Search

Background Whether operative treatment provides benefits for elderly patients with perihilar cholangiocarcinoma is unknown. The aim of this study was to review resection of perihilar cholangiocarcinoma according to age and to clarify its value for octogenarians. Methods Between April 1977 and December 2015, we reviewed consecutive patients who underwent resection for perihilar cholangiocarcinoma with a special focus on patient age. Results During the study interval, 831 patients underwent resection for perihilar cholangiocarcinoma. The median age of the resected patients increased by 11 years over approximately 40 years. Before 2001, no octogenarians underwent operative intervention; however, the proportion of operations for octogenarians increased to 9% after 2010. Further analyses were performed on 643 resected patients between 2001 and 2015. The resectability rate was not different between the octogenarians and the other age groups (71% vs 72.4%). The Charlson Comorbidity Index and preoperative laboratory data were similar between the 2 groups. A less advanced tumor was a predominant feature in the octogenarians compared to the other age groups. Consequently, the procedure used in the octogenarians were less extensive, but the proportion of R0 resection was greater in the octogenarians than in the other age groups (95.% vs 78.3%, P = .008). The ratio of patients who died of other diseases was also greater among octogenarians (29% vs 6.0%, P Conclusion Resection of perihilar cholangiocarcinoma can be performed with low mortality irrespective of age with careful patient selection and offers long-term survival even in octogenarians.

Background Percutaneous transhepatic portal vein embolization (PTPE) can increase the future liver remnant (FLR) volume before extended liver resection; however, there is no current consensus regarding the best embolic material for PTPE. Purpose To evaluate the efficacy of PTPE using gelatin sponge particles and coils. Material and Methods The medical records of 136 patients who underwent PTPE using gelatin sponge particles and metal coils were retrospectively reviewed. We evaluated the procedural details, liver volume on CT, and clinical status before and after PTPE. Results The mean FLR volume increased significantly from 390 ± 147 cm3 to 508 ± 141 cm3 ( P Conclusion PTPE with gelatin sponge particles and coils may impose low physical stress on patients and is a safe method of inducing a significant increase of FLR.

Background/aim Trifluridine/tipiracil (FTD/TPI) is used for metastatic colorectal cancer, that is refractory to 5-fluorouracil (5-FU)-based therapies. However, the impact of pre-exposure to 5-FU on the anti-cancer effect of FTD, which is a key component of FTD/TPI, is unclear. Materials and methods The incorporation into DNA and anti-cancer activity of FTD were analyzed in several cancer cell lines under response to FTD treatment with or without 5-FU pre-exposure. The volumes of tumors in xenografted nude mice were examined among groups that were either untreated or treated with S-1, FTD/TPI or FTD/TPI with pre-exposure to S-1. Results Pre-exposure to 5-FU significantly increased FTD incorporation into DNA and enhanced its anti-cancer effect for viability and proliferation of cancer cells. In the xenograft nude mouse model, the tumor volumes in the FTD/TPI-treated and S-1-pre-exposed group were lower than those in the FTD/TPI-only-treated group. Although both FTD dose and exposure time in the FTD/TPI-treated and S-1-pre-exposed mice were smaller than those in the FTD/TPI-only-treated mice, the incorporated FTD in the tumors in the former group was 86.5% of that in the latter group. Conclusion Pre-exposure to 5-FU enhanced the incorporation into DNA and the anti-cancer effect of FTD in the context of colorectal cancer. Our data indicate the potential for a new sequential therapy using S-1 and FTD/TPI to improve prognosis of colorectal cancer.

Background Cholangiolocellular carcinoma (CoCC) is a rare liver tumor arising from the canals of Hering found between the cholangioles and interlobular bile ducts. Although morphologically CoCC mimics intrahepatic cholangiocarcinoma (ICC), CoCC exhibits a unique intermediate biologic behavior between hepatocellular carcinoma (HCC) and ICC. Curative resection is required for prolonged survival in patients with CoCC. However, effective therapy for postoperative hepatic recurrence has not yet been standardized. Case presentation A 40-year-old man had an asymptomatic liver mass found during a regular medical examination. Contrast-enhanced computed tomography revealed a well-enhanced mass, 15 cm in diameter, in the right liver. He underwent right hemihepatectomy at a local hospital under the preoperative diagnosis of hepatocellular carcinoma. Pathologic examination confirmed a moderately differentiated tubular adenocarcinoma, leading to a diagnosis of ordinary ICC. Twelve months after surgery, he was referred to our hospital due to three hepatic recurrences in the left medial segment. He underwent partial hepatectomy for the recurrence, followed by adjuvant chemotherapy using gemcitabine alone. After the second hepatectomy, hepatic recurrences developed an additional seven times. The numbers and sizes of the recurrent tumors were very limited at each recurrence, satisfying the standard criteria for percutaneous radiofrequency ablation (RFA) for the treatment of HCC. All lesions were treated by percutaneous RFA, although this was an exceptional approach for ICC. He is now alive without evidence of disease 9.2 years after the first hepatectomy. Because his clinical outcome was satisfactory and not compatible with the typical negative outcomes of ordinary ICC, we re-reviewed the histological findings of his tumor. The tumor was composed of small gland-forming cells proliferating in an anastomosing pattern; the cell membrane was strongly immunoreactive for epithelial membrane antigen. These findings were in accordance with the typical features of CoCC, revising his final diagnosis from ICC to CoCC. Conclusions This case report demonstrates a satisfactory outcome using repeated local treatments, such as hepatectomy and RFA, for hepatic recurrences of CoCC, suggesting that a localized treatment approach can be considered to be a therapeutic option. We should be careful in making a definitive diagnosis of ICC and ruling out CoCC because the diagnosis potentially dictates the treatment strategy for recurrences.

