death ( P < .05). Severe nuclear atypia of the tumor cells in lymph vessels significantly increased the HRs of tumor recurrence and initial distant organ metastasis ( P < .05). In EHBDCs located in the hilar portion of the extrahepatic bile duct, the presence of nodal tumors with more than 4 mitotic figures significantly increased the HRs of tumor recurrence and initial distant organ metastasis ( P < .05). Several histological characteristics of tumor cells and tumor stromal cells in vessels and lymph nodes have significant effects on tumor progression of EHBDCs.

Cholangiojejunal fistula caused by bile duct stricture after intraoperative injury to the common hepatic artery.

A 68-year-old man, admitted for the treatment of recurrent cholangitis after a pancreatoduodenectomy (PD) performed 3 years previously was diagnosed as having multiple hepaticolithiasis. On laparotomy, the hepatic artery was not recognized. The anastomosed common hepatic duct was obstructed, and a fistula had been formed between the right hepatic duct and the Roux limb of the jejunum. Lithotripsy was performed from this fistula and it was reanastomosed. Angiography was performed postoperatively and it revealed common hepatic artery injury, most likely to have occurred during the previous PD. The patient's postoperative course was uneventful and he has been asymptomatic for 8 months after the operation, indicating that reanastomosis of the fistula can be an effective method. The stricture of the anastomosis was suspected to be mainly due to cholangial ischemia, because no episode of anastomotic leak or retrograde biliary infection had occurred during the PD perioperative period. There are several reports of late stricture of anastomosis 5 or more years after cholangiojejunostomy. This patient, therefore, requires further long-term follow up.

Surgical resection combined with chemotherapy for advanced hepatocellular carcinoma with tumor thrombus: report of 19 cases.

BACKGROUND: Prognosis of hepatocellular carcinoma (HCC) with tumor thrombus in the main portal vein (MPV), inferior vena cava (IVC), or extrahepatic bile duct (EBD) treated by conventional therapies has been considered poor. This study aimed to evaluate the efficacy of hepatic arterial infusion chemotherapy after surgical resection as an adjuvant therapy or as a treatment for intrahepatic recurrence of HCC with tumor thrombus in MPV, IVC, or EBD. METHODS: Nineteen patients with HCC and tumor thrombus in the MPV, IVC, or EBD who underwent hepatectomy with thrombectomy were reviewed retrospectively. RESULTS: The overall 3-year survival rate was 48.5%. Two patients with postoperative residual tumor thrombus died within 6 months owing to rapid progression of the residual tumor thrombus. Five patients survived more than 5 years after their operations. Tumors disappeared completely in 3 patients after hepatic arterial infusion chemotherapy with a combination of cisplatinum and 5-fluorouracil, and the longest survival period was 17 years and 11 months in a patient with EBD thrombus. CONCLUSIONS: If hepatic reserve is satisfactory, an aggressive surgical approach combined with chemotherapy seems to be of benefit for patients having HCC with tumor thrombus in the MPV, IVC, or EBD.

Preliminary study of the anatomy of the venous drainage of the intrahepatic and extrahepatic bile ducts and its relevance to the practice of hepatobiliary surgery.

