Modern work-up and extended resection in perihilar cholangiocarcinoma: the AMC experience

Modern work-up and extended resection in perihilar cholangiocarcinoma: the AMC experience Aim Perihilar cholangiocarcinoma (PHC) is a challenging disease and requires aggressive surgical treatment in order to achieve curation. The assessment and work-up of patients with presumed PHC is multidisciplinary, complex and requires extensive experience. The aim of this paper is to review current aspects of diagnosis, preoperative work-up and extended resection in patients with PHC from the perspective of our own institutional experience with this complex tumor. Methods We provided a review of applied modalities in the diagnosis and work-up of PHC according to current literature. All patients with presumed PHC in our center between 2000 and 2016 were identified and described. The types of resection, surgical techniques and outcomes were analyzed. Results and conclusion Upcoming diagnostic modalities such as Spyglass and combinations of serum biomarkers and molecular markers have potential to decrease the rate of misdiagnosis of benign, inflammatory disease. Assessment of liver function with hepatobiliary scintigraphy provides better information on the future remnant liver (FRL) than volume alone. The selective use of staging laparoscopy is advisable to avoid futile laparotomies. In patients requiring extended resection, selective preoperative biliary drainage is mandatory in cholangitis and when FRL is small (< 50%). Preoperative portal vein embolization (PVE) is used when FRL volume is less than 40% and optionally includes the left portal vein branches to segment 4. Associating liver partition and portal vein ligation for staged hepatectomy (ALPPS) as alternative to PVE is not recommended in PHC. N2 positive lymph nodes preclude long-term survival. The benefit of unconditional en bloc resection of the portal vein bifurcation is uncertain. Along these lines, an aggressive surgical approach encompassing extended liver resection including segment 1, regional lymph- adenectomy and conditional portal venous resection translates into favorable long-term survival. . . . . . . Keywords Perihilar cholangiocarcinoma Klatskin tumor Diagnosis Staging Biomarkers Preoperative assessment . . . Hepato-biliary scintigraphy Biliary drainage Surgical resection Postoperative outcome Introduction the epithelium of the biliary tract and may occur in the whole biliary ductal system. They are sub-classified ac- Cholangiocarcinoma accounts for 3% of all gastrointes- cording to their location, in intrahepatic, perihilar and tinal malignancies worldwide [1]. The tumors arise from distal cholangiocarcinoma [2]. Each entity comes with F. Rassam and E. Roos contributed equally to this work. * F. Rassam Department of Gastroenterology & Hepatology and Tytgat Institute f.rassam@amc.uva.nl for Liver and Intestinal Research, Academic Medical Center, Amsterdam, The Netherlands 1 Department of Medical Oncology, Academic Medical Center, Department of Surgery, Academic Medical Center, Amsterdam, The Netherlands Amsterdam, The Netherlands Department of Radiotherapy, Academic Medical Center, Amsterdam, The Netherlands Department of Radiology and Nuclear Medicine, Academic Medical Department of Pathology, Academic Medical Center, Center, Amsterdam, The Netherlands Amsterdam, The Netherlands 290 Langenbecks Arch Surg (2018) 403:289–307 a specific set of problems and therefore, management finally develops due to local biliary obstruction, patients are requires a tailored approach. often not resectable anymore, and thus not curable. Up to 65– Perihilar cholangiocarcinoma (PHC), also known as 80% of patients have initially unresectable disease due to ex- Klatskin tumor, is the most frequent biliary tract tumor and tensive hepatic artery and/or portal vein infiltration by tumor accounts for approximately 60% of all cholangiocarcinoma’s or distant metastases at time of presentation [10–13]. Of all [3]. This tumor originates in the extrahepatic biliary tract prox- patients who in time undergo a laparotomy, 40–70% ultimate- imal to the origin of the cystic duct, up until the second-degree ly have resectable disease [14–16]. Patients face many obsta- bile ducts. PHC can be subdivided according to proximal ex- cles during diagnosis and work-up for extended resection. tent of the tumor into the bile ducts (Bismuth-Corlette These problems range from confirmation of malignancy to classification) [4](Fig. 1). cholestasis and cholangitis due to biliary obstruction, requir- The incidence of cholangiocarcinoma varies widely be- ing biliary drainage. tween regions. In Asian populations and Chili, parasitic infec- Oncological outcomes depend heavily on the possibility of tions are strongly associated with PHC, showing a peak inci- performing a radical resection. Patients with unresectable dis- dence in Thailand of 87 per 100,000 [3, 5–7]. In Western ease, receiving palliative chemotherapy with gemcitabine and populations the incidence is considerably lower, 1–2per cisplatin, have an overall median survival of approximately 100,000, and PHC is mainly associated with primary scleros- 12 months [17, 18]. In contrast, median survival of patients ing cholangitis (PSC) [8, 9]. with an R0 resection is 30–46 months and 5-year survival Early symptoms are not specific and patients typically pres- rates range from 25 to 40% [19, 20]. The aggressive surgical ent with the sequelae of biliary obstruction. When jaundice approach necessary to achieve an R0 resection however, is Fig. 1 Bismuth-Corlette classification for staging of perihilar cholangiocarcinoma Langenbecks Arch Surg (2018) 403:289–307 291 associated with significant postoperative morbidity and mor- metastases (n = 68) or unfitness for major resection (n = tality with reported morbidity rates ranging from 60 to 70% 53) (Table 1). [21] and mortality rates as high as 5–18% [19, 22–24]. It is The remaining 321 (53.0%) patients underwent laparoto- therefore crucial to optimize patients before exposing them to my; 120 (19.8%) patients were deemed unresectable on the this high-risk surgery. basis of intraoperative findings. The main reasons were locally The aim of this review is to elaborate current diagnosis and advanced disease (n = 43), N2 lymph node metastases, (n = work-up and to review the issues of extended resection in 39), liver metastases (n = 11), peritoneal or other distant me- patients presenting with a hilar lesion suspicious of PHC, from tastases (n = 26) or major liver resection precluded by comor- the perspective of the long-standing experience with this com- bidities (n =1) (Table 1). plex tumor in our referral center. A total of 201 patients underwent extrahepatic bile duct resection in the majority of cases combined with (extended) The AMC experience; the denominator of patients liver resection. Of these patients, 66 (32.8%) underwent a left referred with (suspected) PHC hemihepatectomy, 8 (4.0%) underwent an extended left hemihepatectomy, 31 (15.4%) patients underwent a right Between 2000 and 2016, a total of 606 patients with lesions hemihepatectomy, 51 (25.4%) underwent an extended right suspicious of PHC have been referred to our center. Patients hemihepatectomy, 8 (4.0%) patients underwent resection of were discussed in our HPB oncology multidisciplinary meet- 1 or 2 segments and the remainder of 37 (18.4%) patients ing, consisting of experienced hepatobiliary surgeons, dedi- underwent bile duct resection alone (Table 2). cated endoscopists, (interventional, abdominal and nuclear) Based on pathological examination of the resection speci- radiologists, radiotherapists, nurse practitioners, medical on- mens, 170 (84.6%) patients had PHC and 31 (15.4%) had cologists and pathologists. benign disease (either unspecified sclerosing cholangitis or A total of 285 (47.0%) patients were deemed IgG4-associated cholangitis) (Fig. 2). unresectable, of which 228 (37.6%) were found to be Severe complications (Clavien-Dindo grade 3 or higher) unresectable at initial presentation on the basis of imag- were observed in 93(46.3%) of resected patients. Of all pa- ing studies. The remaining 57 patients were staged with tients who underwent resection, 18 (9.0%) died within the first unresectable disease after diagnostic laparoscopy 90 days. (Fig. 2). The main reason for unresectability was locally The median survival after resection of confirmed malignan- advanced disease (n = 104), N2 lymph node metastases cy was 52.6 months. The 5-year survival after resection was (n = 29), liver metastases (n = 27), peritoneal or distant 44.3% (Fig. 3). Fig. 2 Flow diagram of patients referred to the AMC with suspicion on PHC (2000–2016) 292 Langenbecks Arch Surg (2018) 403:289–307 Table 1 Reasons for unresectability in patients referred with PHC Reason for unresectability Patients n (%) Initially unresectable 285 After imaging/laboratory assessment 228 After diagnostic laparoscopy 57 Locally advanced disease 104 (36.5%) LN metastases 29 (10.2%) Liver metastases 27 (9.5%) Peritoneal/distant metastases 68 (23.9%) Unfit for surgery 53 (18.6%) Missing 4 (1.4%) Unresectable during laparotomy 120 Locally advanced disease 43 (35.8%) LN metastases 39 (32.5%) Liver metastases 11 (9.2%) Fig. 3 Overall survival in 170 patients undergoing resection of pathology Peritoneal/distant metastases 26 (21.7%) proven PHC in the AMC. The 5-year survival rate after resection was Unfit for surgery 1 (0.8%) 44.3% diagnostic modality for years [31–33]. PHC, however, fre- Pitfalls of diagnosis quently shows a submucosal growth pattern resulting in a low sensitivity of brush cytology of 27–56% [31, 34, 35]. Differentiation between malignant and benign disease Various techniques have been investigated to increase sensi- tivity of cytological samples. These include fluorescence in In patients with a presumed PHC, it is highly desirable to obtain situ hybridization (FISH) which is reported to increase sensi- a definitive diagnosis (Fig. 4). Benign biliary tract strictures are tivity to 69–93% [36–38]. Mutation analysis has not been difficult to differentiate from malignant disease [25–27]. In re- used widely, but seems mainly to increase specificity [39]. cent years, IgG4-associated cholangitis (IAC) has been identi- The use of stiffer bristles or repeated brushings also has not fied as a disease entity that may mimic PHC, both clinically as increased the diagnostic yield of brush cytology [40–42]. on imaging studies. It belongs to the spectrum of IgG4-related Other endoscopic techniques have emerged as well. The use disease, a systemic disease which can affect many other organs of endobiliary forceps biopsy during ERCP resulted in a as well [28–30]. Of all resections for presumed PHC worldwide, higher detection rate ranging from 44 to 89% [43, 44]. The 8–22% of patients turned out to have a benign disease on mi- technique however is challenging, especially in more proxi- croscopical examination of the resection specimen [26]. mal lesions as it is difficult to navigate and position the for- ceps. Consequently, it has not found wide application [44–46]. Endoscopic techniques Endoscopic retrograde Cholangioscopy offers direct visualization of biliary stric- cholangiopancreatography (ERCP) combined with brush- tures and seems to improve the diagnostic yield of routine cytology for microscopical examination has been the standard cytology. Percutaneous cholangioscopy requires percutaneous biliary access and multiple dilatations to allow access of the Table 2 Types of resection undertaken in 201 patients with presumed cholangioscope. Single operator cholangioscopy (Spyglass, PHC Boston scientific, Natick, MA, US) is introduced through a Type of resection Patients n (%) duodenoscope and is used in combination with Spybite Biopsy Forceps [47, 48]. Using these techniques, the sensitiv- Total number of patients 201 ity in diagnosis of biliary strictures has increased to 74.7% Left hemihepatectomy 66 (32.8%) [49]. Cholangioscopy enables targeted biopsies increasing Right hemihepatectomy 8 (4.0%) sensitivity and specificity to detect PHC to 66 and 97%, re- Extended left hemihepatectomy 31 (15.4%) spectively, in a meta-analysis. Single operator cholangioscopy Extended right hemihepatectomy 51 (25.4%) seems a useful new step in centers experienced with ERCP Segmentectomy (≤ 3 Couinaud segments) 8 (4.0%) and brush cytology. Only local excision of hilar bile ducts 37 (18.4%) Alternatively, intraductal ultrasound (IDUS) enables de- Including portal vein resection 30/151 (19.9%), 50 missing tailed imaging of the bile ducts and periductal tissue. IDUS Langenbecks Arch Surg (2018) 403:289–307 293 Fig. 4 Flowchart showing work- Suspected PHC up and treatment of patients suspected of PHC Biomarkers Benign CT/MRI/MRCP IAC assessment Assessment of resectability (Table 3) and staging (Table 4) Bili > 30 μmol/L Potenally respectable Staging laparoscopy Biliary drainage Hepatobiliary scingraphy < 40% FRL CT-volumetry < 2.7%/min/m Preoperave RTx in drained paents PVE Laparotomy Unresectable Resectable Palliave drainage with metal stents Hepatectomy + bile duct resecon Follow up has been reported to improve diagnostic accuracy of ERCP Biomarkers are also needed to monitor patients with an from 58% to 90% [33, 50]. However, stents that are often increased risk of PHC such as in primary sclerosing required to drain obstructed bile ducts make the interpretation cholangitis [56]. The conventional serum markers CA19–9 of IDUS difficult. If this is the case, the use of endoscopic and CEA are frequently used in gastrointestinal malignancies. ultrasonography (EUS) in combination with fine needle aspi- However, the diagnostic value of CA19–9 is debated because ration (FNA) may be preferable [50–52]. These techniques of its variable sensitivity of 33–93% and specificity of 67– however, require specific expertise to reach their maximum 98%. Its use as a prognostic biomarker seems more valuable potential and their success rates must be partially attributed to [57–60]. Furthermore, CA19–9 may be elevated in benign the experience of their users. biliary disease and/or in the presence of cholestasis, impairing its use as a reliable biomarker especially in biliary tumors [61]. Serum markers The limited ability to reliably acquire tissue The same applies for CEA with a sensitivity of 33–84% and samples has resulted in an ongoing quest for serum bio- specificity of 50–88% [58] in pancreato-biliary malignancies. markers. The additional use of serum markers to distinguish IAC from PHC has been an area of extensive research. Serum Staging and resectability IgG4 levels (ULN = 1.4 mg/ml) have limited diagnostic value when only slightly increased, since up to 15% of patients with Criteria for the assessment of resectability PHC have elevated sIgG4 levels as well [53, 54]. Recently a new test has been developed measuring the IgG4/IgG RNA Initial imaging is crucial in establishing diagnosis and in de- ratio. This test distinguishes IAC accurately (94% sensitivity, termining whether a patient is a candidate for resection. The 99% specificity) from PHC and primary sclerosing cholangitis goal of curative resection is to achieve negative margins (R0) [55]. The value of this test awaits further clinical assessment. while preserving sufficient volume and function of remnant 294 Langenbecks Arch Surg (2018) 403:289–307 liver with adequate portal venous and hepatic arterial blood morphological growth patterns that can be recognized on imag- supply. Factors to consider to determine resectability are in- ing to enhance the diagnostic confidence, determine manage- cluded in Table 3 [62]. ment and to provide additional information on prognosis. Unresectability can result from either extensive local disease However, imaging can also lead to confusion due to overlapping (including vascular and nodal involvement), presence of distant appearances with other hepatobiliary diseases, including benign metastases or comorbidity of the patient. Local unresectability lesions. Important conditions to consider are other causes of can be due to involvement of the portal vein and hepatic artery biliary dilatation such as choledocholithiasis, PSC, IAC and on the side of the future remnant liver without the possibility of biliary dilatation due to centrally located colorectal metastases a vascular reconstruction, extensive bilateral proximal infiltra- [65]. Ultrasound is usually the initial test to evaluate patients tion of the tumor into secondary biliary radicles (segmental bile with suspected bile duct obstruction [66, 67], and may provide ducts) and/or massive extension of tumor into the liver paren- informationonthe levelofobstructioninthe biliarytree. chyma. Furthermore, extrahepatic metastases including