The present study aimed to assess the safety and feasibility of laparoscopic extended pelvic surgery for primary or recurrent rectal cancer. The data on 77 patients, who underwent extended pelvic surgery between February 2008 and June 2014, were retrospectively analyzed. The patients were divided, based on their treatment history, into an open surgery (OS) group (n = 41) and a laparoscopic surgery (LS) group (n = 36). The operative time in the LS group was significantly longer than that in the OS group (766 vs. 561 min; p

The diagnostic accuracy of biliary cytology is limited. A novel sandwich enzyme-linked immunosorbent assay that combined Wisteria floribunda agglutinin (WFA) and anti-sialylated mucin 1 (MUC1) monoclonal antibody to target bile samples was recently developed. This study was designed to verify the diagnostic accuracy of WFA-sialylated MUC1 as a sensitive biliary biomarker for human biliary tract cancer. Bile samples from 27 patients with benign disease and 174 patients with biliary tract cancer were analyzed. A receiver-operated characteristic curve analysis for biliary WFA-sialylated MUC1 and serum CA19-9 levels was performed to determine the cutoff value for the prediction of the presence of biliary tract cancer. Biliary WFA-sialylated MUC1 levels were significantly higher in the biliary tract cancer group compared with the benign group (P

Right-sided hepatectomy is often selected for perihilar cholangiocarcinoma, due to the anatomic consideration that “the left hepatic duct is longer than that of the right hepatic duct”. However, only one study briefly mentioned the length of the hepatic ducts. Our aim is to investigate whether the consideration is correct. In surgical study, the lengths of the resected bile duct were measured using pictures of the resected specimens in 475 hepatectomized patients with perihilar cholangiocarcinoma. In radiological study, the estimated lengths of the bile duct to be resected were measured using cholangiograms reconstructed from computed tomography images in 61 patients with distal bile duct obstruction. In surgical study, the length of the resected left hepatic duct was 25.1 ± 6.4 mm in right trisectionectomy (n = 37) and 14.9 ± 5.7 mm in right hepatectomy (n = 167). The length of the right hepatic duct was 14.1 ± 5.7 mm in left hepatectomy (n = 149) and 21.3 ± 6.4 mm in left trisectionectomy (n = 122). In radiological study, the lengths of the bile duct corresponding to the surgical study were 34.1 ± 7.8, 22.4 ± 7.1, 20.8 ± 4.8, and 31.6 ± 5.3 mm, respectively. Both studies determined that the lengths of the resected bile ducts were (1) similar between right and left hepatectomies, (2) significantly shorter in right hepatectomy than in left trisectionectomy, and (3) the longest in right trisectionectomy. The aforementioned anatomical assumption is a surgeon’s biased view. Based on our observations, a flexible procedure selection is recommended.

Inflammation-based prognostic scores have prognostic value in several kinds of cancer. However, little is known about their value in perihilar cholangiocarcinoma. We evaluated whether inflammation-based prognostic scores are associated with survival of patients with perihilar cholangiocarcinoma. Inflammation-based scores (i.e., the modified Glasgow Prognostic Score (mGPS), neutrophil–lymphocyte ratio, platelet–lymphocyte ratio, and prognostic nutritional index) were retrospectively evaluated in 534 patients who underwent resection for perihilar cholangiocarcinoma. Blood samples obtained 1–3 days before surgery after jaundice had fully resolved with biliary drainage and after cholangitis had subsided were used to obtain the scores. Of the four scores evaluated, the mGPS showed prognostic value, whereas the remaining three scores did not. Patients with an mGPS of 0 had significantly better survival than patients with an mGPS of 1 or 2 (41.9 % vs 26.3 % at 5 years, P

Background Transhepatic portal vein (PV) stenting has been shown to be one of the most important treatments for patients with PV stenosis caused by hepatopancreatobiliary malignancy. Methods Ten consecutive patients with PV stenosis caused by the recurrence of a perihilar biliary malignancy underwent transhepatic PV stenting. A self-expandable metallic stent was deployed at the stenosis site. The patients were retrospectively analyzed with regard to the procedure, complications, and survival after the stent placement. Results The median interval between the primary resection and the PV stenting was 22 months. The initial hepatic resection was a left trisectionectomy with caudate lobectomy in seven patients, a left hepatectomy with caudate lobectomy in one patient, a right anterior sectionectomy with caudate lobectomy following a left hepatectomy in one patient and a partial liver resection in one patient. The angle of the PV around the stenosis was greater in the patients with PV stenosis located in the right posterior PV. Eight patients with successful PV stent placement were able to receive anticancer treatment, with a median survival of 14 months. The remaining two patients without successful PV stent placement survived less than 6 months. Conclusions Portal vein stenting might offer relief from the symptoms associated with PV hypertension and the opportunity for sustainable anticancer therapy in patients with recurrent perihilar biliary malignancy.

When compared with other diseases, few authors have reported on single-incision laparoscopic surgery (SILS) for liver cysts. We herein describe our experience with SILS for a huge liver cyst with the application of an umbilical Z-shaped incision using a gel port and a high-density monopolar saline-cooled radiofrequency device with a successful outcome. An 80-old-year woman was diagnosed with a huge liver cyst with abdominal pain. She underwent percutaneous drainage of the liver cyst and injection of both absolute ethanol and an antimicrobial agent into the liver cyst at the previous hospital. Because of re-expansion of the liver cyst and symptom recurrence, we performed SILS for the liver cyst. An umbilical Z-shaped incision was made for gel port placement. After aspiration of the cystic fluid and dissection of the thin cystic wall by laparoscopic coagulating shears, the thick cystic wall was divided using an endoscopic linear stapler to avoid bleeding and bile leakage. After wide fenestration, a high-density monopolar saline-cooled radiofrequency device was applied for the ablation of the remnant membrane of the cystic wall. All maneuvers could be performed only through the gel port. The patient was discharged 4 days after surgery and was satisfied with the cosmetic results. This case shows that the application of an umbilical Z-shaped incision using a gel port and a high-density monopolar saline-cooled radiofrequency device is useful for the accomplishment of SILS for a huge liver cyst.

The purpose of this study was to evaluate the value of preoperative screening colonoscopies in patients with biliary tract cancer. A total of 544 patients with biliary tract cancer who underwent preoperative screening colonoscopies between January 2005 and December 2012 were retrospectively analyzed. Synchronous colorectal neoplasia was detected in 199 patients (36.7 %), with adenocarcinomas detected in 21 (3.9 %) patients, carcinoids in two (0.4 %) patients, and adenomas in 176 (32.4 %) patients. Of those with adenomas, 32 patients were diagnosed with advanced adenomas, defined as adenomas with a maximum diameter of >1 cm, villous histology, or high-grade dysplasia because these characteristics implied the risk of malignant transformation. Fifty-five (10.1 %) of the patients with colorectal neoplasia required resection (11 surgical and 44 endoscopic resections). There were no major adverse events related to the resection. Univariate and multivariate analyses revealed that smoking status [ex-smoker + current smoker vs. non-smoker: odds ratio (OR) 2.32; 95 % confidence interval (CI) 1.30−4.21] and advanced age (≥70 vs. ≤69 years: OR 2.22; 95 % CI 1.24−3.91) were independent risk factors of having a colorectal neoplasia that required resection. In patients with biliary tract cancer, preoperative screening colonoscopy was feasible and provided valuable clinical information. Synchronous colorectal neoplasia was detected in a substantial number of patients. Preoperative screening colonoscopies should be considered especially in high-risk patients such as smokers and elderly patients.