BACKGROUND: Although there have been many studies of the arterial supply of the biliary system, attempts to study the corresponding venous drainage have been few and all have been incomplete. The purpose of the present investigation is to describe the anatomy of the venous drainage of both the intrahepatic and extrahepatic bile ducts and to determine its relevance to hepatobiliary surgery. METHODS: The intrahepatic and extrahepatic venous drainage of the bile ducts was investigated in seven specimens by injecting a solution of 10% gelatin coloured with Alcian blue into the portal vein or the superior mesenteric vein to outline the venous drainage. The specimens were dissected under loop magnification and representative drawings were obtained. RESULTS: The surface of the intrahepatic and extrahepatic bile ducts was covered by a fine venous plexus. On the surface of the supraduodenal common hepatic duct and common bile duct the venous plexus drained laterally into marginal veins, usually two in number and known as the 3 o'clock and 9 o'clock marginal veins. Inferiorly the marginal veins and the venous plexus communicated with the pancreaticoduodenal venous plexus, which in its turn drained into the posterosuperior pancreaticoduodenal vein, a branch of the superior mesenteric vein. Superiorly the marginal veins divided into a number of branches. Some branches followed the left and right hepatic ducts into the liver, communicating with the venous plexus and the adjacent branches of the portal vein. Other branches of variable size entered either segment IV or the caudate lobe or process via the hilar venous plexus. A most important finding was that even after dividing the bile duct and all communicating veins at the upper border of the duodenum, the venous plexus and the marginal veins filled normally to the level of transection. This occurred almost certainly by retrograde filling from above. CONCLUSION: The satisfactory results of end-to-end anastomosis in whole liver transplantation depends partly on the presence of adequate venous drainage. This has been amply demonstrated by the injection studies. This would indicate that the poor results of end-to-end repair of the bile duct after surgical trauma results from other factors such as poor technique, devascularization of the cut ends due to trauma, and carrying out the anastomosis under tension. After resection of the hilum for cholangiocarcinoma the venous drainage of the left and right hepatic ducts and their branches depends mainly on the communications between the venous plexus on the ducts and the adjacent branches of the portal vein, even at a lobular or sinusoidal level. The satisfactory results obtained after anastomosis of the left and right hepatic ducts or their branches to a Roux loop ofjejunum attest to this. This applies also to the transplantation of segments II and III in paediatric patients from related adult donors and in patients receiving split liver transplants. Finally, the venous drainage at the bifurcation of the common hepatic duct has been shown to enter the caudate lobe and segment IV directly. This suggests that a hilar cholangiocarcinoma may metastasize to these segments, and perhaps partly explain the significantly better long-term results when the caudate lobe and segment IV are resected en bloc with the cholangiocarcinoma as part of modern radical surgery for this condition.

Pseudo-obstruction of the extrahepatic bile duct due to artifact from arterial pulsatile compression: a diagnostic pitfall of MR cholangiopancreatography.

PURPOSE: To evaluate the frequency of artifact from arterial pulsatile compression as the cause of pseudo-obstruction of the extrahepatic bile duct at magnetic resonance (MR) cholangiopancreatography (MRCP) and specify the causative vessels. MATERIALS AND METHODS: In 234 patients (102 men, 132 women; age range, 25-80 years), MRCP images obtained by using a single-shot turbo spin-echo sequence were reviewed to assess pseudo-obstruction of the extrahepatic bile duct caused by vascular compression. Dual-phase spiral computed tomography, contrast material-enhanced three-dimensional MR angiography, and/or digital subtraction angiography also were performed to determine the vessel that caused the pseudo-obstruction. RESULTS: Thirty-six pseudo-obstructions due to vascular compression were found in 33 (14%) patients. The common hepatic duct (27 [75%] sites) was the most common pseudo-obstruction site, followed by the left hepatic duct (four [11%] sites), proximal common bile duct (three [8%] sites), and right hepatic duct (two [6%] sites). The causative vessels were identified as the right hepatic artery at 24 (67%) sites; gastroduodenal artery, two (6%) sites; cystic artery, two (6%) sites; proper hepatic artery, one (3%) site; and an unspecified branch of the common hepatic artery, seven (19%) sites. CONCLUSION: At MRCP, pseudo-obstruction of the extrahepatic bile duct can be caused by pulsatile vascular compression of the hepatic and gastroduodenal arteries, and it should not be misdiagnosed as a bile duct tumor or biliary stone.

Laparoscopic cholecystectomy, Calot's triangle, and variations in cystic arterial supply.