distant lymph node metastases beyond the hepatoduodenal ligament Cross-sectional studies CT and MRI are commonly used in (N2 nodes) are associated with poor survival and in most cen- various combinations with cholangiographic studies, in the ters, are considered as unresectable as well. diagnosis and preoperative planning of PHC. CT with iv con- It should be emphasized that local resectability depends on trast offers the opportunity to assess full extension of the tu- biliary anatomy at the liver hilum. The hepatic duct confluence mor in detail and determine resectability [68, 69]. If PHC is is defined by the convergence of the right and left hepatic ducts, suspected, imaging is preferably performed before stenting for at which site many anatomic variations exist [63]. In 20% of biliary drainage, since the images will be obscured by the cases, the anterior and posterior sectorial branches of the right plastic or metal stent. In general, PHC may be recognized by ductal system drain directly into the main hepatic duct. This dilated bile ducts, lack of communication between the left and may give rise to confusion as when a hilar tumor involving the right first-order bile ducts, crowding of bile ducts, ductal wall right anterior and posterior sectorial branches in these cases is thickening and enhancement, and lobar atrophy. In some combined with segmental involvement on the left side, the cases, a solid (mass forming) or papillary mass (intraductal tumor is defined by the Bismuth-Corlette classification as type growth type) may be seen. IV, which in many textbooks is considered unresectable. A type The early arterial and late portal venous phases of a CT- IV tumor in this situation however does not preclude a radical scan aid to assess the relationship between tumor and resection using an extended left hepatectomy. The same holds (branches of) the hepatic artery and portal vein, which is im- true for a tumor extending into the right sectorial ducts involv- portant in determining resectability [70, 71]. Key elements for ing a low inserting segment 4 duct of the left biliary system. staging in imaging are defined in Table 4. According to a Although defined as Bismuth-Corlette type IV, this tumor can of course be radically resected using an extended right Table 4 Key elements for staging of PHC hemihepatectomy. Resectability depends on hilar biliary anat- Key elements necessary for staging PHC omy and it is therefore important that resectability is assessed Location of primary tumor by hepatobiliary surgeons with expertise in PHC [64]. Intra- or extrahepatic Proximal common hepatic duct Imaging Confluence of the left and right hepatic duct Left or right hepatic duct Imaging plays a decisive role in the diagnosis, staging and as- Intraductal growth type sessment of resectability. PHC manifests with various Local extension Segmental duct involvement (including Bismuth-Corlette classification) Table 3 Criteria for the assessment of resectability in PHC Mentioning biliary variant anatomy Criteria for the assessment of resectability Vascular involvement (portal vein and/or hepatic arteries, including vascular variations and presence of stenosis of celiac axis or mesenteric Presence of (extra) hepatic metastases artery) Presence of lymph node metastases confined to hepatoduodenal ligament Lymph nodes (N1) or lymph node metastases along the common hepatic artery Regional N1; cystic duct, common bile duct, proper hepatic artery and and/or celiac axis (N2) portal vein nodes Possibility of achieving free ductal margins on the side of the FRL Metastatic N2; common hepatic artery, periaortic, pericaval, superior Involvement of portal vein bifurcation mesenteric or celiac artery nodes Involvement of hepatic artery branches Distant metastasis Volume and function of FRL Noncontiguous liver, peritoneum, bone, other Langenbecks Arch Surg (2018) 403:289–307 295 meta-analysis by Ruys et al., sensitivity and specificity of CT is based on pathology assessment of the resection specimen were 89 and 92% for assessment of portal vein involvement and is mainly used postoperatively as a prognostic tool. The (encasement or occlusion are strong evidence), 84 and 93% Bismuth-Corlette classification system, introduced in 1975, is for hepatic artery involvement and 61 and 88% for lymph used to describe proximal involvement of tumor into the bile node metastases, respectively [68]. ducts [4]. This system is mainly informative to surgeons for MRI with iv contrast provides an acceptable alternative to planning of the type of resection, but does not determine re- CT in the evaluation of PHC. Both CT and MRI have similar sectability since other parameters such as distant metastases staging accuracy, including that of nodal staging [72]. The and vascular involvement are not included. The Blumgart advantage of MRI is that combined with cholangiography classification system takes in addition to bile duct involve- (MRC), it provides anatomical definition of the biliary tree. ment, portal vein involvement and lobar atrophy into account Whether CT or MRI is used should be based on local expertise as well [82]. However, since its introduction in 1998, the in- and accessibility to one of these modalities [73]. dications for (extended) resections have expanded rendering [ F]-FDG PET-CT has no additional value in the diagnosis the Blumgart system now less applicable. The classification and staging of PHC. In the hilar area, it is difficult to distin- system proposed by the International Cholangiocarcinoma guish tumor from concomitant inflammation. Furthermore, Group for the Staging of PHC takes into account most of the for the identification of nodal involvement, [ F]-FDG PET- variables used in the previous systems: suspicious lymph CT has a sensitivity and specificity of 67 and 68%, respective- nodes, extent of bile duct involvement, extent of vascular ly [74, 75]. Hence, it does not provide additional diagnostic involvement, suspected tumor size and lobar atrophy. As in yield in comparison with CT. the other systems, the information is largely descriptive [83]. The staging systems used to date are mainly surgery orient- Cholangiography MR cholangiography (MRC) combined ed. Each has its merits, but all are limited to the anatomical with MRI has comparable staging accuracy with that of CT description of the tumor and are therefore limited in their ability combined with direct cholangiography [72]. Alternatively, di- to predict the likelihood of an R0 resection. Furthermore stag- rect cholangiography using ERCP or percutaneous ing systems have been criticized for having poor predictable transhepatic cholangiography (PTC) can also be used. PTC quality in different populations [20, 79, 84]. Ideally, a staging may be more helpful in assessing the extent of proximal tumor system would preoperatively predict the likelihood of resect- infiltration. ERCP can also be combined with cytological or able disease along with as well, prognostic value. tissue sampling, albeit sensitivity and specificity are low (see above). A major disadvantage of direct cholangiography is its Staging laparoscopy invasiveness, including the risk of inducing infection, pancre- atitis, bleeding, inflammation and pain. Direct cholangiogra- For optimal determination of resectability, patients with po- phy for diagnostic purposes is therefore, only rarely per- tentially resectable PHC may undergo staging laparoscopy to formed. Especially ERCP entails retrograde contamination detect the presence of occult tumor manifestations. Staging of the obstructed bile ducts with increased risk of cholangitis. laparoscopy may detect small liver and/or peritoneal metasta- Subsequent drainage of the visualized bile ducts using one or ses that are undetectable on routine imaging avoiding a futile more stents is therefore mandatory. ERCP and PTC are pref- laparotomy [84–86, 155]. A thorough inspection of the liver, erably used for therapeutic purposes to drain the obstructed gallbladder, hepatoduodenal ligament and peritoneum is un- bile ducts in the palliative setting or preoperatively, to prepare dertaken. The lesser sac is routinely opened and the common the patient for resection. In the latter situation, the aim is to hepatic artery is examined, lymph node station 8 (N2) is then drain the biliary system of the future remnant liver while leav- identified and biopsied for pathological evaluation. All other ing the part to be resected alone. suspicious lesions, based on intraoperative inspection or pre- vious imaging, are biopsied for histopathological analysis. Staging systems Although not widely used, the combination with laparoscopic ultrasound has been reported to increase the yield of the stag- There are many factors associated with resectability, prognosis ing procedure to some extent. In a meta-analysis by Coelen and prediction of long and short-term survival after resection et al., which included 832 potentially resectable PHC patients, of PHC [15, 76–79]. The most commonly used staging sys- a pooled sensitivity of 52.2% was found to detect tems include the American Joint Committee on Cancer unresectability [14]. Based on our own experience in 273 (AJCC) staging system with incorporated TNM classification, patients undergoing staging laparoscopy for PHC, we devel- the Bismuth-Corlette system, the Blumgart T-staging system oped a risk score that estimates the chance of unresectability. (MSKCC classification) and a classification recently proposed This risk score includes the following factors: tumor size, by the International Cholangiocarcinoma Group for the stag- portal vein involvement, suspected lymph-node metastases ing of PHC [14, 15, 20, 77, 79–81]. The AJCC staging system and suspected (extra) hepatic metastases. It showed good 296 Langenbecks Arch Surg (2018) 403:289–307 discrimination between resectable and unresectable disease The FRL is delineated on the SPECT images for calcula- (AUC 0.77, 0.68–0.86 95% CI) [16]. tion of the functional share (%). This is then multiplied by the total liver function (MUR) to calculate function of specifically Assessment of future remnant liver the FRL. This method provides visual and quantitative infor- mation on regional liver function. Functional share of the FRL Liver volumetry is corrected for body surface area (BSA, m )using the Mosteller formula, to individualize the results based on the Since extended liver resections are often required, it is critical individual metabolic needs [100]. The current cutoff for a safe to assess the FRL preoperatively where CT-volumetric analy- resection is a FRL function of at least 2.7%/min/m [97]. sis is the standard technique. The segments of the FRL are HBS can be used in patients with normal or impaired qual- delineated on the CT images and the ratio of the remnant liver ity of liver parenchyma alike using the same cutoff value. and the total liver, with subtraction of tumor volume is calcu- MUR has been shown to correlate well with ICG clearance lated. This delineation technique gives an indirect measure- [101]. A limitation of using HBS in patients with PHC is that ment of the liver function [87, 88]. the uptake of bilirubin is competitive with mebrofenin as both It is assumed that a FRL-volume of > 25–30% is consid- are taken up by the same hepatocyte transporters [102]. In ered a safe cutoff for patients with healthy liver parenchyma, these patients, hepatocyte function is likely to be decreased whereas > 40% is used in patients with compromised liver, which will be reflected by HBS, with additional underestima- like in patients with (post)cholestatic liver that is damaged tion due to competition. These receptors are downregulated by longstanding biliary obstruction and possible cholangitis during hyperbilirubinemia, but their expression gradually nor- [88–90]. In literature the acceptable minimum volume of FRL malizes after drainage [103]. Considering this interaction, in regard with parenchymal disease is variable and controver- HBS should not be performed in patients with high bilirubin sial (10–40%) [91–93]. In PHC, a FRL volume of more than levels (> 30 μmol/L) and is usually postponed until adequate 40% is usually considered. A disadvantage of FRL volumetry biliary drainage has been achieved [104]. is that individual patient characteristics are not taken into ac- count and that the delineating technique is prone to error [94, Preoperative preparation of the patient 95]. Especially in patients with compromised liver, discrepan- cies have been reported between CT volumetry and postoper- Obstructive jaundice and biliary drainage ative outcomes [96] because the quality of the liver parenchy- ma is not taken into consideration [87, 97]. Patients with PHC usually present with obstructive jaundice. This phenomenon has a negative effect on liver function, in- Liver functional tests creases the risk of biliary infection and impairs cellular immu- nity [105]. Preoperative biliary drainage is used to create a Because liver volume does not equal liver function and func- safer environment prior to liver surgery. It reduces jaundice, tion is not homogeneously distributed in the liver [97], we rely improves liver function and the patient’s condition, at the more on assessment of the function of the FRL, rather than on same time improving the ability of the liver to regenerate 99m volume alone. Tc-mebrofenin hepatobiliary scintigraphy postoperatively [85, 105–109]. The impact of these effects is (HBS) is a validated quantitative dynamic liver function test particularly high in patients with an insufficient FLR and pre- for which mebrofenin, an iminodiacetic (IDA) derivate, is operative drainage has shown to improve outcomes especially used as a tracer. This agent is mainly taken up by the hepato- in patients requiring extended resections [108]. On the other cytes and is subsequently excreted in the bile without under- hand, drainage-related complications may be severe and it is going any biotransformation. The hepatic uptake is mediated therefore advisable to only drain patients with a substantially by the same transport mechanisms as that of various endo- and increased bilirubin and small FLR [85, 110]. Drainage-related exogenous substances, making it an ideal agent to assess liver complications such as cholangitis may severely deteriorate a function. HBS consists of an early dynamic phase, acquired patient’s condition and increase the risk of postoperative mor- directly after intravenous injection of mebrofenin, during bidity and mortality [50, 85, 105, 111]. Preoperative which the mebrofenin uptake rate (MUR, %/min) is measured cholangitis is caused by contamination of the biliary tract dur- [98]. This corresponds with the total liver function. ing drainage procedures. It is therefore advisable to give pro- Immediately afterwards, a SPECT acquisition combined with phylactic antibiotics previous to any drainage procedure [3, low-dose CT is made, falling in the period in which 23, 107, 112]. Any episode of cholangitis induced after biliary mebrofenin is accumulated in the liver. The SPECT data pro- decompression should be treated with antibiotics and addi- vide information on three-dimensional, segmental distribution tional drainage or drain revision if necessary [21]. of function. The low-dose CT is solely used for attenuation Refractory cholangitis is often caused by incomplete biliary correction and anatomical mapping [99]. drainage and requires adequate endoscopic or percutaneous Langenbecks Arch Surg (2018) 403:289–307 297 stenting of residual, obstructed parts of the biliary tract. of a range of interleukins and growth factors that induce hy- Patients should not undergo surgery earlier than that they have pertrophy of the non-embolized lobe. In our cohort of PHC fully recovered from cholangitis [21]. patients, the cutoff for proceeding with PVE is a FRL volume The optimal drainage method is still a much-debated topic, of less than 40% and/or function less than 2.7%/min/m .Inthe in which surgeons tend to favor the percutaneous approach for absence of cholangitis, the biliary system of the embolized reasons of direct access to the biliary duct and postoperative use lobe needs not be drained, since unilateral cholestasis may of the intraluminal drain(s) across the hepaticojejunostomy. even have a synergistic effect on the hypertrophy response Although percutaneous biliary drainage (PTBD) seems as- of the non-embolized lobe. After a period of 3 weeks after sociated with higher postoperative morbidity, further prospec- PVE, CT and HBS are repeated and reassessed. We have tive studies are needed to better define the optimal mode of shown that functional increase occurs more rapidly than vol- biliary drainage in PHC [112, 113]. Furthermore, PTBD might ume, suggesting a shorter