ObjectivesThis randomized clinical trial was designed to investigate whether inchinkoto has a hepatoprotective effect on postoperative outcome after major hepatectomy.MethodsSixty-one patients scheduled for major hepatectomy were randomly assigned to one of two groups in which preoperative inchinkoto was (inchinkoto group, n = 30) or was not (non-inchinkoto group, n = 31) administered. Inchinkoto was administered for at least 7 days before surgery. The primary endpoint was the incidence of post-hepatectomy liver damage. The expression of nuclear factor E2-related factor 2 (Nrf2) and other oxygen stress-related markers in the liver were also determined.ResultsThere was no significant difference in clinical characteristics between the inchinkoto and non-inchinkoto groups. Serum levels in liver function tests and incidences of post-hepatectomy liver failure did not differ significantly between the two groups. However, there was a significantly higher induction of antioxidant factors in the liver, such as Nrf2 protein and heme oxygenase-1 mRNA, after hepatectomy in the inchinkoto group than in the non-inchinkoto group.ConclusionsThe preoperative administration of inchinkoto did not have a significant impact on the overall outcome of major hepatectomy. However, inchinkoto induced the expression of Nrf2 during hepatectomy and may have exerted an antioxidative effect on the liver.

Generalization of laparoscopic pelvic surgery has brought about profound knowledge of the pelvic anatomy and has encouraged expansion of indications for laparoscopic surgery to extended pelvic surgery. Pelvic exenteration (PE) is still a demanding surgical procedure and remains an essential technique for pelvic surgery although minimally invasive and function-preserving surgery is in the mainstream of surgical treatment. However, the techniques of laparoscopic PE (LPE) have been rarely explained nor has its feasibility been fully evaluated. The aim of this study was to describe important technical points and to assess the feasibility of LPE for pelvic malignancies. Data on 67 patients with pelvic malignancies, who underwent PE between June 2006 and August 2014, were analyzed retrospectively. LPE has been indicated since 2013. Patients were divided into the LPE group (n = 9) and the conventional open PE (OPE) group (n = 58). Operative time in the LPE and OPE groups was similar (935 vs. 883 min, p = 0.398). Intraoperative blood loss in the LPE group was significantly less than that in the OPE group (830 vs. 2769 ml, p = 0.003). Pathological R0 resection rate was similar in both groups (77.8 vs. 75.9 %). Overall incidence of any complication and major complications were much lower in the LPE group (66.7 and 0 %) compared to the OPE group (89.7 and 32.8 %), although not statistically significant (p = 0.094 and 0.053, respectively). Postoperative hospital stay was significantly shorter in the LPE group than in the OPE group (27 vs. 43 days, p = 0.003). We confirmed that LPE for pelvic malignancies resulted in less blood loss, a lower complication rate, and shorter postoperative hospital stay compared to OPE. LPE performed by an experienced pelvic surgeon was safe and efficient, and might be a promising option for carefully selected patients.

Background The occurrence of bacterial translocation (BT) to the mesenteric lymph nodes following the Pringle maneuver is well established; however, the incidence of BT to the portal circulation remains unclear. Methods Portal blood of patients with suspected hilar malignancy who underwent major hepatobiliary resection with cholangiojejunostomy was sampled three times during surgery: immediately after laparotomy (PV-1); before liver transection and after skeletonization of the hepatoduodenal ligament (PV-2); and after completion of the liver transection (PV-3). The samples were analyzed for microbes with a bacterium-specific ribosomal RNA-targeted reverse transcription-polymerase chain reaction method. Results Fifty patients were enrolled in the study, with a mean total Pringle time of 86 min. Microbes in the portal blood were detected in 11 (22%) of the 50 patients. The occurrence of microbes was not different among the PV-1 samples (8% = 4/50), PV-2 samples (14% = 7/50), and PV-3 samples (14% = 7/50) (P = 0.567). Obligate anaerobes were predominantly detected. The positivity of the PV-3 samples showed no correlation with the total Pringle time or with the occurrence of postoperative infectious complications. The total Pringle time did not affect the surgical outcomes, including infectious complications, liver failure, or mortality. The concentrations of aspartate aminotransferase and alanine aminotransferase on postoperative day 1 significantly correlated with the total Pringle time. Conclusions The intermittent Pringle maneuver is unlikely to induce BT to the portal circulation and is safe, even in difficult, complicated hepatobiliary resections requiring long clamping times.

Major hepatectomy with extrahepatic bile duct resection for perihilar cholangiocarcinoma remains a highly morbid procedure. The association between preoperative sarcopenia and postoperative morbidity/mortality has been reported for various types of surgeries. The objective was to analyze the relationship between preoperative sarcopenia and postoperative morbidity/mortality in patients who underwent major hepatectomy with extrahepatic bile duct resection. This study included 256 patients who underwent major hepatectomy with extrahepatic bile duct resection from 2008 to 2014. Preoperative sarcopenia was assessed by a measurement of the total psoas muscle area (TPA). The measured TPA was normalized by height. Preoperative sarcopenia was defined as the presence of a normalized TPA in the lowest sex-specific tertile. A total of 54 males and 31 females were determined to have preoperative sarcopenia. The length of the postoperative hospital stay for patients with sarcopenia was significantly longer than for those without sarcopenia (39 vs 30 days, p

Although an aggressive surgical approach to perihilar cholangiocarcinoma (PHC) has improved survival, a prognosis of advanced PHC remains unsatisfactory. The overexpression of mesenchymal–epithelial transition factor (MET) and recepteur d’origine nantais (RON) has been shown to be associated with poor prognosis in some types of cancer. One hundred sixty-nine patients who underwent histologically curative resection for PHC were subjected to immunohistochemical analysis for MET and RON. The association between a positive expression of MET or RON and clinicopathologic features as well as the patients’ prognosis were analyzed. There were 27 patients (16 %) who had a positive expression for both MET and RON. Although clinicopathologic features in the either MET- or RON-negative group were not significantly different compared to the both MET- and RON-positive group, the prognosis tended to be worse in the patients with both MET and RON positivity. When the analysis was limited to patients with advanced-stage disease (stage III and IVa), a multivariate analysis revealed that both MET and RON positivity and lymph node metastasis were identified as independent poor prognostic factors. The overall survival rate for patients with both MET and RON positivity was worse than that with either MET or RON negativity in patients with advanced PHC. The poor prognosis in these patients was not associated with unfavorable clinicopathologic features. The examination of MET and RON expression in PHC may enable a tailored method for patient classification that could not otherwise be achieved using the conventional pathologic classification system.