BACKGROUND: The extrahepatic biliary tree with the exact anatomic features of the arterial supply observed by laparoscopic means has not been described heretofore. Iatrogenic injuries of the extrahepatic biliary tree and neighboring blood vessels are not rare. Accidents involving vessels or the common bile duct during laparoscopic cholecystectomy, with or without choledocotomy, can be avoided by careful dissection of Calot's triangle and the hepatoduodenal ligament. METHODS: We performed 244 laparoscopic cholecystectomies over a 2-year period between January 1, 1995 and January 1, 1997. RESULTS: In 187 of 244 consecutive cases (76.6%), we found a typical arterial supply anteromedial to the cystic duct, near the sentinel cystic lymph node. In the other cases, there was an atypical arterial supply, and 27 of these cases (11.1%) had no cystic artery in Calot's triangle. A typical blood supply and accessory arteries were observed in 18 cases (7.4%). CONCLUSION: Young surgeons who are not yet familiar with the handling of an anatomically abnormal cystic blood supply need to be more aware of the precise anatomy of the extrahepatic biliary tree.

Analysis of the arterial supply of the extrahepatic bile ducts and its clinical significance.

The purpose of this study is to describe the arterial supply of the entire extrahepatic bile duct system. The cross-sectional area of all arteries that supply the ducts is measured under an operating microscope in 50 adult cadavers injected with red latex through the aorta. The extrahepatic bile duct system is divided into four topographic portions: cystic duct and gallbladder, right and left hepatic ducts, bile (common) duct and including its supra-retroduodenal parts, and the pancreatic and intraduodenal portions. The arterial supply to each portion is carefully detailed. The ducts are supplied by more than seven arteries, of which the major arteries are the cystic artery, posterior superior pancreaticoduodenal artery, right hepatic artery, and retroportal artery. Collectively they provide 94.5% of the blood supply to the ducts. Arteries form three types of anastomotic patterns on the walls of the ducts, suggesting that ductal incisions can be made in ways that least disturb the blood supply. The patterns are: a network, a longitudinal anastomotic chain, and an arterial circle. These data emphasize the importance of the arterial supply in biliary surgery and especially the treatment of hemobilia.

Ischemic biliary stricture due to lymph node dissection in the hepatoduodenal ligament.

We report on 2 cases of ischemic biliary stricture after radical lymphadenectomy in the hepatoduodenal ligament with skeletonization of the extrahepatic bile ducts for malignant diseases. In both cases, histologic examination of the subsequently resected biliary strictures revealed evidence of ischemia. Skeletonization of the bile ducts may result in ischemia and stricture formation. In patients who require thorough dissection of the hepatoduodenal lymph nodes, bile duct resection should be considered.

Assessment of portal vein invasion by bile duct cancer using intraductal ultrasonography.

BACKGROUND AND STUDY AIMS: We recently reported on the contribution of intraductal ultrasonography (IDUS) to the regional staging of bile duct cancer, and we present here the first detailed study of the value of IDUS in assessing the portal vein invasion by bile duct cancer. PATIENTS AND METHODS: Preoperative assessment of portal vein invasion was performed by IDUS via a percutaneous tract or via transpapillary route in 18 patients with extrahepatic bile duct cancer. Various probes, with diameters of 1.4, 2.0, 2.4, 2.6, and 3.2 mm, and frequencies of 7.5, 15, 20, and 30 MHz, were used. All patients additionally underwent endoscopic ultrasonography (EUS) and angiography. In the first six cases, the IDUS and EUS images were analyzed retrospectively without the knowledge of operative results or the other imaging tests. In the remaining 12 cases, IDUS and EUS images were prospectively reviewed prior to surgery, without knowledge of the angiographic findings. The gold standard for the results of IDUS, EUS and angiography was the histopathological findings in 17 resected tumors, and the intraoperative findings in one patient who did not undergo resective surgery. RESULTS: IDUS was able to demonstrate the portal vein in all cases. Its accuracy in diagnosing portal vein invasion was 100% for all locations. EUS was useful in assessing portal vein invasion at the middle and distal bile duct (the accuracy was 91%), but was not useful in assessing invasion at the proximal bile duct (the accuracy was 57%). CONCLUSIONS: IDUS proved useful for assessing the extension of cancer invasion into the portal vein, even in proximal bile duct tumors.