waiting time until resection can take be complicated by portal vein thrombosis or seeding metasta- place [119]. sis that may change resectability of the tumor [110, 114]. For In the series reported by the Nagoya group, PVE showed to now, endoscopic biliary drainage (EBD) is still the preferred improve the surgical outcomes of PHC [10]. PVE is consid- method in most Western countries. ered a safe procedure with an overall morbidity rate of 2.2%. Endo-nasobilairy drainage (ENBD) is the advocated meth- Most common complications are hematoma, hemobilia, septic od in many Asian countries. As in PTBD, it provides more complications, backflow of embolization material and throm- precise information on the extent of cancer along the bile ducts bosis in the FRL [120]. Olthof et al. analyzed the incidence of [115]. Some authors reported less complications and a high postoperative liver failure in a combined series of two Western success rate of ENBD compared to EBD [116, 117]. However, centers specialized in PHC. A risk score was proposed to others reported comparable results with EBD and PTBD select candidates for PVE based on FRL volume combined [110]. Western centers generally do not perform ENBD be- with jaundice at presentation, preoperative cholangitis and cause the nasal tubes easily dislocate and from the patients’ preoperative bilirubin level > 50 μmol/L [121]. perspective, are usually less well tolerated. ENBD drains bile Accelerated tumor growth due to PVE does not seem to externally via the naso-gastro-duodenal tubes, precluding bile influence the survival of PHC patients [122, 123]. PVE how- entering the intestinal system and therefore demands bile sup- ever, does predetermine the side of the resection and in case of pletion. This is then only possible via the oral route or a sec- new findings that may require a change of strategy, this cannot ond gastroduodenal tube. be reversed. If the patient becomes unresectable due to disease Several retrospective studies have been performed progression in the waiting time, the atrophy-hypertrophy re- concerning the optimal drainage method [118] mainly empha- action stabilizes with time and the overall liver volume and sizing that each method comes with its own set of complica- function remain unchanged. However, the persistence of the tions such as cholangitis, pancreatitis or vascular complica- atrophied, usually contaminated cholestatic liver lobe can be tions. Until evidence has been presented, EBD remains the accompanied with adverse effects such as liver abscess, com- reference method in most Western countries [107, 109]. plicating further palliative treatment of the patient who typi- The balance between the benefits and risks of biliary drain- cally will need repeated treatment with biliary stents [124]. age is fragile and drainage strategies should be optimized in Additional embolization of segment 4 in preparation of ex- order to minimalize the risk of intrinsic complications. Due to tended right hemihepatectomy is an option depending on the these risks, it may be advisable to undertake surgery without target increase of FRL volume that needs to be attained. To this prior drainage provided there is a surplus of remnant liver end, the left portal vein branches to segment 4 are occluded volume. Wiggers et al. showed that with a FLR > 50% preop- along with embolization of the right portal venous system. The erative biliary drainage was of no added value [88]. technique is challenging and requires an experienced interven- Hence, in patients requiring extended resection, we now tional radiologist since access to the left portal venous system use selective preoperative biliary drainage of only the future can give additional risk of injury. Backflow of embolization remnant liver when FRL is small (< 50%) whereas complete material into the left portal venous system can lead to inadver- preoperative biliary drainage is mandatory in the event of tent embolization and thrombosis of the portal veins supplying (recent) cholangitis. the FRL. Alternatively, to decrease these risks, partial emboli- zation of only segment 4a can be performed [120, 156]. Portal vein embolization Preoperative radiotherapy If the FRL has not sufficient volume or function to undergo a safe resection, portal vein embolization (PVE) is the standard There is no general consensus regarding the use of neoadju- intervention to increase the functional capacity of the FRL. vant therapy for PHC. Low-dose preoperative radiotherapy The local hemodynamic changes proposedly result in a release (3 × 3.5 Gy in 3 days prior to resection) was instituted in our 298 Langenbecks Arch Surg (2018) 403:289–307 center in patients with resectable PHC who received preoper- sided approach comprising an anatomical left ative drainage to prevent seeding metastases. The increased hemihepatectomy, whereas coming from the right, an extended risk of seeding metastases after biliary drainage is an area of right hemihepatectomy is required to include the central sector. debate [114, 125] with various outcomes reported in literature Intraoperative frozen-section pathological examination of [86, 126]. In our cohort, seeding metastasis in up to 20% of the resection margins of the biliary ducts is performed to con- patients after endoscopic stenting has been observed in the firm radicality at the ductal level. In case of residual tumor in laparotomy scar or drain tract [86] . This complication was the resection margin, the level of biliary resection is extended associated with tumor cells contained in the bile that inevita- although in our series, survival was worse in these patients bly contaminated the operative field after bile duct transection compared to patients who had an initial free margin [108]. in the course of resection. After using preoperative radiation, Some surgeons routinely drain the biliary ducts of the remnant no catheter tract recurrences after drainage have been reported liver after reconstruction using trans-anastomotic tubes. We [125]. However, there is no evidence for this concept that is usually do not internally drain the hepaticojejunostomy but uniquely applied in our center [127]. A recent study conducted when there are PTC drains in place, these are positioned in two Western specialized centers, did not show an associa- across the anastomoses allowing access for possible postoper- tion of seeding metastases with center or mode of preoperative ative cholangiography. As recently reported, leaving the drainage, i.e. endoscopic or percutaneous drainage [125]. drains open is not advised since the loss of bile negatively New prospective studies are needed to develop guidelines on influences postoperative regeneration of the liver remnant this topic. [130]. Surgical aspects Resection of the portal vein bifurcation: unconditional or on demand General considerations Controversy exists regarding the unconditional, simultaneous The goal of surgical treatment is to achieve an R0 resection of en bloc resection of the portal vein bifurcation with the tumor. the tumor along with clearance of the regional lymph nodes. The hilar vessels run adjacent to the tumor with less than Because of the central location of the tumor at the liver hilum 1 mm between tumor and portal vein while perineural infiltra- and its proximal extension into the segmental bile ducts, com- tion of the tumor along the bile ducts is a common feature of plete resection requires excision of the extrahepatic biliary duct PHC. Because of this anatomical proximity, a no-touch tech- in combination with extended liver resection. The close relation nique was proposed by the Berlin group in which resection of the tumor with the right and left portal vein and the hepatic includes unconditional excision of the portal vein bifurcation artery branches often demands concomitant vascular resections en bloc with tumor excision and hepatectomy. Using this on- and reconstruction. The Japanese surgeons were the first to cological strategy, dissection of the hepatic hilum is avoided show in the nineties of the previous century, that this aggressive and the risk of dissemination of tumor cells minimized. This approach resulted in improved long-term survival [128]. Still technique however is less feasible in patients requiring left following these lines, radical resection entails excision of the hepatectomies due to the fact that the right hepatic artery usu- liver hilum with (extended) hemihepatectomy including seg- ally crosses the hepatic hilum directly anterior or posterior of ment IV and the caudate lobe, complete lymphadenectomy of the tumor, unless there is a displaced right hepatic artery orig- the hepatoduodenal ligament and excision of the portal vein inating from the superior mesenteric artery that runs along the bifurcation when involved [129]. Additionally, arterial resec- right-lateral margin of the hepatoduodenal ligament. Also, re- tions are undertaken in order to achieve an R0 resection (Fig. 5). construction of the right portal branches with the main stem is The type of resection depends on location of the tumor and technically more demanding. biliary anatomy at the hepatic duct confluence, the radial and Multivariate analysis of the Berlin series of resected PHC longitudinal extent of tumor into the intrahepatic bile ducts and showed that portal vein resection was the only significant its association with adjacent periductal structures, portal vein factor to influence patient survival after confirmed R0 resec- and hepatic arteries. When viewing cross-sectional imaging tion. The 5-year survival rate of curative liver resection in their studies, it is important to perceive the tumor in a three- series was 65% with portal vein resection as compared to 28% dimensional fashion. The tumor extends along the right and left without [131]. Other authors advise to only perform portal biliary ducts into the liver and at the same time, in anterior and vein resection when during exploration, the portal venous posterior direction into the ducts of segment 4 and segment 1, bifurcation/contralateral portal venous branch is found to be respectively. Complete resection therefore should include the invaded by tumor. central sector of the liver along the antero-posterior axis includ- In our center, we do not advocate the unconditional exci- ing the segments 1 and 4 [62]. As the central sector anatomi- sion of the portal vein bifurcation, also because portal venous cally is part of the left hemi-liver, there is an advantage of a left- reconstruction has been associated with an increased Langenbecks Arch Surg (2018) 403:289–307 299 Fig. 5 Extended resection for PHC should include the central sector (segment 4) with segment 1 along the antero-posterior axis of the liver. Depending on the predominant side of the tumor, a left (extended) or right extended hemihepatectomy is chosen for en bloc resection of the hilar area postoperative morbidity rate [88]. Modern preoperative imag- tumor involvement. The right hepatic artery with its right an- ing techniques now accurately demonstrate portal venous in- terior and posterior branches however, is frequently infiltrated volvement. Relying on this information, we only resect the by tumor. The choice of performing a right or left resection is portal vein bifurcation en bloc with the tumor right away when often dictated by the side of the liver in which the hepatic there is evidence of vascular involvement on CT-scan. artery branches are free. In PHC predominantly involving Otherwise, the decision to resect the portal vein bifurcation the left liver, (extended) left hemihepatectomy with concom- is made intraoperatively. This policy has led to portal vein itant resection of the right hepatic artery is hampered by pre- resection in approximately 20% of our cases (Table 2). serving a tumor free, intrahepatic distal stump for arterial re- Survival analysis of our series showed an overall 5-year sur- construction especially when tumor mass is substantial. vival of 44.3%, which is comparable with the 5-year survival Microsurgical techniques are usually applied to create a safe of 43% reported by the Berlin group in a series of patients that anastomosis with the right posterior branch of the right hepatic underwent R0 resection using unconditional en bloc portal artery in these cases. Combined arterial resection and recon- vein resection [132]. struction with portal vein resection is controversial. In the Concomitant resection of the portal vein bifurcation with Nagoya cohort of resected patients, this challenging technique extended hemihepatectomy is followed by end-to-end anasto- was associated with a mortality of 2% and a 5-year survival of mosis of the left portal vein with the main portal venous stem. 30% for patients with advanced cholangiocarcinoma [133]. Complete mobilization of the left portal vein by detaching all Preoperative embolization of the branch of the hepatic ar- side-branches to segments 4 and 1 facilitates reconstruction. tery feeding the future remnant liver and subsequent excision As the caliber of the left portal vein usually is much smaller, it is not attractive in our view, because the biliary anastomosis is important to bevel the anastomosis after oblique clamping depends on arterial periductal arterial perfusion and loss of in order to prevent stenosis of the anastomosis. With arterial blood supply will lead to failure of the (extended) left hepatectomy, reconstruction of the right portal hepaticojejunostomies. Arterialization of the portal vein can branches with the main stem is technically more demanding. be used as last resort when arterial perfusion of the remnant liver is sacrificed, however is preferably applied for salvage of the remnant liver [134]. Concomitant resection of the hepatic artery branch feeding the remnant liver Concomitant resection of segment 1 Adequate arterial perfusion is crucial to function of the rem- nant liver. The left hepatic artery runs along the medial side of As pointed out above, the caudate lobe is part of the central the hepatoduodenal ligament and is therefore less at risk for antero-posterior axis of the liver and is preferably resected 300 Langenbecks Arch Surg (2018) 403:289–307 en bloc with the tumor and liver hilum. Although the seg- Parenchyma sparing liver resection ment 1 bile ducts often drain into the left ductal system, they may drain into any part of the hepatic duct confluence and As volume and function of the remnant liver are the most these ducts are frequently involved by tumor as well. critical factors for postoperative outcomes, parenchyma pre- Routine S1 resection en bloc with (extended) serving techniques can be applied in selected cases. These hemihepatectomy has therefore been implemented at our techniques can be used as an alternative to PVE or in addition institution since 1998 and has increased the rate of R0 re- to PVE, in order to spare as much functional liver tissue as sections and has resulted in improved survival [135]. En possible. bloc excision of segment 1 is therefore recommended with In right-sided tumors that require an extended right resection of PHC. hemihepatectomy, the cranial part of segment 4 (i.e. 4a) may be preserved depending on the level of involvement of the segment 4 bile duct. Free margin of the cut segment 4 bile duct Right or left (extended) hepatectomy is checked using frozen-section pathological examination. In case of a left-sided tumor, a modified extended left The decision to perform a right or left hepatectomy depends hemihepatectomy may be undertaken. Extending left resec- on local tumor extension, portal venous and hepatic arterial tion to include segment 5, the adjacent part of segment 8 involvement and the FRL volume and function. For optimal may be preserved. Whether this can be performed depends preoperative preparation of the patient, such as biliary drain- on the proximal extent of the tumor into the right segmental age or the need for PVE, it is important to preoperatively ducts, and the anatomy of the right sectoral ducts (B5/8 and determine the side of the liver to be resected. B6/7) in relation to the right hepatic duct and hepatic duct If the tumor extent and FRL volume allow both options, a confluence. right-sided (extended) hepatectomy is often preferred because A pitfall of sparing portions of the central sectoral segments it is more likely to achieve oncological radicality and is more a 4 and 8 is cutting off their portal venous and arterial blood straightforward procedure for several reasons [136]. Firstly, supply by resection of the tumor. The central position of the because the biliary confluence is located on the right side of tumor often requires sacrifice of the middle hepatic artery to the hepatoduodenal ligament, a right hepatectomy allows segment 