The clinicopathologic significance of mucin production in patients with papillary cholangiocarcinoma (PCC) is still controversial. We aimed at clarifying the similarities and differences between PCC cases with and without mucin secretion with regard to biological behavior and clinical course. Among 644 patients with surgically resected cholangiocarcinoma (1998–2011), 184 (28 %) patients were considered to have PCC and were enrolled in the study. Those patients were divided into two groups based on whether their PCC was mucin-producing (PCC-M, n = 89) or not (PCC-NM, n = 95). The presence of mucin secretion was determined by the cut surface of the specimens and by pathologic examination. The clinicopathological features of PCC-M and PCC-NM largely overlapped. No significant between-group differences in malignant potential characteristics, including the depth of invasion, pathological T classification, and regional/periaortic lymph node metastasis, were observed (P = 0.193, 0.181, 0.083, and 0.674, respectively). However, a few clinicopathological differences existed between the two PCC types, i.e., the predominant histological type and epithelial subtype (P

Objective: To report a single-incision laparoscopic cholecystectomy (SILC) for a patient with cholecystitis that required endoscopic nasogallbladder drainage (ENGBD). Clinical Presentation and Intervention: A 75-year-old man was diagnosed with moderate acute cholecystitis and underwent antiplatelet therapy for a history of brain infarction. An ENGBD was performed as an initial treatment for his cholecystitis. After recovery from the cholecystitis, a SILC was performed using a SILS Port with an additional forceps. Because neither Rouviere's sulcus nor Calot's triangle could be identified with a favorable laparoscopic view, the fundus-first procedure was selected. The patient's postoperative course was uneventful, and he was discharged from the hospital on day 3 after surgery. Conclusion: In this case of a patient who had cholecystitis that required ENGBD, a SILC was successful performed using a combination of SILS Port with additional forceps and fundus-first procedure.

Background The choleretic effects of Inchinkoto (ICKT), a hepatoprotective herbal medicine, are variable among patients. This study sought to investigate the correlation between the choleretic effects of ICKT and the intestinal microenvironment in patients with biliary obstruction. Methods Patients with biliary obstruction who underwent external biliary drainage were enrolled. The concentrations of total bilirubin and bile acid in the bile, and genipin, a major active ingredient of ICKT, in the bloodstream before and after ICKT administration were measured. Feces were collected from the patients to determine bacterial count and organic acid concentrations. Results Samples from 37 patients were collected and analyzed. The serum concentration of genipin increased 3 hours after ICKT administration and showed a positive correlation with the percent changes of biliary concentrations of bile acid, total bilirubin, and direct bilirubin. Serum genipin concentration also showed a positive correlation with the fecal concentrations of representative obligate anaerobes such as the Clostridium leptum subgroup, Bacteroides fragilis group, Bifidobacterium, and the Atopobium cluster. In sharp contrast, so-called harmful bacteria such as Clostridium difficile, Enterobacteriaceae, and Enterococcus showed a negative correlation with the concentration of genipin. Genipin concentration after ICKT administration showed a positive correlation with the fecal concentration of short chain fatty acids such as propionic acid and butyric acid, and a negative correlation with the fecal concentration of lactic acid. Conclusions The absorption of genipin was variable among patients. This variability may be associated with the fecal microenvironment profile and partly explain the variable choleretic effects of ICKT among patients.

The tumor-suppressive role of trefoil factor family (TFF) members in gastric carcinogenesis has been suggested, but their significance and mechanisms in other digestive diseases remain elusive. To clarify the role of TFF1 in pancreatic carcinogenesis, we performed IHC on human samples, transfected siRNA against TFF1 into pancreatic cancer cell lines, and employed mouse models in which PanIN development and loss of TFF1 occur simultaneously. In human samples, the expression of TFF1 was specifically observed in pancreatic intraepithelial neoplasm (PanIN), but was frequently lost in the invasive component of pancreatic ductal adenocarcinoma (PDAC). When the expression of TFF1 was suppressed in vitro, pancreatic cancer cell lines showed enhanced invasive ability and features of epithelial-mesenchymal transition (EMT), including upregulated Snail expression. TFF1 expression was also observed in PanIN lesions of Pdx-1 Cre; LSL-KRASG12D (KC) mice, a model of pancreatic cancer, and loss of TFF1 in these mice resulted in the expansion of PanIN lesions, an EMT phenotype in PanIN cells, and an accumulation of cancer-associated fibroblasts (CAFs), eventually resulting in the development of invasive adenocarcinoma. This study indicates that the acquisition of TFF1 expression is an early event in pancreatic carcinogenesis and that TFF1 might act as a tumor suppressor to prevent EMT and the invasive transformation of PanIN.

This study investigated the impact of gastrojejunal anatomic position on the incidence of delayed gastric emptying after pancreatoduodenectomy.A total of 160 patients were included in the retrospective analysis. The relative anatomic position of the gastrojejunostomy was evaluated using coronal and sagittal plane computed tomography images on postoperative day 7; the coronal cardia anastomotic angle and the sagittal fundus anastomotic angle were measured. In the validation study, 64 consecutive patients were enrolled, and gastric emptying was evaluated using water-soluble contrast medium. The extent of gastric emptying was graded as grade I (no gastric dilatation and no stasis), grade II (gastric dilatation but no stasis), or grade III (gastric dilatation and stasis).Patients with grades B (n = 8) and C (n = 22) delayed gastric emptying were included in the delayed gastric emptying group (n = 30), and the others were included in the nondelayed gastric emptying group (n = 130). The coronal cardia anastomotic angle was not significantly different between the 2 groups, whereas the sagittal fundus anastomotic angle was significantly greater in the delayed gastric emptying group compared to the nondelayed gastric emptying group (median 50.3 vs 64.5 degrees, P .001). Multivariate analysis, including various risk factors of delayed gastric emptying, indicated that a sagittal fundus anastomotic angle60 degrees was the only independent risk factor of delayed gastric emptying (odds ratio, 16.59). In the validation study, the median degree of sagittal fundus anastomotic angle increased as the gastric emptying grade increased (grade I, 44.3 degrees; grade II, 55.3 degrees; grade III, 60.7 degrees; P = .014 by analysis of variance).The gastrojejunal anatomic position after pancreatoduodenectomy has a significant impact on the incidence of delayed gastric emptying. (Surgery 2017;160:XXX-XXX.).