4 or the right-anterior portal vein and hepatic artery more complete resection of the tumor. Also, the right hepatic branches to segment 8 leading to parenchymal infarction. duct is often short (< 1 cm) or even absent in case of a triple Another possibility for parenchymal preservation is hepatic confluence, while the left hepatic duct has a relatively performing a central liver resection (mesohepatectomy) when long and straight course until reaching the border of the left the bile ducts of segments 6 and 7 and the left lateral segments 2 and 3 are not infiltrated by the tumor. This complex proce- portal vein and branching off into the ducts of segments 2 and 3[137]. Therefore, tumors that invade the right sectoral ducts dure includes resection of the central sectors of the liver in- and even the segmental ducts to segment 4 (Bismuth-Corlette cluding segments 4, 5 and 8. In these cases, multiple jejunal type IIIa-IV) can be radically resected by extended right anastomoses with the remaining intrahepatic segmental bile hemihepatectomy. A disadvantage of a right-sided approach ducts are required [129]. A formal central resection is only is that segments 2 and 3 are small and that in many patients possible when the vascular structures supplying the left lateral preoperative right PVE is necessary before undertaking ex- segments as well as the right-posterior segments 6 and 7 are tended right hemihepatectomy. free of the tumor and can be preserved. Tumors predominantly involving the left ductal system (Bismuth-Corlette type IIIb-IV) require a left-sided ap- ALPPS proach. The advantage of a left-sided resection is that the remnant liver, i.e. the right liver segments, usually has more In situ split of the liver in combination with portal vein ligation volume and resection can be extended farther into the right (ALPPS) has been introduced as a method to induce rapid liver. Of note, the volume of segments 6 + 7 usually exceeds hypertrophy of the FRL. Because of the higher mortality and that of segments 2 + 3 which may direct the choice of a right morbidity reported in the initial series of ALPPS, this method or left-sided approach. A formal extended left along with great interest has generated a heated discussion in hemihepatectomy following the medial margin of the right the surgical community [138, 139]. The advantage of ALPPS hepatic vein is technically more difficult and depending on is debated in extended right hemihepatectomy as compared to involvement of the segment 8 ducts, part of the anterior right complete embolization of the right portal venous system in- sector may be preserved (see below). A down-side of a left cluding segment 4 as described above. Several authors have resection is that construction of the (often multiple) biliary reported their results of ALPPS in patients with PHC. Due to anastomoses may be more complex with a higher risk of stenting of the biliary system and ensuing cholangitis, patients biliary complications. were at increased risk of interstage morbidity and mortality Langenbecks Arch Surg (2018) 403:289–307 301 [140, 141]. ALPPS for PHC demonstrated poor outcomes Liver failure is a dreaded complication after extensive hepa- with 48% perioperative mortality in the ALPPS registry tectomy and is a major cause of mortality in patients with PHC [140]. We therefore for now, do not recommend ALPPS for [21, 148]. The risk of postoperative liver failure is increased resection of PHC and rather consider PVE with selective em- due to the combination of intra-operative blood loss, a small bolization of the left portal vein branches to segment 4 for FLR and cholestasis [21, 88, 121]. Reported liver failure augmentation of FRL volume in patients requiring extended ranges from 3 to 25% [31, 88, 143, 144]. Biliary leakage from right hemihepatectomy. either the hepaticojejunal anastomosis or the parenchymal dis- section surface ranges from 6 to 29% [31, 88, 143]. Infections range from 23 to 66% and bleeding complications from 4 to The extent of lymphadenectomy 9% [15, 31, 88, 145, 146](Table 5). Standard lymphadenectomy includes resection of lymph nodes around the extrahepatic bile duct, the portal vein Discussion and future perspectives and hepatic artery, as well as the lymphatic channels and nerves contained in the hepatoduodenal ligament. The The management of perihilar cholangiocarcinoma is complex number of lymph nodes resected is also relevant as less and requires close multidisciplinary collaboration between than 4 lymph nodes evaluated in the specimen was iden- hepatobiliary surgeons, endoscopists, radiologists, medical tified as a poor prognostic factor for time to recurrence oncologists and pathologists. In this review, we provide a [142]. Lymphnodemetastasesthatare limitedtothe he- summary of the current diagnosis and work-up in the light patic pedicle or the hepatoduodenal ligament (N1) are of extended resection and elaborate on future perspectives. included in the field of resection, but those along the Establishing the diagnosis of PHC is still one of the most common hepatic artery and/or celiac axis (N2) are consid- challenging aspects of the diagnostic work-up. New diagnos- ered distant metastases. N2-disease has a poor prognosis tic endoscopic techniques such as SpyGlass and endoscopic and disease specific survival of patients with para-aortic ultrasound enable more precise biopsies, resulting in increased lymph node metastasis was similar to M1 patients, sug- sensitivity and specificity in diagnosing biliary strictures [33]. gesting that survival is not influenced by the extent of These techniques will likely decrease the rate of misdiagnoses lymph node dissection, but rather by the presence of N2 and bring down the number of futile resections performed for disease [12, 78]. Therefore, we do not recommend routine benign disease. The currently used conventional tumor marker lymphadenectomy beyond the hepatoduodenal ligament. CA19–9 is not particularly sensitive or specific. A combina- tion of different markers seems more useful in the diagnosis Complications and follow-up of PHC. Biomarkers such as CYFRA21-1 and MUC-5 need to be evaluated in larger cohorts to assess its Postoperative morbidity and mortality of patients with PHC is clinical value. Molecular techniques such as circulating significant. Reported mortality ranges from 5 to 18% even in miRNA’s and Tumor Educated Platelets (TEP) represent an high volume centers, and morbidity as high as 60–70%, with exciting area with great promise in this field [149, 150]. For around 50% severe complications (Clavien-Dindo grade III or now, approximately 50% of patients with suspicion on PHC higher) [21]. Table 5 shows the incidence of the most common undergo resection without a confirmed tissue diagnosis. complications as reported in literature together with the com- CT-volumetry has traditionally been the golden standard plications recorded in our own series. Risks are particularly for assessment of a sufficient FRL. However, not only the high in patients with tumors requiring an extended resection. quantity but also the quality of the FRL is important while liver volume does not correlate with liver function. In our cohort, total and regional (segmental) liver function was pre- Table 5 Complications and reported incidence in a selection of operatively evaluated using HBS. This quantitative method literature reports including the AMC series allows measurement of FRL-function and can be used in pa- Complication type Incidence literature Incidence AMC tients with impaired quality of liver parenchyma using the same cutoff value. A limitation of using HBS in patients with Liver failure 3–25% [31, 88, 143, 144] 19% (29/156*) PHC is that bilirubin induces competitive uptake with Biliary leakage 6–29% [31, 88, 143] 30% (47/156) mebrofenin as both are taken up by the same hepatocyte trans- Bleeding 4–9% [15, 31, 88, 145, 146] 8% (13/156) porters. In patients with obstructive cholestasis, HBS may Multi organ failure 1–3% [32, 87, 135] 2% (3/156) underestimate liver function when the biliary system is not Infections 23–66% [15, 31, 88, 145, 146] 22% (35/156) completely drained. Mortality 5–17% [88, 143, 147, 157] 9% (18/201) The additional value of staging laparoscopy in the future is *Total cohort: n = 156, missing n =45 questionable, considering the low yield and further 302 Langenbecks Arch Surg (2018) 403:289–307 improvements in accuracy of preoperative imaging tech- in a highly selected group of patients [153, 154]. The idea of a niques. Using a risk score allows to predict unresectable short course of radiation preceding resection was to eradicate PHC at staging laparoscopy in order to make a more selective free floating tumor cells in the bile, thus preventing viable approach to staging laparoscopy. tumor cells of contaminating the peritoneal surface. There is Since patients with PHC typically present with obstructive however no evidence for this concept. jaundice, decompression of the biliary tract is a much-debated The challenge in the coming years is to reduce morbidity topic. For the past, few years it has become clear that drainage and mortality associated with extended resections for PHC. of the biliary tract comes with a serious risk of drainage- Optimizing preoperative workup is key to achieve improved related complications. Since obstructive jaundice impairs liver outcomes after extended resections. regeneration, biliary drainage is still advised in case of a small FLR. The optimal drainage method has yet to be established. In The Netherlands the DRAINAGE trial is underway to eval- Conclusion uate outcomes of PTBD vs. EBD in resectable PHC [108, 151]. In this multi-center trial with an all-comers design, all The field of work-up in PHC is changing with the introduction patients with a presumed PHC and cholestasis are randomized of newer modalities that have emerged over the past few to undergo PTBD or EBD. The study is powered for drainage- years. Upcoming diagnostic modalities and molecular tech- related complications and postoperative outcomes. For now, niques might help to decrease the rate of misdiagnosis of be- we use selective preoperative, endoscopic biliary drainage of nign, inflammatory disease. Assessment of liver function with only the future remnant liver when FRL is small (< 50%) hepatobiliary scintigraphy provides better information on the unless mandated by cholangitis. FRL than volume alone. The selective use of staging laparos- The most important prognostic factor for long-term surviv- copy is advisable to avoid futile laparotomies. In patients re- al of PHC is a margin negative resection of the hilar tumor. In quiring extended resection, selective preoperative biliary experienced hands, even Bismuth-Corlette type IV tumors can drainage is mandatory in cholangitis and when FRL is small be resected with curative intent. R0 resection requires an ag- (< 50%). Preoperative PVE is used when FRL volume is less gressive surgical approach encompassing hilar resection in than 40% and optionally includes the left portal vein branches combination with extended liver resection, frequently accom- to segment 4. ALPPS as alternative to PVE is not recommend- panied with vascular reconstructions. These extended resec- ed in PHC. N2 positive lymph nodes preclude long-term sur- tions are associated with higher morbidity and mortality rates vival. The benefit of unconditional en bloc resection of the than experienced in liver resections without bile duct resec- portal vein bifurcation is uncertain. Although still associated tion, probably because of the sequelae of obstructive jaundice. with considerable morbidity and mortality, an aggressive sur- Survival after resection is however favorable, with 5-year gical approach encompassing extended liver resection includ- overall survival rates comparable with survival after extended ing segment 1, regional lymphadenectomy and conditional liver resection for colorectal liver metastases. portal venous resection offers the only chance for long-term PVE is a widely accepted interventional procedure to in- survival. crease FRL volume and function before undertaking major liver resection. This method of liver augmentation is especial- Authors’ contributions F. Rassam: study conception and design, acquisi- ly of benefit in patients with PHC who require extended liver tion of data, analysis and interpretation of data, drafting of manuscript. E. Roos: study conception and design, acquisition of data, analysis and resection in predamaged livers. We therefore advocate the interpretation of data, drafting of manuscript. K.P. van Lienden: critical liberal use of PVE in patients with PHC in whom the FRL is revision of manuscript. J.E. van Hooft: critical revision of manuscript. below 40% of total liver volume. It is important to note that to H.J. Klümpen: critical revision of manuscript. G. van Tienhoven: critical obtain the maximum hypertrophy effect of segments 2 and 3, revision of manuscript. R.J. Bennink: critical revision of manuscript. M.R. Engelbrecht critical revision of manuscript. A. Schoorlemmer: crit- the side-branches of the left portal vein to segment 4 can be ical revision of manuscript. U.H.W. Beuers: critical revision of manu- embolized as well. Obviously, selective embolization of the script. J. Verheij: critical revision of manuscript. M.G. Besselink: critical segment 4 branches requires expertise of the interventional revision of manuscript. O.R.C. Busch: critical revision of manuscript. radiologist as available in specialized centers. Although the T.M. van Gulik: Study conception and design, critical revision of first successful case of ALPPS was reported by Schlitt in a manuscript. patient with PHC, the use of ALPPS in PHC as alternative to Compliance with ethical standards PVE is not recommended because of the reported high mor- bidity and mortality of the procedure in this category of pa- Conflict of interest The authors declare that they have no conflict of tients [152]. interest. There are no established strategies regarding the use of neo-adjuvant therapies in PHC. The only exception is neo- Ethical approval This article does not contain any studies with human adjuvant chemo-radiation therapy prior to liver transplantation participants or animals performed by any of the authors. Langenbecks Arch Surg (2018) 403:289–307 303 Open Access This article is distributed under the terms of the Creative resectability and outcomes in 380 patients. J Am Coll Surg 215: Commons Attribution 4.0 International License (http:// 343–355. https://doi.org/10.1016/j.jamcollsurg.2012.05.025 creativecommons.org/licenses/by/4.0/), which permits unrestricted use, 16. Coelen RJS, Ruys AT, Wiggers JK, Nio CY, Verheij J, Gouma DJ, distribution, and reproduction in any medium, provided you give appro- Besselink MGH, Busch ORC, van Gulik TM (2016) Development priate credit to the original author(s) and the source, provide a link to the of a risk score to predict detection of metastasized or locally ad- Creative Commons license, and indicate if changes were made. vanced perihilar cholangiocarcinoma at staging laparoscopy. Ann Surg Oncol 23(S5):904–910. https://doi.org/10.1245/s10434-016- 5531-6 17. Valle J, Wasan H, Palmer DH et al (2010) Cisplatin plus gemcitabine versus gemcitabine for biliary tract cancer. N Engl J References Med 4:395–397. https://doi.org/10.1586/egh.10.45 18. Wyluda E, Yee NS (2015) Systemic treatment of advanced biliary tract carcinoma: emerging roles of targeted therapy and molecular 1. Aljiffry M, Abdulelah A, Walsh M et al (2009) Evidence-based profiling. Clin Cancer Drugs 2:80–86 approach to cholangiocarcinoma: a systematic review of the cur- 19. Ito F, Cho CS, Rikkers LF, Weber SM (2009) Hilar cholangiocar- rent literature. J Am Coll Surg 208:134–147. https://doi.org/10. cinoma: current management. Ann Surg 250:210–218. https://doi. 1016/j.jamcollsurg.2008.09.007 org/10.1097/SLA.0b013e3181afe0ab 2. Gatto M, Alvaro D (2010) New insights on cholangiocarcinoma. 20. Groot Koerkamp B, Wiggers JK, Gonen M et al (2015) Survival World J Gastrointest Oncol 2:136–145. https://doi.org/10.4251/ after resection of perihilar cholangiocarcinoma-development and wjgo.v2.i3.136 external validation of a prognostic nomogram. Ann Oncol 26: 3. Khan SA, Thomas HC, Davidson BR, Taylor-Robinson SD 1930–1935. https://doi.org/10.1093/annonc/mdv279 (2005) Cholangiocarcinoma. Lancet (London, England) 366: 21. Coelen RJS, Olthof PB, van Dieren S et al (2016) External vali- 1303–1314. https://doi.org/10.1016/S0140-6736(05)67530-7 dation of the estimation of physiologic ability and surgical stress 4. Bismuth H, Nakache R, Diamond T (1992) Management strate- (E-PASS) risk model to predict operative risk in perihilar cholan- gies in resection for hilar cholangiocarcinoma. Ann Surg 215(1): giocarcinoma. JAMA Surg 147:26–34. https://doi.org/10.1001/ 31–38. https://doi.org/10.1097/00000658-199201000-00005 jamasurg.2016.2305 5. Vaeteewoottacharn K, Seubwai W, Bhudhisawasdi V et al (2014) 22. Hemming AW, Reed AI, Fujita S et al (2005) Surgical manage- Potential targeted therapy for liver fluke associated cholangiocar- ment of hilar cholangiocarcinoma. Ann Surg 241:693–699-702. cinoma. J Hepatobiliary Pancreat Sci 21:362–370. https://doi.org/ https://doi.org/10.1097/01.sla.0000160701.38945.82 10.1002/jhbp.65 23. Anderson JE, Hemming AW, Chang DC et al (2012) Surgical 6. Ong CK, Subimerb C, Pairojkul C et al (2012) Exome sequencing management trends for cholangiocarcinoma in the USA 1998- of liver flukeg-associated cholangiocarcinoma. Nat Genet 44: 2009. J Gastrointest Surg 16:2225–2232. https://doi.org/10.1007/ 690–693. https://doi.org/10.1038/ng.2273 s11605-012-1980-9 7. Rizvi S, Gores GJ (2013) Pathogenesis, diagnosis, and manage- 24. Nagino M, Ebata T, Yokoyama Y et al (2013) Evolution of surgi- ment of cholangiocarcinoma. Gastroenterology 145:1215–1229. cal treatment for perihilar cholangiocarcinoma: a single-center 34- https://doi.org/10.1053/j.gastro.2013.10.013 year review of 574 consecutive resections. Ann Surg 258:129– 8. Sarkar S, Bowlus CL (2016) Primary sclerosing cholangitis: mul- 140. https://doi.org/10.1097/SLA.0b013e3182708b57 tiple phenotypes, multiple approaches. Clin Liver Dis 20:67–77. 25. Maillette de Buy Wenniger LJ, Beuers U (2015) Immunoglobulin https://doi.org/10.1016/j.cld.2015.08.005 G4-related cholangiopathy: clinical and experimental insights. 9. Boberg KM, Schrumpf E, Bergquist A et al (2000) Curr Opin Gastroenterol 31:252–257. https://doi.org/10.1097/ Cholangiocarcinoma in primary sclerosing cholangitis: K-ras mu- mog.0000000000000170 tations and Tp53 dysfunction are implicated in the neoplastic 26. Hubers LM, Maillette de Buy Wenniger LJ, Doorenspleet ME et al development. J Hepatol 32:374–380 (2015) IgG4-associated cholangitis: a comprehensive review. Clin 10. Nagino M, Ebata T, Yokoyama Y et al (2013) Evolution of surgi- Rev Allergy Immunol 48:198–206. https://doi.org/10.1007/ cal treatment for perihilar cholangiocarcinoma: a single-center 34- s12016-014-8430-2 year review of 574 consecutive resections. Ann Surg 258:129– 27. Zaydfudim VM, Wang AY, De Lange EE et al (2015) IgG4- 140. https://doi.org/10.1097/SLA.0b013e3182708b57 associated cholangitis can mimic hilar cholangiocarcinoma. Gut 11. Govil S, Reddy MS, Rela M (2014) Surgical resection techniques Liver 9:556–560. https://doi.org/10.5009/gnl14241 for locally advanced hilar cholangiocarcinoma. Langenbeck's 28. Zen Y, Britton D, Mitra V et al (2015) A global proteomic study Arch Surg 399:707–716. https://doi.org/10.1007/s00423-014- identifies distinct pathological features of IgG4-related and prima- 1216-4 ry sclerosing cholangitis. Histopathology. https://doi.org/10.1111/ 12. Regimbeau JM, Fuks D, Le Treut YP et al (2011) Surgery for hilar his.12813 cholangiocarcinoma: a multi-institutional update on practice and 29. Deshpande V, Zen Y, Chan JK et al (2012) Consensus statement outcome by the AFC-HC study group. J Gastrointest Surg 15: on the pathology of IgG4related disease. Mod Pathol 25:11811192 480–488. https://doi.org/10.1007/s11605-011-1414-0 www.modernpathology.org. https://doi.org/10.1038/modpathol. 13. Hartog H, Ijzermans JNM, van Gulik TM, Koerkamp BG (2016) 2012.72 Resection of perihilar cholangiocarcinoma. Surg Clin North Am 30. Kamisawa T, Zen Y, Pillai S, Stone JH (2015) IgG4-related dis- 96:247–267. https://doi.org/10.1016/j.suc.2015.12.008 ease. Lancet 385:1460–1471. https://doi.org/10.1016/S0140- 6736(14)60720-0 14. Coelen RJS, Ruys AT, Besselink MGH, Busch ORC, van Gulik TM (2016) Diagnostic accuracy of staging laparoscopy for detect- 31. Kloek JJ, van Deldein OM, Erdogan D et al (2008) Differentiation ing metastasized or locally advanced perihilar cholangiocarcino- of malignant and benign proximal bile duct strictures: the diag- ma: a systematic review and meta-analysis. Surg Endosc 30(10): nostic dilemma. World J Gastroenterol 14:5032–5038. https://doi. 1–11. https://doi.org/10.1007/s00464-016-4788-y org/10.3748/wjg.14.5032 15. Matsuo K, Rocha FG, Ito K et al (2012) The blumgart preopera- 32. Corvera CU, Blumgart LH, Darvishian F et al (2005) Clinical and tive staging system for hilar cholangiocarcinoma: analysis of pathologic features of proximal biliary strictures masquerading as 304 Langenbecks Arch Surg (2018) 403:289–307 hilar cholangiocarcinoma. J Am Coll Surg 201:862–869. https:// 48. Kurihara T, Yasuda I, Isayama H et al (2016) Diagnostic and therapeutic single-operator cholangiopancreatoscopy in doi.org/10.1016/j.jamcollsurg.2005.07.011 biliopancreatic diseases: prospective multicenter study in Japan. 33. Victor DW, Sherman S, Karakan T, Khashab MA (2012) Current World J Gastroenterol 22:1891–1901. https://doi.org/10.3748/ endoscopic approach to indeterminate biliary strictures. World J wjg.v22.i5.1891 Gastroenterol 18:6197–6205. https://doi.org/10.3748/wjg.v18. 49. Navaneethan U, Hasan M, Lourdusamy V et al (2015) Single- i43.6197 operator cholangioscopy and tarhetted biopsies in the diagnosis 34. Esnaola NF, Meyer JE, Karachristos A, Maranki JL, Camp ER, of inditerminat ebiliary strictures: a systematic review. Denlinger CS (2016) Evaluation and management of intrahepatic Gastrointest Endosc 82:608–614. https://doi.org/10.1007/978-1- and extrahepatic cholangiocarcinoma. Cancer 122(9):1349–1369. 4939-2914-6 https://doi.org/10.1002/cncr.29692 50. Hara K, Yamao K, Mizuno N, Hijioka S, Imaoka H, Tajika M, 35. Navaneethan U, Njei B, Lourdusamy V et al (2016) Comparative Tanaka T, Ishihara M, Okuno N, Hieda N, Yoshida T, Niwa Y effectiveness of biliary brush cytology and intraductal biopsy for (2016) Endoscopic ultrasonography-guided biliary drainage: who, detection of malignant biliary strictures: a systematic review and when, which, and how? World J Gastroenterol 22(3):1297–1303. meta-analysis. Gastrointest Endosc. 1:23–30. https://doi.org/10. https://doi.org/10.3748/wjg.v22.i3.1297 1007/s40778-014-0003-z.Genome 51. Lee JH, Salem R, Aslanian H et al (2004) Endoscopic ultrasound 36. Barr Fritcher EG, Kipp BR, Halling KC, Clayton AC (2014) and fine-needle aspiration of unexplained bile duct strictures. Am FISHing for pancreatobiliary tract malignancy in endoscopic J Gastroenterol 99:1069–1073. https://doi.org/10.1111/j.1572- brushings enhances the sensitivity of routine cytology. 0241.2004.30223.x Cytopathology 25:288–301. https://doi.org/10.1111/cyt.12170 52. Itoi T, Itokawa F, Uraoka T et al (2013) Novel EUS-guided 37. Liew ZH, Loh TJZ, Lim TKH et al (2017) Role of fluorescence in gastrojejunostomy technique using a new double-balloon enteric situ hybridization in diagnosing cholangiocarcinoma in indetermi- tube and lumen-apposing metal stent (with videos). Gastrointest nate biliary strictures. J Gastroenterol Hepatol. https://doi.org/10. Endosc 78:934–939. https://doi.org/10.1016/j.gie.2013.09.025 1111/jgh.13824 53. Masaki Y, Kurose N, Yamamoto M, Takahashi H, Saeki T, Azumi 38. Barr Fritcher EG, Voss JS, Brankley SM et al (2015) An optimized A, Nakada S, Matsui S, Origuchi T, Nishiyama S, Yamada K, set of fluorescence in situ hybridization probes for detection of Kawano M, Hirabayashi A, Fujikawa K, Sugiura T, Horikoshi pancreatobiliary tract cancer in cytology brush samples. M, Umeda N, Minato H, Nakamura T, Iwao H, Nakajima A, Gastroenterology 149:1813–1824e1. https://doi.org/10.1053/j. Miki M, Sakai T, Sawaki T, Kawanami T, Fujita Y, Tanaka M, gastro.2015.08.046 Fukushima T, Eguchi K, Sugai S, Umehara H (2012) Cutoff 39. Gonda TA (2017) Mutation profile and fluorescence in situ hy- values of serum IgG4 and histopathological IgG4+ plasma cells bridization analyses increase detection of malignancies in biliairy for diagnosis of patients with IgG4-related disease. Int J strictures. Clin Gastroenterol Hepatol. https://doi.org/10.1016/j. Rheumatol 2012:0–5. https://doi.org/10.1155/2012/580814 cgh.2016.12.013.This 54. Oseini AM, Chaiteerakij R, Shire AM et al (2011) Utility of serum 40. Fogel EL, DeBellis M, McHenry L et al (2006) Effectiveness of a immunoglobulin G4 in distinguishing immunoglobulin G4- new long cytology brush in the evaluation of malignant biliary associated cholangitis from cholangiocarcinoma. Hepatology 54: obstruction: a prospective study. Gastrointest Endosc 63:71–77. 940–948. https://doi.org/10.1002/hep.24487 https://doi.org/10.1016/j.gie.2005.08.039 55. Doorenspleet ME, Hubers LM, Culver EL et al (2016) IgG4+ B- 41. Coté GA, Sherman S (2011) Biliary stricture and negative cytol- cell receptor clones distinguish IgG4-related disease from primary ogy: what next? Clin Gastroenterol Hepatol 9:739–743. https:// Sclerosing cholangitis and biliary/pancreatic malignancies. doi.org/10.1016/j.cgh.2011.04.011 Hepatology 64(2):1–49. https://doi.org/10.1002/hep.28568 42. De Bellis M, Fogel EL, Sherman S et al (2003) Influence of stric- 56. Charatcharoenwitthaya P, Enders FB, Halling KC, Lindor KD ture dilation and repeat brushing on the cancer detection rate of (2008) Utility of serum tumor markers, imaging, and biliary cy- brush cytology in the evaluation of malignant biliary obstruction. tology for detecting cholangiocarcinoma in primary sclerosing Gastrointest Endosc 58:176–182. https://doi.org/10.1067/mge. cholangitis. Hepatology 48:1106–1117. https://doi.org/10.1002/ 2003.345 hep.22441 43. Fukuda Y, Tsuyuguchi T, Sakai Yet al (2005) Diagnostic utility of 57. Grunnet M, Mau-Sørensen M (2014) Serum tumor markers in bile peroral cholangioscopy for various bile-duct lesions. Gastrointest duct cancer—a review. Biomarkers 19:437–443. https://doi.org/ Endosc 62:374–382. https://doi.org/10.1016/j.gie.2005.04.032 10.3109/1354750X.2014.923048 44. Kawashima H, Itoh A, Ohno E et al (2012) Transpapillary biliary 58. Viterbo D, Gausman V, Gonda T (2016) Diagnostic and therapeu- forceps biopsy to distinguish benign biliary stricture from malig- tic biomarkers in pancreaticobiliary malignancy. World J nancy: how many tissue samples should be obtained? Dig Endosc Gastrointest Endosc 8:128–142. https://doi.org/10.4253/wjge.v8. 24:22–27. https://doi.org/10.1111/j.1443-1661.2012.01253.x i3.128 45. Kitajima Y, Ohara H, Nakazawa T et al (2007) Usefulness of 59. Liu SL, Song ZF, Hu QG et al (2010) Serum carbohydrate antigen transpapillary bile duct brushing cytology and forceps biopsy for (CA) 19-9 as a prognostic factor in cholangiocarcinoma: a meta- improved diagnosis in patients with biliary strictures. J analysis. Front Med China 4:457–462. https://doi.org/10.1007/ Gastroenterol Hepatol 22:1615–1620. https://doi.org/10.1111/j. s11684-010-0240-1 1440-1746.2007.05037.x 60. Patel AH, Harnois DM, Klee GG et al (2000) The utility of CA 19- 46. Lin LF, Siauw CP, Ho KS, Tung JN (2003) Guidewire technique 9 in the diagnoses of cholangiocarcinoma in patients without pri- for endoscopic transpapillary procurement of bile duct biopsy mary sclerosing cholangitis. Am J Gastroenterol 95:204–207. specimens without endoscopic sphincterotomy. Gastrointest https://doi.org/10.1016/S0002-9270(99)00744-3 Endosc 58:272–274. https://doi.org/10.1067/mge.2003.329 61. Chapman MH, Sandanayake NS, Andreola F et al (2011) 47. Chen YK, Pleskow DK (2007) SpyGlass single-operator peroral Circulating CYFRA 21-1 is a specific diagnostic and prognostic cholangiopancreatoscopy system for the diagnosis and therapy of biomarker in biliary tract cancer. J Clin Exp Hepatol 1:6–12. bile-duct disorders: a clinical feasibility study (with video){a fig- https://doi.org/10.1016/S0973-6883(11)60110-2 ure is presented}. Gastrointest Endosc 65:832–841. https://doi. 62. Van Gulik TM, Kloek JJ, Ruys AT et al (2011) Multidisciplinary org/10.1016/j.gie.2007.01.025 management of hilar cholangiocarcinoma (Klatskin tumor): Langenbecks Arch Surg (2018) 403:289–307 305 extended resection is associated with improved survival. Eur J external validation of a prognostic nomogram. Ann Oncol 26: 1930–19 Surg Oncol 37:65–71. https://doi.org/10.1016/j.ejso.2010.11.008 35. https://doi.org/10.1093/annonc/mdv279 63. Chaib E, Kanas AF, Galvão FHF, D’Albuquerque LAC (2014) 79. Buettner S, van Vugt JLA, Gani F et al (2016) A comparison of Bile duct confluence: anatomic variations and its classification. prognostic schemes for perihilar cholangiocarcinoma. J Surg Radiol Anat 36:105–109. https://doi.org/10.1007/s00276- Gastrointest Surg 20:1716–1724. https://doi.org/10.1007/s11605- 013-1157-6 016-3203-2 64. Van Gulik TM, Dinant S, Busch ORC et al (2007) Original article: 80. Groot Koerkamp B, Wiggers JK, Allen PJ et al (2014) American new surgical approaches to the Klatskin tumour. Aliment joint committee on cancer staging for resected perihilar cholangio- Pharmacol Ther 26(Suppl 2):127–132. https://doi.org/10.1111/j. carcinoma: a comparison of the 6th and 7th editions. HPB 16: 1365-2036.2007.03485.x 1074–1082. https://doi.org/10.1111/hpb.12320 65. Wiggers JK, te Riele WW, van Dongen TH et al (2016) Combined 81. Deoliveira ML, Schulick RD, Nimura Y et al (2011) New staging liver and extrahepatic bile duct resection for biliary invasion of system and a registry for perihilar cholangiocarcinoma. colorectal metastasis: a case-cohort analysis and systematic re- Hepatology 53:1363–1371. https://doi.org/10.1002/hep.24227 view. HepatoBiliary Surg Nutr 5(4):350–357. https://doi.org/10. 82. Ding G, Yang Y, Cao L et al (2015) A modified Jarnagin-Blumgart 21037/hbsn.2016.05.01 classification better predicts survival for resectable hilar cholan- 66. Olthof SC, Othman A, Clasen S et al (2016) Imaging of cholan- giocarcinoma. World J Surg Oncol 13:1–9. https://doi.org/10. giocarcinoma. Visc Med 32:402–410. https://doi.org/10.1159/ 1186/s12957-015-0526-5 83. Nagino M (2011) Perihilar cholangiocarcinoma: a much needed 67. Blackbourne L, Earnhardt R, Sistrom C et al (1994) The sensitiv- but imperfect new staging system. Nat Rev Gastroenterol Hepatol ity and role of ultrasound in the evaluation of biliary obstruction. 8:252–253. https://doi.org/10.1038/nrgastro.2011.67 Am Surg 60:683–690 84. Steyerberg EW, Vergouwe Y (2014) Towards better clinical pre- 68. Ruys AT, Busch OR, Rauws EA et al (2013) Prognostic impact of diction models: seven steps for development and an ABCD for preoperative imaging parameters on resectability of hilar cholan- validation. Eur Heart J 35:1925–1931. https://doi.org/10.1093/ giocarcinoma. HPB Surg 2013:657309. https://doi.org/10.1155/ eurheartj/ehu207 2013/657309 85. Iacono C, Ruzzenente A, Campagnaro T, Bortolasi L, 69. Engelbrecht MR, Katz SS, Van Gulik TM et al (2015) Imaging of Valdegamberi A, Guglielmi A (2013) Role of preoperative biliary perihilar cholangiocarcinoma. Am J Roentgenol 204:782–791. drainage in jaundiced patients who are candidates for https://doi.org/10.2214/AJR.14.12830 pancreatoduodenectomy or hepatic role of preoperative biliary 70. Mar WA, Shon AM, Lu Y et al (2016) Imaging spectrum of chol- drainage in jaundiced patients who arecandidates for angiocarcinoma: role in diagnosis, staging, and posttreatment pancreatoduodenectomy or hepatic resection: highlights and evaluation. Abdom Radiol 41:553–567. https://doi.org/10.1007/ drawbacks. Ann Surg 257(2):191–204. https://doi.org/10.1097/ s00261-015-0583-9 SLA.0b013e31826f4b0e 71. Choi YH, Lee JM, Lee JY et al (2008) Biliary malignancy: value 86. Ten Hoopen-Neumann H, Gerhards MF, Van Gulik TM et al of arterial, pancreatic, and hepatic phase imaging with (1999) Occurrence of implantation metastases after resection of multidetector-row computed tomography. J Comput Assist klatskin tumors. Dig Surg 16:209–213. https://doi.org/10.1159/ Tomogr 32:362–368. https://doi.org/10.1097/RCT. 0b013e318126c134 87. Martel G, Cieslak KP, Huang R et al (2015) Comparison of tech- 72. Hee SP, Lee JM, Choi JY et al (2008) Preoperative evaluation of niques for volumetric analysis of the future liver remnant: impli- b i l e du ct ca nc e r : M R I comb ine d w i t h M R cations for major hepatic resections. HPB 17:1051–1057. https:// cholangiopancreatography versus MDCT with direct cholangiog- doi.org/10.1111/hpb.12480 raphy. Am J Roentgenol 190:396–405. https://doi.org/10.2214/ 88. Wiggers JK, Koerkamp BG, Cieslak KP et al (2016) AJR.07.2310 Postoperative mortality after liver resection for perihilar 73. Yeh BM, Liu PS, Soto JA et al (2009) MR imaging and CT of the cholangiocarcinoma: development of a risk score and im- biliary tract. Radiographics 29:1669–1688. https://doi.org/10. portance of biliary drainage of the future liver remnant. J 1148/rg.296095514 Am Coll Surg 223:321–331.e1. https://doi.org/10.1016/j. 74. Ruys AT, Bennink RJ, Van Westreenen HL et al (2011) FDG- jamcollsurg.2016.03.035 positron emission tomography/computed tomography and stan- 89. Hammond JS, Guha IN, Beckingham IJ, Lobo DN (2011) dardized uptake value in the primary diagnosis and staging of hilar Prediction, prevention and management of postresection liver fail- cholangiocarcinoma. HPB 13:256–262. https://doi.org/10.1111/j. ure. Br J Surg 98:1188–1200. https://doi.org/10.1002/bjs.7630 1477-2574.2010.00280.x 90. Van Den Broek E, Dijkstra MJJ, Krijgsman O et al (2015) High 75. Annunziata S, Pizzuto D, Caldarella C et al (2014) Diagnostic prevalence and clinical relevance of genes affected by chromo- accuracy of Fluorine-18-Fluorodeoxyglucose positron emission somal breaks in colorectal cancer. PLoS One 10:1–14. https:// tomography in the evaluation of the primary tumor in patients with doi.org/10.1371/journal.pone.0138141 cholangiocarcinoma: a meta-analysis. Eur J Nucl Med Mol 91. Cieslak KP, Bennink RJ, de Graaf W et al (2016) Measurement of Imaging 41:S488. https://doi.org/10.1007/s00259-014-2901-9 liver function using hepatobiliary scintigraphy improves risk as- 76. Chaiteerakij R, Harmsen WS, Marrero CR et al (2015) A new sessment in patients undergoing major liver resection. HPB 18: clinically based staging system for perihilar cholangiocarcinoma. 