Minor hepatectomy following liver partition between the right anterior and posterior sectors requires technical ingenuities. In such hepatectomy, we used three-dimensional (3D) print; therefore, our procedure was introduced.Digital segmentation of anatomical structures from multidetector-row computed tomography images utilized the original software 'PLUTO', which was developed by Graduate School of Information Science, Nagoya University. After changing the final segmentation data to the stereolithography files, 3D-printed liver at 70% scale was produced. The support material was washed and mould charge was removed from 3D-printed hepatic veins. The surface of 3D-printed model was abraded and coated with urethane resin paint. After natural drying, 3D-printed hepatic veins were coloured by injection of a dye. The 3D-printed portal veins were whitish because mould charge remained. All procedures after 3D printing were performed by hand work. A 3D-printed model of the right posterior sector and a 3D-printed model of other parenchyma were produced, respectively. Measuring the length between the main structures on the liver surface and the planned partition line on the 3D-printed model, land mark between the right anterior and posterior sectors on the real liver surface was produced with scale adjustment.Minor hepatectomy following liver partition between the right anterior and posterior sectors was performed referring to 3D-printed model. The planned liver partition and resection were successful.Application of 3D-printed liver to minor hepatectomy following liver partition between the right anterior and posterior sectors is easy and a suitable procedure.

Objectives Trefoil Factor Family protein 1 (TFF1) is secreted from mucus-producing cells. The relationship between TFF1 expression and clinical outcome in pancreatic ductal adenocarcinoma (PDAC) remains unknown. We aimed to evaluate the prognostic significance of TFF1 expression in PDAC. Methods TFF1 expression was examined on paraffin-embedded sections from 91 patients with resected PDAC using immunohistochemistry. The relationships between TFF1 expression and clinicopathological features were analyzed. Results Among 91 PDAC patients, 71 patients (79.7%) showed TFF1 expression in cancer cells. In a subgroup of 71 patients, TFF1 expression was predominantly observed in the central part of the tumor, whereas TFF1 expression in the invasion front was reduced in 33 patients (46.4%). A significant correlation between preserved TFF1 expression in the invasion front and lymph node metastasis was observed. Univariate survival analysis revealed that preserved TFF1 expression in the invasion front, positive lymphatic invasion, lymph node metastasis and R1 resection was a significant poor prognostic factor in TFF1-positive PDAC patients. Conclusions TFF1 expression is frequently lost or decreased in the invasion front of human PDAC, and preserved TFF1 expression in the invasion front might predict poor survival in patients with PDAC.

Background Mucin-producing cholangiocarcinoma (MPCC) is an uncommon tumour that is clinically characterized by mucin-hypersecretion. Because the initial symptoms of MPCC may be attributed to the viscus mucobilia, the primary tumour mass may potentially be unrecognizable. We report an interesting case of curatively resected occult MPCC in situ. Case presentation A 70-year-old man was referred to our hospital with increased levels of biliary enzymes. Multidetector row computed tomography (MDCT) demonstrated a diffuse dilatation of the entire biliary system without evidence of tumour mass. Additionally, there were numerous variably sized cysts throughout the liver. The cyst of S4 was the largest, followed by that of S1, which connected with the right hepatic duct. Endoscopic retrograde cholangiography showed intrabiliary mucus, predominantly in the left hepatic duct, but failed to show a communication of both cysts with the bile duct. We clinically suspected that minute MPCC was present within the S1 cyst and performed left hepatectomy, caudate lobectomy, and resection of the extrahepatic bile duct. Macroscopically, papillary adenocarcinoma in situ was present in the S1 cyst, and a final diagnosis of MPCC originating from the bile duct of the caudate lobe was made. Conclusions For MPCC, in practice, we should consider the possibility that this tumour can be occult. In this complicated setting, demonstrating the communication to the responsible dilated duct is a clue to the diagnosis. Multidirectional MDCT images succeeded in specifically demonstrating this communication, which is insensitive to the presence of excessive mucobilia.

The presence of cholangitis has been shown to impair liver regeneration capacity after major hepatectomy in a rat cholangitis model. It is unclear, however, whether the presence of cholangitis has any impact on liver generation in clinical settings.To determine the effects of preoperative cholangitis on hepatic regeneration rates after preoperative portal vein embolizations (PVEs) and postoperative courses after major hepatectomies in humans.From 1991 to 2012, 450 patients underwent preoperative PVEs and subsequent major hepatectomies. Among them, 72 patients (16.0%) had preoperative cholangitis. The volume change of the nonembolized lobe after PVE and the postoperative outcomes after a major hepatectomy were compared between cholangitis and noncholangitis groups.The average volume increase in the nonembolized lobe after PVE was almost identical in both the cholangitis (10.0%) and noncholangitis (9.5%) groups. The average term required to acquire institutional safety criteria, however, was longer in the cholangitis group (24.3 days) compared with the noncholangitis group (18.3 days) (P.001). The postoperative maximum serum total bilirubin levels (5.7 mg/dL vs 8.1 mg/dL, P = .035), morbidity rate (56% vs 78%, P = .001), and postoperative hospital stay (44 days vs 53 days, P = .021) were all greater in the cholangitis group compared with the noncholangitis group. With multivariate logistic regression analyses, the presence of preoperative cholangitis was identified as one of the independent risk factors for postoperative morbidity.These results indicate that patients with preoperative cholangitis should be carefully managed during their perioperative periods of PVE and after major hepatectomies.

Left hepatic trisectionectomy is a challenging procedure. For an anatomically correct resection, it is necessary to have understanding of the right intersectional plane; however, little is known on this issue. The purpose of this study was to investigate the 3D anatomy of the right intersectional plane and to enable safe and precise left trisectionectomy. A virtual left trisectionectomy was performed using 3D-processing software, in patients who underwent computed tomography. The reconstructed images were reviewed, and attention was paid to the extent of the right hepatic vein (RHV) exposure on the transected plane and the type of the inferior right hepatic vein (IRHV). Of the 200 study patients, 109 (54.5 %) patients showed complete exposure of the RHV on the transected plane, whereas the remaining 91 exhibited partial exposure. In the 109 patients with complete exposure, 58 (53.2 %) patients had no IRHV and the remaining 51 had a small IRHV. None of the patients had a large IRHV. In contrast, of the 91 patients with partial exposure, only 10 (11.0 %) patients had no IRHV, 35 (38.5 %) had a small IRHV, and 46 (50.5 %) patients had a large IRHV. The incidence of IRHV types was significantly different between the two groups (P

Hepatectomy for an invisible small tumor using intraoperative ultrasonography requires technical ingenuity. We used a 3D print of a liver to perform a hepatectomy on two patients with synchronous multiple liver metastases from colorectal cancer. Because of preoperative chemotherapy, one of the tumors became smaller and invisible to ultrasonography in each case. We present our procedure here. Multidetector-row computed tomography images of anatomical structures were digitally segmented using the original software “PLUTO,” which was developed at the Graduate School of Information Science, Nagoya University. After converting the final segmentation data to stereolithography files, a 3D printed liver at a 70 % scale was produced. The support material was washed and the mold charge was removed from the 3D-printed hepatic veins. The surface of the 3D-printed model was abraded and coated with urethane resin paint. After air-drying, the 3D-printed hepatic veins were colored by injecting a dye. The 3D printed portal veins were whitish because mold charge remained. All procedures after 3D printing were performed by hand. Hepatectomy for the small tumor that is invisible to intraoperative ultrasonography was performed by referring to a 3D-printed model. The planned resections were successful with histologically negative surgical margins. The application of a 3D-printed liver to perform a hepatectomy for a small tumor that is invisible to intraoperative ultrasonography is an easy and feasible procedure. Use of 3D-printing technology in hepatectomy requires further improvement and automation of hand work after the 3D print has been made.