773–780. https://doi.org/10.1016/j.hpb.2016.06.006 Am J Gastroenterol 2014:1881–1890. https://doi.org/10.1038/ajg. 92. Cieslak KP, Runge JH, Heger M et al (2014) New perspectives in 2014.327.A the assessment of future remnant liver. Dig Surg 31:255–268. 77. Zaydfudim VM, Clark CJ, Kendrick ML et al (2013) Correlation https://doi.org/10.1159/000364836 of staging systems to survival in patients with resected hilar chol- 93. Shoup M, Gonen M, D’Angelica M et al (2003) Volumetric anal- angiocarcinoma. Am J Surg 206:159–165. https://doi.org/10. ysis predicts hepatic dysfunction in patients undergoing major 1016/j.amjsurg.2012.11.020 liver resection. J Gastrointest Surg 7:325–330. https://doi.org/10. 78. Groot Koerkamp B, Wiggers JK, Gonen M et al (2015) Survival 1016/S1091-255X(02)00370-0 after resection of perihilar cholangiocarcinoma-development and 306 Langenbecks Arch Surg (2018) 403:289–307 94. Vauthey JN, Abdalla EK, Doherty DA et al (2002) Body surface Endosc Other Interv Tech 31:422–429. https://doi.org/10.1007/ area and body weight predict total liver volume in western adults. s00464-016-4993-8 Liver Transpl 8:233–240. https://doi.org/10.1053/jlts.2002.31654 111. Fang Y, Gurusamy KS, Wang Q et al (2013) Meta-analysis of 95. Ribero D, Chun YS, Vauthey JN (2008) Standardized liver randomized clinical trials on safety and efficacy of biliary drainage volumetry for portal vein embolization. Semin Intervent Radiol before surgery for obstructive jaundice. Br J Surg. 100:1589– 25:104–109. https://doi.org/10.1055/s-2008-1076681 1596. https://doi.org/10.1002/bjs.9260 96. Dinant S, de Graaf W, Verwer BJ et al (2007) Risk assessment of 112. Celotti A, Solaini L, Montori G et al (2017) Preoperative biliary posthepatectomy liver failure using hepatobiliary scintigraphy and drainage in hilar cholangiocarcinoma: systematic review and me- CT volumetry. J Nucl Med 48:685–692. https://doi.org/10.2967/ ta-analysis. Eur J Surg Oncol 43:4–11. https://doi.org/10.1016/j. jnumed.106.038430 ejso.2017.04.001 97. de Graaf W, van Lienden KP, Dinant S et al (2010) Assessment of 113. Wiggers JK, Coelen RJS, Rauws EAJ et al (2015) Preoperative future remnant liver function using hepatobiliary scintigraphy in endoscopic versus percutaneous transhepatic biliary drainage in patients undergoing major liver resection. J Gastrointest Surg 14: potentially resectable perihilar cholangiocarcinoma 369–378. https://doi.org/10.1007/s11605-009-1085-2 (DRAINAGE trial): design and rationale of a randomized con- 98. Bennink RJ, Dinant S, Erdogan D et al (2004) Preoperative as- trolled trial. Bmc gastroenterol 14:1–8. https://doi.org/10.1186/ sessment of postoperative remnant liver function using s12876-015-0251-0 hepatobiliary scintigraphy. J Nucl Med 45:965–971 114. Komaya K, Ebata T, Yokoyama Y et al (2017) Verification of the 99. de Graaf W, van Lienden KP, van Gulik TM, Bennink RJ (2010) oncologic inferiority of percutaneous biliary drainage to endo- scopic drainage: a propensity score matching analysis of resectable 99mTc-Mebrofenin hepatobiliary scintigraphy with SPECT for the assessment of hepatic function and liver functional volume perihilar cholangiocarcinoma. Surgery (United States) 161:394– before partial hepatectomy. J Nucl Med 51:229–236. https://doi. 404. https://doi.org/10.1016/j.surg.2016.08.008 org/10.2967/jnumed.109.069724 115. Nimura Y (2008) Preoperative biliary drainage before resection 100. Du Bois D, Du Bois E (1989) A formula to estimate the approx- for cholangiocarcinoma (pro). HPB 10:130–133. https://doi.org/ imate surface area if height and weight be known. Nutrition 5: 10.1080/13651820801992666 303–311 116. Kawakami H, Kuwatani M, Onodera M et al (2011) Endoscopic nasobiliary drainage is the most suitable preoperative biliary drain- 101. Erdogan D, Heijnen BHM, Bennink RJ et al (2004) Preoperative assessment of liver function: a comparison of 99mTc-Mebrofenin age method in the management of patients with hilar cholangio- scintigraphy with indocyanine green clearance test. Liver Int 24: carcinoma. J Gastroenterol 46:242–248. https://doi.org/10.1007/ 117–123. https://doi.org/10.1111/j.1478-3231.2004.0901.x s00535-010-0298-1 102. De Graaf W, Häusler S, Heger M et al (2011) Transporters in- 117. Kawashima H, Itoh A, Ohno E et al (2012) Preoperative endo- scopic nasobiliary drainage in 164 consecutive patients with volved in the hepatic uptake of 99mTc-mebrofenin and indocya- nine green. J Hepatol 54:738–745. https://doi.org/10.1016/j.jhep. suspected perihilar cholangiocarcinoma. Ann Surg 257:1. https:// 2010.07.047 doi.org/10.1097/SLA.0b013e318262b2e9 103. Schaap FG, van der Gaag NA, Gouma DJ, Jansen PLM (2009) 118. Farges O, Regimbeau JM, Fuks D et al (2013) Multicentre High expression of the bile salt-homeostatic hormone fibroblast European study of preoperative biliary drainage for hilar cholan- growth factor 19 in the liver of patients with extrahepatic chole- giocarcinoma. Br J Surg 100:274–283. https://doi.org/10.1002/ stasis. Hepatology 49:1228–1235. https://doi.org/10.1002/hep. bjs.8950 22771 119. De Graaf W, Van Lienden KP, Van Den Esschert JW et al (2011) 104. Keppler D (2014) Special section on transporters in toxicity and Increase in future remnant liver function after preoperative portal disease—minireview the roles of MRP2, MRP3, OATP1B1 and vein embolization. Br J Surg 98:825–834. https://doi.org/10.1002/ OATP1B3 in conjugated hyperbilirubinemia. Drug Metab Dispos bjs.7456 42(4):561–565. https://doi.org/10.1124/dmd.113.055772 120. Van Gulik TM, Van Den Esschert JW, De Graaf W et al (2009) 105. Gaag NA, Van Der KJJ, de Castro SMM (2009) Preoperative Controversies in the use of portal vein embolization. Dig Surg 25: biliary drainage in patients with obstructive jaundice: history and 436–444. https://doi.org/10.1159/000184735 current status. J Gastrointest Surg 13:814–820. https://doi.org/10. 121. Olthof PB, Wiggers JK, Groot Koerkamp B et al (2017) 1007/s11605-008-0618-4 Postoperative liver failure risk score: identifying patients with 106. Kennedy TJ, Yopp A, Qin Y et al (2009) Role of preoperative resectable perihilar cholangiocarcinoma who can benefit from biliary drainage of liver remnant prior to extended liver resection portal vein embolization. J Am Coll Surg 225(3):387–394. for hilar cholangiocarcinoma. HPB (Oxford) 11:445–451. https:// https://doi.org/10.1016/j.jamcollsurg.2017.06.007 doi.org/10.1111/j.1477-2574.2009.00090.x 122. Hayashi S, Baba Y, Ueno K et al (2007) Acceleration of primary 107. Al Mahjoub A, Menahem B, Fohlen A et al (2017) Preoperative liver tumor growth rate in embolized hepatic lobe after portal vein biliary drainage in patients with resectable perihilar cholangiocar- embolization. Acta Radiol 48:721–727. https://doi.org/10.1080/ cinoma: is percutaneous transhepatic biliary drainage safer and 02841850701424514 more effective than endoscopic biliary drainage? A meta-analysis. 123. Glantzounis GK, Tokidis E, Basourakos SP et al (2017) The role J Vasc Interv Radiol 28:576–582. https://doi.org/10.1016/j.jvir. of portal vein embolization in the surgical management of primary 2016.12.1218 hepatobiliary cancers. A systematic review. Eur J Surg Oncol 43: 108. Wiggers JK, Koerkamp BG, Coelen RJ et al (2015) Preoperative 32–41. https://doi.org/10.1016/j.ejso.2016.05.026 biliary drainage in perihilar cholangiocarcinoma: identifying pa- 124. Huisman F, Cieslak KP, Van Lienden KP, Bennink RJ, Van Gulik tients who require percutaneous drainage after failed endoscopic TM (2017) Liver related complications in unresectable disease drainage. Endoscopy 47:1124–1131 after portal vein embolization. Hepatobiliary Surg Nutr 6:379– 109. Farges O, Regimbeau JM, Fuks D et al (2013) Multicentre 386. European study of preoperative biliary drainage for hilar cholan- 125. Wiggers JK, Groot Koerkamp B, Coelen RJ et al (2015) giocarcinoma. Br J Surg. 100:274–283. https://doi.org/10.1002/ Percutaneous preoperative biliary drainage for resectable perihilar bjs.8950 cholangiocarcinoma: no association with survival and no increase 110. Jo JH, Chung MJ, Han DH et al (2017) Best options for preoper- in seeding metastases. Ann Surg Oncol 22:1156–1163. https://doi. ative biliary drainage in patients with Klatskin tumors. Surg org/10.1245/s10434-015-4676-z Langenbecks Arch Surg (2018) 403:289–307 307 126. Heimbach JK, Sanchez W, Rosen CB, Gores GJ (2011) Trans- intent resection. J Am Coll Surg 8:583–592. https://doi.org/10. peritoneal fine needle aspiration biopsy of hilar cholangiocarcino- 1002/aur.1474.Replication ma is associated with disease dissemination. HPB 13:356–360. 143. Van Riel WG, Van Golen RF, Reiniers MJ et al (2016) How much https://doi.org/10.1111/j.1477-2574.2011.00298.x ischemia can the liver tolerate during resection? Hepatobiliary 127. Grendar J, Grendarova P, Sinha R, Dixon E (2014) Neoadjuvant Surg Nutr 5(1):58–71. https://doi.org/10.3978/j.issn.2304-3881. therapy for downstaging of locally advanced hilar cholangiocarci- 2015.07.05 noma: a systematic review. HPB 16:297–303. https://doi.org/10. 144. van de Kerkhove MP, de Jong KP, Rijken AM et al (2003) MARS 1111/hpb.12150 treatment in posthepatectomy liver failure. Liver Int 23:44–51 128. Nishio H, Nagino M, Nimura Y (2005) Surgical management of 145. Rocha FG, Matsuo K, Blumgart LH, Jarnagin WR (2010) Hilar hilar cholangiocarcinoma: the Nagoya experience. HPB 7:259– cholangiocarcinoma: the Memorial Sloan-Kettering Cancer 262. https://doi.org/10.1080/13651820500373010 Center experience. J Hepatobiliary Pancreat Sci 17:490–496. 129. Van Gulik TM, Ruys AT, Busch ORC et al (2011) Extent of liver https://doi.org/10.1007/s00534-009-0205-4 resection for hilar cholangiocarcinoma (klatskin tumor): how 146. Jarnagin WR, Fong Y, DeMatteo RP et al (2001) Staging, resect- much is enough? Dig Surg 28:141–147. https://doi.org/10.1159/ ability, and outcome in 225 patients with hilar cholangiocarcino- ma. Ann Surg 234:507–517–9. https://doi.org/10.1097/00000658- 130. Olthof PB, Coelen RJS, Wiggers JK et al (2016) External biliary 200110000-00010 drainage following major liver resection for perihilar cholangio- 147. De Jong MC, Marques H, Clary BM et al (2012) The impact of carcinoma: impact on development of liver failure and biliary portal vein resection on outcomes for hilar cholangiocarcinoma: a leakage. HPB 18:1–6. https://doi.org/10.1016/j.hpb.2015.11.007 multi-institutional analysis of 305 cases. Cancer 118(19):4737– 131. Neuhaus P, Jonas S, Bechstein WO et al (1999) Extended resec- 4747. https://doi.org/10.1002/cncr.27492 tions for hilar cholangiocarcinoma. Ann Surg 230:808–818 dis- 148. Koch M, Garden OJ, Padbury R et al (2011) Bile leakage after cussion 819 hepatobiliary and pancreatic surgery: a definition and grading of 132. Neuhaus P, Thelen A, Jonas S et al (2012) Oncological superiority severity by the International Study Group of Liver Surgery. of hilar en bloc resection for the treatment of hilar cholangiocar- Surgery 149:680–688. https://doi.org/10.1016/j.surg.2010.12.002 cinoma. Ann Surg Oncol 19:1602–1608. https://doi.org/10.1245/ 149. Correa-Gallego C, Maddalo D, Doussot A et al (2016) Circulating s10434-011-2077-5 plasma levels of MicroRNA-21 and MicroRNA-221 are potential 133. Nagino M, Nimura Y, Nishio H et al (2010) Hepatectomy with diagnostic markers for primary intrahepatic cholangiocarcinoma. simultaneous resection of the portal vein and hepatic artery for PLoS One 11:1–16. https://doi.org/10.1371/journal.pone.0163699 advanced perihilar cholangiocarcinoma: an audit of 50 consecu- 150. Feller SM, Lewitzky M (2016) Hunting for the ultimate liquid tive cases. Ann Surg 252:115–123. https://doi.org/10.1097/SLA. cancer biopsy—let the TEP dance begin. Cell Commun Signal 0b013e3181e463a7 14:24. https://doi.org/10.1186/s12964-016-0147-9 134. Bhangui P, Salloum C, Lim C et al (2014) Portal vein 151. Dekker AM, Wiggers JK, Coelen RJ et al (2016) Perioperative arterialization: a salvage procedure for a totally de-arterialized blood transfusion is not associated with overall survival or time to liver. The Paul Brousse Hospital experience. HPB 16:723–738. recurrence after resection of perihilar cholangiocarcinoma. HPB https://doi.org/10.1111/hpb.12200 18:262–270. https://doi.org/10.1016/j.hpb.2015.08.004 135. Dinant S, Gerhards MF, Busch ORC et al (2005) The importance of complete excision of the caudate lobe in resection of hilar chol- 152. Schnitzbauer AA, Lang SA, Goessmann H et al (2012) Right angiocarcinoma. HPB (Oxford) 7:263–267. https://doi.org/10. portal vein ligation combined with in situ splitting induces rapid 1080/13651820500372376 left lateral liver lobe hypertrophy enabling 2-staged extended right 136. Uesaka K (2012) Left hepatectomy or left trisectionectomy with hepatic resection in small-for-size settings. Ann Surg 255:405– resection of the caudate lobe and extrahepatic bile duct for hilar 414. https://doi.org/10.1097/SLA.0b013e31824856f5 cholangiocarcinoma (with video). J Hepatobiliary Pancreat Sci 19: 153. Gores GJ, Murad SD, Heimbach JK, Rosen CB (2013) Liver 195–202. https://doi.org/10.1007/s00534-011-0474-6 transplantation for perihilar cholangiocarcinoma. Dig Dis 31: 137. Bismuth H (1982) Surgical anatomy and anatomical surgery of the 126–129. https://doi.org/10.1159/000347207 liver. World J Surg 6:3–9. https://doi.org/10.1007/BF01656368 154. Skipworth JR, Olde Damink SWM, Imber C et al (2011) Review 138. Oldhafer KJ, Stavrou GA, van Gulik TM (2016) ALPPS—where article: surgical, neo-adjuvant and adjuvant management strategies do we stand, where do we go? Ann Surg 263:1. https://doi.org/10. in biliary tract cancer. Aliment Pharmacol Ther 34:1063–1078. 1097/SLA.0000000000001633 https://doi.org/10.1111/j.1365-2036.2011.04851.x 139. Lang H, de Santibanes E, Clavien PA (2017) Outcome of ALPPS 155. Bird N, Elmasry M, Jones R et al (2017) Role of staging laparos- for perihilar cholangiocarcinoma: case-control analysis including copy in the stratification of patients with perihilar cholangiocarci- the first series from the international ALPPS registry. HPB 19: noma. Br J Surg 104:418–425. https://doi.org/10.1002/bjs.10399 379–380. https://doi.org/10.1016/j.hpb.2017.01.024 156. Van Lienden KP, Van Den Esschert JW, De Graaf W et al (2013) 140. Olthof PB, Coelen RJS, Wiggers JK et al (2017) High mortality Portal vein embolization before liver resection: a systematic re- after ALPPS for perihilar cholangiocarcinoma: case-control anal- view. Cardiovasc Intervent Radiol 36:25–34. https://doi.org/10. ysis including the first series from the international ALPPS regis- 1007/s00270-012-0440-y try. HPB 19:381–387. https://doi.org/10.1016/j.hpb.2016.10.008 157. Dumitrascu T, Brasoveanu V, Stroescu C, Ionescu M, Popescu I 141. Li J, Ewald F, Gulati A, Nashan B (2016) Associating liver parti- (2015) Major hepatectomies for perihilar cholangiocarcinoma: tion and portal vein ligation for staged hepatectomy: from techni- predictors for clinically relevant postoperative complications cal evolution to oncological benefit. World J Gastrointest Surg 8: using the international study Group of Liver Surgery definitions. 124–133. https://doi.org/10.4240/wjgs.v8.i2.124 Asian J Surg 39(2):1–9. https://doi.org/10.1016/j.asjsur.2015.04. 142. Groot Koerkamp B, Wiggers JK, Allen PJ et al (2016) Recurrence rate and pattern of perihilar cholangiocarcinoma after curative http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Langenbeck's Archives of Surgery Springer Journals

Modern work-up and extended resection in perihilar cholangiocarcinoma: the AMC experience

Free
19 pages

Loading next page...
 
/lp/springer_journal/modern-work-up-and-extended-resection-in-perihilar-cholangiocarcinoma-IvppHVNwRT
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Medicine & Public Health; General Surgery; Abdominal Surgery; Cardiac Surgery; Thoracic Surgery; Traumatic Surgery; Vascular Surgery
ISSN
1435-2443
eISSN
1435-2451
D.O.I.
10.1007/s00423-018-1649-2
Publisher site
See Article on Publisher Site

Abstract

Aim Perihilar cholangiocarcinoma (PHC) is a challenging disease and requires aggressive surgical treatment in order to achieve curation. The assessment and work-up of patients with presumed PHC is multidisciplinary, complex and requires extensive experience. The aim of this paper is to review current aspects of diagnosis, preoperative work-up and extended resection in patients with PHC from the perspective of our own institutional experience with this complex tumor. Methods We provided a review of applied modalities in the diagnosis and work-up of PHC according to current literature. All patients with presumed PHC in our center between 2000 and 2016 were identified and described. The types of resection, surgical techniques and outcomes were analyzed. Results and conclusion Upcoming diagnostic modalities such as Spyglass and combinations of serum biomarkers and molecular markers have potential to decrease the rate of misdiagnosis of benign, inflammatory disease. Assessment of liver function with hepatobiliary scintigraphy provides better information on the future remnant liver (FRL) than volume alone. The selective use of staging laparoscopy is advisable to avoid futile laparotomies. In patients requiring extended resection, selective preoperative biliary drainage is mandatory in cholangitis and when FRL is small (< 50%). Preoperative portal vein embolization (PVE) is used when FRL volume is less than 40% and optionally includes the left portal vein branches to segment 4. Associating liver partition and portal vein ligation for staged hepatectomy (ALPPS) as alternative to PVE is not recommended in PHC. N2 positive lymph nodes preclude long-term survival. The benefit of unconditional en bloc resection of the portal vein bifurcation is uncertain. Along these lines, an aggressive surgical approach encompassing extended liver resection including segment 1, regional lymph- adenectomy and conditional portal venous resection translates into favorable long-term survival. . . . . . . Keywords Perihilar cholangiocarcinoma Klatskin tumor Diagnosis Staging Biomarkers Preoperative assessment . . . Hepato-biliary scintigraphy Biliary drainage Surgical resection Postoperative outcome Introduction the epithelium of the biliary tract and may occur in the whole biliary ductal system. They are sub-classified ac- Cholangiocarcinoma accounts for 3% of all gastrointes- cording to their location, in intrahepatic, perihilar and tinal malignancies worldwide [1]. The tumors arise from distal cholangiocarcinoma [2]. Each entity comes with F. Rassam and E. Roos contributed equally to this work. * F. Rassam Department of Gastroenterology & Hepatology and Tytgat Institute f.rassam@amc.uva.nl for Liver and Intestinal Research, Academic Medical Center, Amsterdam, The Netherlands 1 Department of Medical Oncology, Academic Medical Center, Department of Surgery, Academic Medical Center, Amsterdam, The Netherlands Amsterdam, The Netherlands Department of Radiotherapy, Academic Medical Center, Amsterdam, The Netherlands Department of Radiology and Nuclear Medicine, Academic Medical Department of Pathology, Academic Medical Center, Center, Amsterdam, The Netherlands Amsterdam, The Netherlands 290 Langenbecks Arch Surg (2018) 403:289–307 a specific set of problems and therefore, management finally develops due to local biliary obstruction, patients are requires a tailored approach. often not resectable anymore, and thus not curable. Up to 65– Perihilar cholangiocarcinoma (PHC), also known as 80% of patients have initially unresectable disease due to ex- Klatskin tumor, is the most frequent biliary tract tumor and tensive hepatic artery and/or portal vein infiltration by tumor accounts for approximately 60% of all cholangiocarcinoma’s or distant metastases at time of presentation [10–13]. Of all [3]. This tumor originates in the extrahepatic biliary tract prox- patients who in time undergo a laparotomy, 40–70% ultimate- imal to the origin of the cystic duct, up until the second-degree ly have resectable disease [14–16]. Patients face many obsta- bile ducts. PHC can be subdivided according to proximal ex- cles during diagnosis and work-up for extended resection. tent of the tumor into the bile ducts (Bismuth-Corlette These problems range from confirmation of malignancy to classification) [4](Fig. 1). cholestasis and cholangitis due to biliary obstruction, requir- The incidence of cholangiocarcinoma varies widely be- ing biliary drainage. tween regions. In Asian populations and Chili, parasitic infec- Oncological outcomes depend heavily on the possibility of tions are strongly associated with PHC, showing a peak inci- performing a radical resection. Patients with unresectable dis- dence in Thailand of 87 per 100,000 [3, 5–7]. In Western ease, receiving palliative chemotherapy with gemcitabine and populations the incidence is considerably lower, 1–2per cisplatin, have an overall median survival of approximately 100,000, and PHC is mainly associated with primary scleros- 12 months [17, 18]. In contrast, median survival of patients ing cholangitis (PSC) [8, 9]. with an R0 resection is 30–46 months and 5-year survival Early symptoms are not specific and patients typically pres- rates range from 25 to 40% [19, 20]. The aggressive surgical ent with the sequelae of biliary obstruction. When jaundice approach necessary to achieve an R0 resection however, is Fig. 1 Bismuth-Corlette classification for staging of perihilar cholangiocarcinoma Langenbecks Arch Surg (2018) 403:289–307 291 associated with significant postoperative morbidity and mor- metastases (n = 68) or unfitness for major resection (n = tality with reported morbidity rates ranging from 60 to 70% 53) (Table 1). [21] and mortality rates as high as 5–18% [19, 22–24]. It is The remaining 321 (53.0%) patients underwent laparoto- therefore crucial to optimize patients before exposing them to my; 120 (19.8%) patients were deemed unresectable on the this high-risk surgery. basis of intraoperative findings. The main reasons were locally The aim of this review is to elaborate current diagnosis and advanced disease (n = 43), N2 lymph node metastases, (n = work-up and to review the issues of extended resection in 39), liver metastases (n = 11), peritoneal or other distant me- patients presenting with a hilar lesion suspicious of PHC, from tastases (n = 26) or major liver resection precluded by comor- the perspective of the long-standing experience with this com- bidities (n =1) (Table 1). plex tumor in our referral center. A total of 201 patients underwent extrahepatic bile duct resection in the majority of cases combined with (extended) The AMC experience; the denominator of patients liver resection. Of these patients, 66 (32.8%) underwent a left referred with (suspected) PHC hemihepatectomy, 8 (4.0%) underwent an extended left hemihepatectomy, 31 (15.4%) patients underwent a right Between 2000 and 2016, a total of 606 patients with lesions hemihepatectomy, 51 (25.4%) underwent an extended right suspicious of PHC have been referred to our center. Patients hemihepatectomy, 8 (4.0%) patients underwent resection of were discussed in our HPB oncology multidisciplinary meet- 1 or 2 segments and the remainder of 37 (18.4%) patients ing, consisting of experienced hepatobiliary surgeons, dedi- underwent bile duct resection alone (Table 2). cated endoscopists, (interventional, abdominal and nuclear) Based on pathological examination of the resection speci- radiologists, radiotherapists, nurse practitioners, medical on- mens, 170 (84.6%) patients had PHC and 31 (15.4%) had cologists and pathologists. benign disease (either unspecified sclerosing cholangitis or A total of 285 (47.0%) patients were deemed IgG4-associated cholangitis) (Fig. 2). unresectable, of which 228 (37.6%) were found to be Severe complications (Clavien-Dindo grade 3 or higher) unresectable at initial presentation on the basis of imag- were observed in 93(46.3%) of resected patients. Of all pa- ing studies. The remaining 57 patients were staged with tients who underwent resection, 18 (9.0%) died within the first unresectable disease after diagnostic laparoscopy 90 days. (Fig. 2). The main reason for unresectability was locally The median survival after resection of confirmed malignan- advanced disease (n = 104), N2 lymph node metastases cy was 52.6 months. The 5-year survival after resection was (n = 29), liver metastases (n = 27), peritoneal or distant 44.3% (Fig. 3). Fig. 2 Flow diagram of patients referred to the AMC with suspicion on PHC (2000–2016) 292 Langenbecks Arch Surg (2018) 403:289–307 Table 1 Reasons for unresectability in patients referred with PHC Reason for unresectability Patients n (%) Initially unresectable 285 After imaging/laboratory assessment 228 After diagnostic laparoscopy 57 Locally advanced disease 104 (36.5%) LN metastases 29 (10.2%) Liver metastases 27 (9.5%) Peritoneal/distant metastases 68 (23.9%) Unfit for surgery 53 (18.6%) Missing 4 (1.4%) Unresectable during laparotomy 120 Locally advanced disease 43 (35.8%) LN metastases 39 (32.5%) Liver metastases 11 (9.2%) Fig. 3 Overall survival in 170 patients undergoing resection of pathology Peritoneal/distant metastases 26 (21.7%) proven PHC in the AMC. The 5-year survival rate after resection was Unfit for surgery 1 (0.8%) 44.3% diagnostic modality for years [31–33]. PHC, however, fre- Pitfalls of diagnosis quently shows a submucosal growth pattern resulting in a low sensitivity of brush cytology of 27–56% [31, 34, 35]. Differentiation between malignant and benign disease Various techniques have been investigated to increase sensi- tivity of cytological samples. These include fluorescence in In patients with a presumed PHC, it is highly desirable to obtain situ hybridization (FISH) which is reported to increase sensi- a definitive diagnosis (Fig. 4). Benign biliary tract strictures are tivity to 69–93% [36–38]. Mutation analysis has not been difficult to differentiate from malignant disease [25–27]. In re- used widely, but seems mainly to increase specificity [39]. cent years, IgG4-associated cholangitis (IAC) has been identi- The use of stiffer bristles or repeated brushings also has not fied as a disease entity that may mimic PHC, both clinically as increased the diagnostic yield of brush cytology [40–42]. on imaging studies. It belongs to the spectrum of IgG4-related Other endoscopic techniques have emerged as well. The use disease, a systemic disease which can affect many other organs of endobiliary forceps biopsy during ERCP resulted in a as well [28–30]. Of all resections for presumed PHC worldwide, higher detection rate ranging from 44 to 89% [43, 44]. The 8–22% of patients turned out to have a benign disease on mi- technique however is challenging, especially in more proxi- croscopical examination of the resection specimen [26]. mal lesions as it is difficult to navigate and position the for- ceps. Consequently, it has not found wide application [44–46]. Endoscopic techniques Endoscopic retrograde Cholangioscopy offers direct visualization of biliary stric- cholangiopancreatography (ERCP) combined with brush- tures and seems to improve the diagnostic yield of routine cytology for microscopical examination has been the standard cytology. Percutaneous cholangioscopy requires percutaneous biliary access and multiple dilatations to allow access of the Table 2 Types of resection undertaken in 201 patients with presumed cholangioscope. Single operator cholangioscopy (Spyglass, PHC Boston scientific, Natick, MA, US) is introduced through a Type of resection Patients n (%) duodenoscope and is used in combination with Spybite Biopsy Forceps [47, 48]. Using these techniques, the sensitiv- Total number of patients 201 ity in diagnosis of biliary strictures has increased to 74.7% Left hemihepatectomy 66 (32.8%) [49]. Cholangioscopy enables targeted biopsies increasing Right hemihepatectomy 8 (4.0%) sensitivity and specificity to detect PHC to 66 and 97%, re- Extended left hemihepatectomy 31 (15.4%) spectively, in a meta-analysis. Single operator cholangioscopy Extended right hemihepatectomy 51 (25.4%) seems a useful new step in centers experienced with ERCP Segmentectomy (≤ 3 Couinaud segments) 8 (4.0%) and brush cytology. Only local excision of hilar bile ducts 37 (18.4%) Alternatively, intraductal ultrasound (IDUS) enables de- Including portal vein resection 30/151 (19.9%), 50 missing tailed imaging of the bile ducts and periductal tissue. IDUS Langenbecks Arch Surg (2018) 403:289–307 293 Fig. 4 Flowchart showing work- Suspected PHC up and treatment of patients suspected of PHC Biomarkers Benign CT/MRI/MRCP IAC assessment Assessment of resectability (Table 3) and staging (Table 4) Bili > 30 μmol/L Potenally respectable Staging laparoscopy Biliary drainage Hepatobiliary scingraphy < 40% FRL CT-volumetry < 2.7%/min/m Preoperave RTx in drained paents PVE Laparotomy Unresectable Resectable Palliave drainage with metal stents Hepatectomy + bile duct resecon Follow up has been reported to improve diagnostic accuracy of ERCP Biomarkers are also needed to monitor patients with an from 58% to 90% [33, 50]. However, stents that are often increased risk of PHC such as in primary sclerosing required to drain obstructed bile ducts make the interpretation cholangitis [56]. The conventional serum markers CA19–9 of IDUS difficult. If this is the case, the use of endoscopic and CEA are frequently used in gastrointestinal malignancies. ultrasonography (EUS) in combination with fine needle aspi- However, the diagnostic value of CA19–9 is debated because ration (FNA) may be preferable [50–52]. These techniques of its variable sensitivity of 33–93% and specificity of 67– however, require specific expertise to reach their maximum 98%. Its use as a prognostic biomarker seems more valuable potential and their success rates must be partially attributed to [57–60]. Furthermore, CA19–9 may be elevated in benign the experience of their users. biliary disease and/or in the presence of cholestasis, impairing its use as a reliable biomarker especially in biliary tumors [61]. Serum markers The limited ability to reliably acquire tissue The same applies for CEA with a sensitivity of 33–84% and samples has resulted in an ongoing quest for serum bio- specificity of 50–88% [58] in pancreato-biliary malignancies. markers. The additional use of serum markers to distinguish IAC from PHC has been an area of extensive research. Serum Staging and resectability IgG4 levels (ULN = 1.4 mg/ml) have limited diagnostic value when only slightly increased, since up to 15% of patients with Criteria for the assessment of resectability PHC have elevated sIgG4 levels as well [53, 54]. Recently a new test has been developed measuring the IgG4/IgG RNA Initial imaging is crucial in establishing diagnosis and in de- ratio. This test distinguishes IAC accurately (94% sensitivity, termining whether a patient is a candidate for resection. The 99% specificity) from PHC and primary sclerosing cholangitis goal of curative resection is to achieve negative margins (R0) [55]. The value of this test awaits further clinical assessment. while preserving sufficient volume and function of remnant 294 Langenbecks Arch Surg (2018) 403:289–307 liver with adequate portal venous and hepatic arterial blood morphological growth patterns that can be recognized on imag- supply. Factors to consider to determine resectability are in- ing to enhance the diagnostic confidence, determine manage- cluded in Table 3 [62]. ment and to provide additional information on prognosis. Unresectability can result from either extensive local disease However, imaging can also lead to confusion due to overlapping (including vascular and nodal involvement), presence of distant appearances with other hepatobiliary diseases, including benign metastases or comorbidity of the patient. Local unresectability lesions. Important conditions to consider are other causes of can be due to involvement of the portal vein and hepatic artery biliary dilatation such as choledocholithiasis, PSC, IAC and on the side of the future remnant liver without the possibility of biliary dilatation due to centrally located colorectal metastases a vascular reconstruction, extensive bilateral proximal infiltra- [65]. Ultrasound is usually the initial test to evaluate patients tion of the tumor into secondary biliary radicles (segmental bile with suspected bile duct obstruction [66, 67], and may provide ducts) and/or massive extension of tumor into the liver paren- informationonthe levelofobstructioninthe biliarytree. chyma. Furthermore, extrahepatic metastases including distant lymph node metastases beyond the hepatoduodenal ligament Cross-sectional studies CT and MRI are commonly used in (N2 nodes) are associated with poor survival and in most cen- various combinations with cholangiographic studies, in the ters, are considered as unresectable as well. diagnosis and preoperative planning of PHC. CT with iv con- It should be emphasized that local resectability depends on trast offers the opportunity to assess full extension of the tu- biliary anatomy at the liver hilum. The hepatic duct confluence mor in detail and determine resectability [68, 69]. If PHC is is defined by the convergence of the right and left hepatic ducts, suspected, imaging is preferably performed before stenting for at which site many anatomic variations exist [63]. In 20% of biliary drainage, since the images will be obscured by the cases, the anterior and posterior sectorial branches of the right plastic or metal stent. In general, PHC may be recognized by ductal system drain directly into the main hepatic duct. This dilated bile