The American Joint Committee on Cancer (AJCC) has recommended that cancers with liver involvement be graded T2b and those with portal vein involvement be graded T3, although the value of staging as prognostic factors remains unclear. We evaluated the current definition of the T2/3 tumors for perihilar cholangiocarcinoma. A total of 202 patients with perihilar cholangiocarcinoma who underwent hepatectomy without vascular resection were enrolled. Clinicopathologic data about invasion of the liver and the unilateral portal vein were evaluated. The liver and the unilateral portal vein were involved in 100 (49.5 %) and 38 (18.8 %) patients, respectively. The survival rates were not significantly different between patients with and without liver invasion (48.6 vs. 52.2 %, respectively, at 5 years, P = 0.157) and between patients with or without unilateral portal vein invasion (43.2 vs. 52.1 %, respectively, at 5 years, P = 0.363). The survival rate of patients with tumors staged pT2b was not significantly different from the rate of patients with pT2a (63.4 vs. 55.6 % at 5 years, P = 0.912), and the pT2b tumor patient survival rate was better than the rate of patients with pT3 (34.9 % at 5 years, P = 0.011). Using multivariate analysis, nodal metastasis (P = 0.003), positive surgical margin (P = 0.010), and Bismuth type IV tumor (P = 0.039) were identified as independent prognostic factors. The liver and the unilateral portal vein are frequently involved in perihilar cholangiocarcinoma. The determinants of the current AJCC T2/3 tumor classifications are rational; however, subdivision of T2 tumors may be of less clinical value.

Background Previously, we reported on the clinical efficacy and safety of portal vein embolization (PVE) with fibrin glue. Our embolic materials for PVE changed from fibrin glue to absolute ethanol (EOH) after 2001 due to prohibition of using fibrin glue for PVE. With introducing our technique of PVE with EOH, we evaluated its safety and efficacy with attention to the amount of EOH. Methods The medical records of 154 patients who underwent PVE using EOH were retrospectively reviewed. Results Changes with time in both the serum levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) after PVE returned to the initial condition within 7 days after PVE. In the 96 patients who underwent CT volumerty 14 to 21 days after PVE, the volume of the embolized lobe decreased from 701 ± 165 cm3 to 549 ± 148 cm3 (P < 0.0001). Meanwhile, the volume of the non-embolized lobe increased from 388 ± 105 cm3 to 481 ± 113 cm3 (P < 0.0001). On simple linear regression, the amount of EOH was positively correlated with both the maximum of AST and that of ALT after PVE; however, it never correlated with changes in liver volume after PVE. Conclusions Portal vein embolization with EOH has a substantial effect on both hypertrophy of the non-embolized lobe and atrophy of the embolized lobe. Quick recoveries of changes with time in AST and ALT after PVE proved that PVE with EOH is a safe procedure. The amount of EOH affected the extent of liver damage but had no clinical effects on changes in liver volume after PVE.

Intraductal papillary neoplasm of the bile duct (IPNB) is a presumed precursor lesion in biliary carcinogenesis, clinicopathologically overlapping with papillary cholangiocarcinomas (PCC); however, because IPNB has no standardized definition, this relationship remains equivocal. Herein, we aimed to develop a new prognostic model for PCC by focusing on the invasive proportion.Among 644 patients with resected cholangiocarcinoma (1998-2011), 184 (28%) had intraductal, exophytic, papillary lesions. These were divided into 4 subsets based on the invasive component: Noninvasive (PCC-1; n = 14), ≤10% (PCC-2; n = 32), 11-50% (PCC-3; n = 60), and50% (PCC-4; n = 78). The remaining 460 were identified as non-PCCs (NPCC).Invasion beyond the duct wall and regional lymph node metastases were more frequent in NPCC than PCC (P.001 for both). Five-year survival was better for PCC (55%) than NPCC (35%; P.001), indicating the papillary component to be a significant, independent prognosticator. PCC-4 and NPCC had similar clinicopathologic features and overlapping survival curves: 33% and 35% at 5 years (P = .835), both less than those of PCC-1, PCC-2, and PCC-3 (respectively, 92%, 74%, and 64% at 5 years; P.005 in all combinations). Multivariate analysis in PCC showed50% invasive component, nodal metastasis, and a positive operative margin as independent predictors.PCC survival decreased with progression of the invasive component. PCC with50% invasive component was clinicopathologically similar to NPCC. Although IPNB might be nosologically applied only for PCC cases with ≤50% invasive component, the present prognostic delineation suggests that all PCC subgroups belonged to a singular disease group.

Cholangiocarcinomas exhibit various modes of local extension, and some tumors can only be completely resected by hepatopancreatoduodenectomy (HPD), which is defined as the resection of the whole extrahepatic biliary system with the adjacent liver and pancreatoduodenum. Since Takasaki et al. introduced HPD for locally advanced gallbladder cancer in 1980, Japanese hepatobiliary surgeons have aggressively challenged this extended procedure for advanced biliary tumors. Early experiences with HPD were frequently associated with liver failure and sequential mortality, leading to an underestimation of the survival benefit of HPD. However, with improvements in surgical techniques and perioperative patient care, including portal vein embolization, over the last two decades, the mortality rate after HPD has gradually decreased. Recent studies have demonstrated a favorable survival in cholangiocarcinoma, provided that R0 resection is achieved. In contrast, HPD for gallbladder cancer remains controversial because of the extremely poor survival, although the study populations have been limited. HPD can be performed with low mortality and offers a better probability of long-term survival in patients with cholangiocarcinoma. We should consider HPD to be a standard approach for laterally advanced cholangiocarcinomas that are otherwise unresectable.

Background Right-sided hepatectomy is often selected for perihilar cholangiocarcinoma because the extrahepatic portion of the left hepatic duct is longer than that of the right hepatic duct. However, the length of resected left hepatic duct in right-sided hepatectomy has not been reported. Methods Patients who underwent right-sided hepatectomy for perihilar cholangiocarcinoma were reviewed retrospectively. Trisectionectomies were performed according to a previously reported technique of anatomical right hepatic trisectionectomy. Right hepatectomy was performed according to standard operative procedures. The length of resected left hepatic duct was measured. Results Thirty-three patients underwent right trisectionectomy and 141 had a right hemihepatectomy. Patients having a trisectionectomy had more advanced tumours and so required combined portal vein resection more frequently. Duration of surgery and blood loss were similar in the two groups. Morbidity and mortality rates tended to be higher following hemihepatectomy than after trisectionectomy. The mean(s.d.) length of resected left hepatic duct was significantly greater in trisectionectomy than in hemihepatectomy (25·0(6·9) versus 14·8(5·3) mm; P < 0·001). In patients with Bismuth type IV tumours, the percentage of negative left hepatic duct margins was significantly higher for trisectionectomy than for hemihepatectomy (89 versus 57 per cent; P = 0·021). Achievement of R0 resection was similar and survival did not differ between the two groups, despite different tumour load. Conclusion Compared with right hemihepatectomy, anatomical right hepatic trisectionectomy provides a greater length of resected hepatic duct, leading to a high proportion of negative proximal ductal margins even in patients with Bismuth type IV tumours.

Background Bismuth type IV perihilar cholangiocarcinoma has traditionally been categorized as unresectable disease. The aim of this study was to review experience with a resection-based strategy in patients who have type IV perihilar cholangiocarcinoma. Methods Medical records of consecutive patients with a diagnosis of type IV perihilar cholangiocarcinoma between 2006 and 2015 were reviewed retrospectively. Primary outcomes assessed were surgical results and long-term survival. Results Of the 332 patients with type IV tumour, 216 (65·1 per cent) underwent resection. Left hepatic trisectionectomy was the most common procedure (112 patients). Combined vascular resection was performed in 131 patients. Median duration of operation was 607 (range 356–1045) min, and blood loss was 1357 (209–10 349) ml. Complications of Clavien–Dindo grade III or more developed in 90 patients (41·7 per cent) and four (1·9 per cent) died from complications within 90 days. Survival rates were better for the 216 patients whose tumours were resected than for the 116 patients with unresected tumours (32·8 versus 1·5 per cent at 5 years; P < 0·001). Patients with pN0 M0 disease after resection had a favourable 5-year survival rate of 53 per cent. Percutaneous transhepatic biliary drainage, blood transfusion, lymph node metastasis and distant metastasis were identified as independent negative prognostic factors for survival. Conclusion Although resection for type IV tumour is technically demanding with high morbidity, it can be performed with low mortality and offers better survival probability in selected patients.

Pancreatic ductal adenocarcinoma (PDAC) is a lethal malignant disease associated with poor prognosis, despite recent medical advances. It is of great importance to understand the initial events and cells of origin of pancreatic cancer to prevent the development and progression of PDAC. There are three distinct precursor lesions that develop into PDAC: pancreatic intraepithelial neoplasms (PanINs), intraductal papillary mucinous neoplasms (IPMNs), and mucinous cystic neoplasms (MCNs). Studies on genetically engineered mouse models have revealed that the initiation and development of these lesions largely depend on genetic alterations. These lesions originate from different populations in the pancreas. PanIN development seems to be the result of the transdifferentiation of acinar cells, whereas IPMNs most likely arise from the progenitor niche of the pancreatic ductal epithelium. Pancreatic carcinogenesis is dependent on various events, including gene alterations, environmental insults, and cell types. However, further studies are needed to fully understand the initial processes of pancreatic cancer.

Background The International Union Against Cancer (UICC) staging system for perihilar cholangiocarcinoma changed in 2009. The aim of this study was to validate and optimize the UICC system for these tumours. Methods This retrospective study was conducted in eight Japanese hospitals between 2001 and 2010. Perihilar cholangiocarcinoma was defined as a cholangiocarcinoma that involves the hilar bile duct, independent of the presence or absence of a liver mass component. The stratification ability of the UICC tumour node metastasis (TNM) system was compared with that of a modified system. Results Of 1352 patients, 35·9, 44·8 and 12·6 per cent had Bismuth type IV tumours, nodal metastasis (N1) and distant metastasis (M1) respectively. T4 tumours (43·2 per cent) and stage IVA (T4 Nany M0; 36·3 per cent) disease were most common. Survival was not significantly different between patients with T3versus T4 tumours (P = 0·284). Survival for patients with stage IVA disease was comparable to that for patients with stage IIIB tumours (T1–3 N1 M0) (P = 0·426). Vascular invasion, pancreatic invasion, positive margin, N1 and M1 status were identified as independent predictors of survival. When Bismuth type IV tumours were removed from the T4 determinants and N1 tumours grouped together, the modified grouping had a higher linear trend χ2 and likelihood ratio χ2 compared with the original system (245·6 versus 170·3 respectively and 255·8 versus 209·3 respectively). Conclusion The present data suggest that minimal modification with removal of Bismuth type IV tumours from the T4 determinants and bundling of N1 disease may enhance the prognostic ability of the UICC system. However, this requires validation on an independent data set.

We report a case of successful right hepatectomy plus pancreatoduodenectomy (Rt-HPD) with arterial reconstruction for extrahepatic bile duct carcinoma with obstruction of the celiac axis in a 76-year-old man. Obstruction of the celiac axis resulted in arterial blood supply to the upper abdominal organs coming from the pancreatic head arcade. The patient underwent arterial reconstruction before the Rt-HPD to maintain the blood supply from the pancreatic head arcade for as long as possible. His postoperative course was uneventful and he was well with no sign of recurrence when last seen, 64 months after surgery. To our knowledge, this is the first description of this modified HPD with arterial reconstruction. Thus, rational surgical planning based on careful preoperative assessment would expand the indications for HPD, even for patients with celiac axis obstruction requiring arterial reconstruction.

Attempts have been made to use bevacizumab (BEV) in an adjuvant or neoadjuvant setting. However, BEV is known to cause various adverse events, and the safety of neoadjuvant BEV has not yet been fully evaluated. This study assessed the postoperative complications in patients receiving neoadjuvant BEV for colorectal cancer.The data for 78 patients with resectable advanced or metastatic colorectal cancer who received neoadjuvant BEV followed by surgical resection were retrospectively analyzed.The median interval between the last BEV dose and surgery was 9 weeks. The most common postoperative complication was pelvic sepsis, which occurred in 11 patients (14 %). A biliary fistula developed in four of 23 patients who underwent liver resection. Anastomotic leakage occurred in six of 24 patients with a colorectal anastomosis, four of whom required re-laparotomy. In a univariate analysis, male gender and a greater intraoperative blood loss were associated with postoperative complications of any grade. Colorectal anastomosis was a risk factor for major complications. In a multivariate analysis, intraoperative blood loss was an independent risk factor for postoperative complications of any grade (HR 6.338; P = 0.003). With regard to major postoperative complications, colorectal primary anastomosis was the only independent predictive risk factor (HR 8.285; P = 0.013).In patients with colorectal cancer who underwent elective surgery after BEV treatment, the interval between BEV and surgery was not a risk factor for postoperative complications (based on a median interval of 9 weeks). Colorectal primary anastomosis was the only independent risk factor for major postoperative complications.

To review our 34-year experience with 574 consecutive resections for perihilar cholangiocarcinoma and to evaluate the progress made in surgical treatment of this disease.Few studies have reported improved surgical outcomes for perihilar cholangiocarcinoma; therefore, it is still unclear whether surgical treatment of this intractable disease has progressed.Between April 1977 and December 2010, a total of 754 consecutive patients with perihilar cholangiocarcinoma were treated, of whom 574 (76.1%) underwent resection. The medical records of these resected patients were retrospectively reviewed.The incidence of major hepatectomies has increased, and limited resections, including central hepatectomies and bile duct resections, were rarely performed. Combined vascular resection was being used more often. Operative time has become shorter, and intraoperative blood loss has also decreased significantly. Because of refinements in surgical techniques and perioperative management, morbidity decreased significantly but was still high, with a rate of 43.1% in the last 5 years. Mortality rate has also decreased significantly (P0.001) from 11.1% (8/72) before 1990 to 1.4% (3/218) in the last 5 years. The ratio of advanced disease defined as pStage IVA and IVB has increased significantly from 49.4% before 2000 to 61.4% after 2001. The disease-specific survival for the 574 study patients (including all deaths) was 44.3% at year 3, 32.5% at year 5, and 19.9% at year 10. The survival was significantly better in the later period of 2001 to 2010 than in the earlier period of 1977 to 2000 (38.1% vs 23.1% at year 5, P0.001). For pM0, R0, and pN0 patients (n = 243), the survival in the later period was good with 67.1% at year 5, which was significantly better than that of the earlier period (P0.001). For pM0, R0, and pN1 patients (n = 142), however, the survival in the later period was similar to that of the earlier period (22.1% vs 14.6% at year 5, P = 0.647). Multivariate analysis revealed that lymph node metastasis was the strongest prognostic indicator.Surgical treatment of perihilar cholangiocarcinoma has been evolving steadily, with expanded surgical indication, decreased mortality, and increased survival. Survival for R0 and pN0 patients was satisfactory, whereas survival for pN1 patients was still poor, suggesting that establishment of effective adjuvant chemotherapy is needed.

Resection of cholangiocarcinoma often results in a positive ductal margin, from carcinoma in situ (CIS) near the main tumor; however, the biological behavior of the residual CIS after surgical resection remains equivocal. We report a case of late local recurrence of CIS, defined as long-term tumor progression from CIS residue at the ductal stump. The patient, a 73-year-old man, had undergone bile duct resection for distal cholangiocarcinoma, leaving positive ductal margins with CIS. A biliary stricture was found 10 years later at the site of anastomosis, and right hepatectomy with pancreatoduodenectomy was performed. Based on histological analogy and the evidence of remnant CIS, a final diagnosis of late local recurrence from the CIS foci was made. This uncommon mode of recurrence should be considered in patients with early-stage disease with expected favorable survival because salvage surgery is feasible for selected patients.

Background A salvage hepatectomy for an “inoperable” advanced perihilar tumor initially treated with a self-expanding metallic stent (SEMS) is challenging, and its safety and survival benefits remain unclear. The aim of this study was to report our experiences with this difficult resection. Methods This study involved 10 consecutive patients with suspected perihilar cholangiocarcinoma who underwent SEMS placement at a local hospital and were referred to our clinic for possible resection as their last option. Their medical records were analyzed retrospectively. Results Tumor extent was first reevaluated using multidetector-row computed tomography. Of the 10 patients, 4 were diagnosed as inoperable owing to locally advanced tumors ( n = 3) or poor physical condition ( n = 1). In the remaining 6 patients, after additional biliary drainage, a salvage hepatectomy was performed, including a right hepatectomy with a caudate lobectomy in 5 patients and a central bisectionectomy with a caudate lobectomy in 1. A combined portal vein resection was required in 3 patients, and a combined pancreatoduodenectomy was performed in 2 patients. R0 resection was achieved in 5 patients, and all patients tolerated the resection. Three patients died of recurrence, and the remaining 3 were alive without recurrence at the time of publication, 1 of whom has survived >10 years. Conclusion Pre-resection SEMS placement does not preclude a subsequent hepatectomy for patients with advanced perihilar tumors. Salvage hepatectomy, although technically demanding, is feasible and can revise the palliative scenario and benefit selected patients treated initially with an SEMS.

To analyze lymph node status in resected perihilar cholangiocarcinoma, to clarify which index (ie, location, number, or ratio of involved nodes) is better for staging, and to determine the minimum requirements for node examination.In the TNM classification for perihilar cholangiocarcinoma, the number or ratio of involved nodes is not considered for nodal staging. The minimum requirement for histologic examination of lymph nodes is arbitrary.This study involved 320 patients with perihilar cholangiocarcinoma who underwent resection from January 2000 to December 2009 at Nagoya University Hospital. The relationship between lymph node status and patient survival was retrospectively analyzed.Total lymph node counts (TLNCs), ie, the number of lymph nodes examined histologically, averaged 12.9 ± 8.3 (range: 1-59). Lymph node metastasis was found in 146 (45.6%) patients and was an independent, powerful prognostic factor. The survival rates were not significantly different between patients with regional node metastasis alone and those with distant node metastasis (19.2% vs 11.5% at 5 years, P = 0.058). The survival for patients with multiple node metastases was significantly worse than that for patients with single metastasis (12.1% vs 27.6% at 5 years, P = 0.002), regardless of the presence or absence of distant lymph node metastasis. The survival for patients with lymph node ratios (LNRs) of 0.2 or less was significantly better than that for patients with LNRs greater than 0.2 (21.4% vs 13.5% at 5 years, P = 0.032). Upon multivariate analysis of the 146 patients with lymph node metastasis, the number of involved nodes (single vs multiple) was identified as an independent prognostic factor (RR of 1.61, P = 0.045), whereas the locations (regional alone vs distant) and ratios (LNR ≤ 0.2 vs LNR0.2) of involved nodes were not. When the 148 pN0-R0 patients were divided into 3 groups (ie, those with TLNC ≥ 8, with TLNC = 5, 6, or 7, and with TLNC ≤ 4), survivals were identical between the first and second groups, whereas they were largely different between the former two and the third.Lymph node metastasis is a powerful, independent prognostic factor in perihilar cholangiocarcinoma and is better classified based not on location but on the number of involved nodes. To adequately assess nodal status, histologic examination of 5 or more nodes is recommended.