ducts, lack of communication between the left and may give rise to confusion as when a hilar tumor involving the right first-order bile ducts, crowding of bile ducts, ductal wall right anterior and posterior sectorial branches in these cases is thickening and enhancement, and lobar atrophy. In some combined with segmental involvement on the left side, the cases, a solid (mass forming) or papillary mass (intraductal tumor is defined by the Bismuth-Corlette classification as type growth type) may be seen. IV, which in many textbooks is considered unresectable. A type The early arterial and late portal venous phases of a CT- IV tumor in this situation however does not preclude a radical scan aid to assess the relationship between tumor and resection using an extended left hepatectomy. The same holds (branches of) the hepatic artery and portal vein, which is im- true for a tumor extending into the right sectorial ducts involv- portant in determining resectability [70, 71]. Key elements for ing a low inserting segment 4 duct of the left biliary system. staging in imaging are defined in Table 4. According to a Although defined as Bismuth-Corlette type IV, this tumor can of course be radically resected using an extended right Table 4 Key elements for staging of PHC hemihepatectomy. Resectability depends on hilar biliary anat- Key elements necessary for staging PHC omy and it is therefore important that resectability is assessed Location of primary tumor by hepatobiliary surgeons with expertise in PHC [64]. Intra- or extrahepatic Proximal common hepatic duct Imaging Confluence of the left and right hepatic duct Left or right hepatic duct Imaging plays a decisive role in the diagnosis, staging and as- Intraductal growth type sessment of resectability. PHC manifests with various Local extension Segmental duct involvement (including Bismuth-Corlette classification) Table 3 Criteria for the assessment of resectability in PHC Mentioning biliary variant anatomy Criteria for the assessment of resectability Vascular involvement (portal vein and/or hepatic arteries, including vascular variations and presence of stenosis of celiac axis or mesenteric Presence of (extra) hepatic metastases artery) Presence of lymph node metastases confined to hepatoduodenal ligament Lymph nodes (N1) or lymph node metastases along the common hepatic artery Regional N1; cystic duct, common bile duct, proper hepatic artery and and/or celiac axis (N2) portal vein nodes Possibility of achieving free ductal margins on the side of the FRL Metastatic N2; common hepatic artery, periaortic, pericaval, superior Involvement of portal vein bifurcation mesenteric or celiac artery nodes Involvement of hepatic artery branches Distant metastasis Volume and function of FRL Noncontiguous liver, peritoneum, bone, other Langenbecks Arch Surg (2018) 403:289–307 295 meta-analysis by Ruys et al., sensitivity and specificity of CT is based on pathology assessment of the resection specimen were 89 and 92% for assessment of portal vein involvement and is mainly used postoperatively as a prognostic tool. The (encasement or occlusion are strong evidence), 84 and 93% Bismuth-Corlette classification system, introduced in 1975, is for hepatic artery involvement and 61 and 88% for lymph used to describe proximal involvement of tumor into the bile node metastases, respectively [68]. ducts [4]. This system is mainly informative to surgeons for MRI with iv contrast provides an acceptable alternative to planning of the type of resection, but does not determine re- CT in the evaluation of PHC. Both CT and MRI have similar sectability since other parameters such as distant metastases staging accuracy, including that of nodal staging [72]. The and vascular involvement are not included. The Blumgart advantage of MRI is that combined with cholangiography classification system takes in addition to bile duct involve- (MRC), it provides anatomical definition of the biliary tree. ment, portal vein involvement and lobar atrophy into account Whether CT or MRI is used should be based on local expertise as well [82]. However, since its introduction in 1998, the in- and accessibility to one of these modalities [73]. dications for (extended) resections have expanded rendering [ F]-FDG PET-CT has no additional value in the diagnosis the Blumgart system now less applicable. The classification and staging of PHC. In the hilar area, it is difficult to distin- system proposed by the International Cholangiocarcinoma guish tumor from concomitant inflammation. Furthermore, Group for the Staging of PHC takes into account most of the for the identification of nodal involvement, [ F]-FDG PET- variables used in the previous systems: suspicious lymph CT has a sensitivity and specificity of 67 and 68%, respective- nodes, extent of bile duct involvement, extent of vascular ly [74, 75]. Hence, it does not provide additional diagnostic involvement, suspected tumor size and lobar atrophy. As in yield in comparison with CT. the other systems, the information is largely descriptive [83]. The staging systems used to date are mainly surgery orient- Cholangiography MR cholangiography (MRC) combined ed. Each has its merits, but all are limited to the anatomical with MRI has comparable staging accuracy with that of CT description of the tumor and are therefore limited in their ability combined with direct cholangiography [72]. Alternatively, di- to predict the likelihood of an R0 resection. Furthermore stag- rect cholangiography using ERCP or percutaneous ing systems have been criticized for having poor predictable transhepatic cholangiography (PTC) can also be used. PTC quality in different populations [20, 79, 84]. Ideally, a staging may be more helpful in assessing the extent of proximal tumor system would preoperatively predict the likelihood of resect- infiltration. ERCP can also be combined with cytological or able disease along with as well, prognostic value. tissue sampling, albeit sensitivity and specificity are low (see above). A major disadvantage of direct cholangiography is its Staging laparoscopy invasiveness, including the risk of inducing infection, pancre- atitis, bleeding, inflammation and pain. Direct cholangiogra- For optimal determination of resectability, patients with po- phy for diagnostic purposes is therefore, only rarely per- tentially resectable PHC may undergo staging laparoscopy to formed. Especially ERCP entails retrograde contamination detect the presence of occult tumor manifestations. Staging of the obstructed bile ducts with increased risk of cholangitis. laparoscopy may detect small liver and/or peritoneal metasta- Subsequent drainage of the visualized bile ducts using one or ses that are undetectable on routine imaging avoiding a futile more stents is therefore mandatory. ERCP and PTC are pref- laparotomy [84–86, 155]. A thorough inspection of the liver, erably used for therapeutic purposes to drain the obstructed gallbladder, hepatoduodenal ligament and peritoneum is un- bile ducts in the palliative setting or preoperatively, to prepare dertaken. The lesser sac is routinely opened and the common the patient for resection. In the latter situation, the aim is to hepatic artery is examined, lymph node station 8 (N2) is then drain the biliary system of the future remnant liver while leav- identified and biopsied for pathological evaluation. All other ing the part to be resected alone. suspicious lesions, based on intraoperative inspection or pre- vious imaging, are biopsied for histopathological analysis. Staging systems Although not widely used, the combination with laparoscopic ultrasound has been reported to increase the yield of the stag- There are many factors associated with resectability, prognosis ing procedure to some extent. In a meta-analysis by Coelen and prediction of long and short-term survival after resection et al., which included 832 potentially resectable PHC patients, of PHC [15, 76–79]. The most commonly used staging sys- a pooled sensitivity of 52.2% was found to detect tems include the American Joint Committee on Cancer unresectability [14]. Based on our own experience in 273 (AJCC) staging system with incorporated TNM classification, patients undergoing staging laparoscopy for PHC, we devel- the Bismuth-Corlette system, the Blumgart T-staging system oped a risk score that estimates the chance of unresectability. (MSKCC classification) and a classification recently proposed This risk score includes the following factors: tumor size, by the International Cholangiocarcinoma Group for the stag- portal vein involvement, suspected lymph-node metastases ing of PHC [14, 15, 20, 77, 79–81]. The AJCC staging system and suspected (extra) hepatic metastases. It showed good 296 Langenbecks Arch Surg (2018) 403:289–307 discrimination between resectable and unresectable disease The FRL is delineated on the SPECT images for calcula- (AUC 0.77, 0.68–0.86 95% CI) [16]. tion of the functional share (%). This is then multiplied by the total liver function (MUR) to calculate function of specifically Assessment of future remnant liver the FRL. This method provides visual and quantitative infor- mation on regional liver function. Functional share of the FRL Liver volumetry is corrected for body surface area (BSA, m )using the Mosteller formula, to individualize the results based on the Since extended liver resections are often required, it is critical individual metabolic needs [100]. The current cutoff for a safe to assess the FRL preoperatively where CT-volumetric analy- resection is a FRL function of at least 2.7%/min/m [97]. sis is the standard technique. The segments of the FRL are HBS can be used in patients with normal or impaired qual- delineated on the CT images and the ratio of the remnant liver ity of liver parenchyma alike using the same cutoff value. and the total liver, with subtraction of tumor volume is calcu- MUR has been shown to correlate well with ICG clearance lated. This delineation technique gives an indirect measure- [101]. A limitation of using HBS in patients with PHC is that ment of the liver function [87, 88]. the uptake of bilirubin is competitive with mebrofenin as both It is assumed that a FRL-volume of > 25–30% is consid- are taken up by the same hepatocyte transporters [102]. In ered a safe cutoff for patients with healthy liver parenchyma, these patients, hepatocyte function is likely to be decreased whereas > 40% is used in patients with compromised liver, which will be reflected by HBS, with additional underestima- like in patients with (post)cholestatic liver that is damaged tion due to competition. These receptors are downregulated by longstanding biliary obstruction and possible cholangitis during hyperbilirubinemia, but their expression gradually nor- [88–90]. In literature the acceptable minimum volume of FRL malizes after drainage [103]. Considering this interaction, in regard with parenchymal disease is variable and controver- HBS should not be performed in patients with high bilirubin sial (10–40%) [91–93]. In PHC, a FRL volume of more than levels (> 30 μmol/L) and is usually postponed until adequate 40% is usually considered. A disadvantage of FRL volumetry biliary drainage has been achieved [104]. is that individual patient characteristics are not taken into ac- count and that the delineating technique is prone to error [94, Preoperative preparation of the patient 95]. Especially in patients with compromised liver, discrepan- cies have been reported between CT volumetry and postoper- Obstructive jaundice and biliary drainage ative outcomes [96] because the quality of the liver parenchy- ma is not taken into consideration [87, 97]. Patients with PHC usually present with obstructive jaundice. This phenomenon has a negative effect on liver function, in- Liver functional tests creases the risk of biliary infection and impairs cellular immu- nity [105]. Preoperative biliary drainage is used to create a Because liver volume does not equal liver function and func- safer environment prior to liver surgery. It reduces jaundice, tion is not homogeneously distributed in the liver [97], we rely improves liver function and the patient’s condition, at the more on assessment of the function of the FRL, rather than on same time improving the ability of the liver to regenerate 99m volume alone. Tc-mebrofenin hepatobiliary scintigraphy postoperatively [85, 105–109]. The impact of these effects is (HBS) is a validated quantitative dynamic liver function test particularly high in patients with an insufficient FLR and pre- for which mebrofenin, an iminodiacetic (IDA) derivate, is operative drainage has shown to improve outcomes especially used as a tracer. This agent is mainly taken up by the hepato- in patients requiring extended resections [108]. On the other cytes and is subsequently excreted in the bile without under- hand, drainage-related complications may be severe and it is going any biotransformation. The hepatic uptake is mediated therefore advisable to only drain patients with a substantially by the same transport mechanisms as that of various endo- and increased bilirubin and small FLR [85, 110]. Drainage-related exogenous substances, making it an ideal agent to assess liver complications such as cholangitis may severely deteriorate a function. HBS consists of an early dynamic phase, acquired patient’s condition and increase the risk of postoperative mor- directly after intravenous injection of mebrofenin, during bidity and mortality [50, 85, 105, 111]. Preoperative which the mebrofenin uptake rate (MUR, %/min) is measured cholangitis is caused by contamination of the biliary tract dur- [98]. This corresponds with the total liver function. ing drainage procedures. It is therefore advisable to give pro- Immediately afterwards, a SPECT acquisition combined with phylactic antibiotics previous to any drainage procedure [3, low-dose CT is made, falling in the period in which 23, 107, 112]. Any episode of cholangitis induced after biliary mebrofenin is accumulated in the liver. The SPECT data pro- decompression should be treated with antibiotics and addi- vide information on three-dimensional, segmental distribution tional drainage or drain revision if necessary [21]. of function. The low-dose CT is solely used for attenuation Refractory cholangitis is often caused by incomplete biliary correction and anatomical mapping [99]. drainage and requires adequate endoscopic or percutaneous Langenbecks Arch Surg (2018) 403:289–307 297 stenting of residual, obstructed parts of the biliary tract. of a range of interleukins and growth factors that induce hy- Patients should not undergo surgery earlier than that they have pertrophy of the non-embolized lobe. In our cohort of PHC fully recovered from cholangitis [21]. patients, the cutoff for proceeding with PVE is a FRL volume The optimal drainage method is still a much-debated topic, of less than 40% and/or function less than 2.7%/min/m .Inthe in which surgeons tend to favor the percutaneous approach for absence of cholangitis, the biliary system of the embolized reasons of direct access to the biliary duct and postoperative use lobe needs not be drained, since unilateral cholestasis may of the intraluminal drain(s) across the hepaticojejunostomy. even have a synergistic effect on the hypertrophy response Although percutaneous biliary drainage (PTBD) seems as- of the non-embolized lobe. After a period of 3 weeks after sociated with higher postoperative morbidity, further prospec- PVE, CT and HBS are repeated and reassessed. We have tive studies are needed to better define the optimal mode of shown that functional increase occurs more rapidly than vol- biliary drainage in PHC [112, 113]. Furthermore, PTBD might ume, suggesting a shorter waiting time until resection can take be complicated by portal vein thrombosis or seeding metasta- place [119]. sis that may change resectability of the tumor [110, 114]. For In the series reported by the Nagoya group, PVE showed to now, endoscopic biliary drainage (EBD) is still the preferred improve the surgical outcomes of PHC [10]. PVE is consid- method in most Western countries. ered a safe procedure with an overall morbidity rate of 2.2%. Endo-nasobilairy drainage (ENBD) is the advocated meth- Most common complications are hematoma, hemobilia, septic od in many Asian countries. As in PTBD, it provides more complications, backflow of embolization material and throm- precise information on the extent of cancer along the bile ducts bosis in the FRL [120]. Olthof et al. analyzed the incidence of [115]. Some authors reported less complications and a high postoperative liver failure in a combined series of two Western success rate of ENBD compared to EBD [116, 117]. However, centers specialized in PHC. A risk score was proposed to others reported comparable results with EBD and PTBD select candidates for PVE based on FRL volume combined [110]. Western centers generally do not perform ENBD be- with jaundice at presentation, preoperative cholangitis and cause the nasal tubes easily dislocate and from the patients’ preoperative bilirubin level > 50 μmol/L [121]. perspective, are usually less well tolerated. ENBD drains bile Accelerated tumor growth due to PVE does not seem to externally via the naso-gastro-duodenal tubes, precluding bile influence the survival of PHC patients [122, 123]. PVE how- entering the intestinal system and therefore demands bile sup- ever, does predetermine the side of the resection and in case of pletion. This is then only possible via the oral route or a sec- new findings that may require a change of strategy, this cannot ond gastroduodenal tube. be reversed. If the patient becomes unresectable due to disease Several retrospective studies have been performed progression in the waiting time, the atrophy-hypertrophy re- concerning the optimal drainage method [118] mainly empha- action stabilizes with time and the overall liver volume and sizing that each method comes with its own set of complica- function remain unchanged. However, the persistence of the tions such as cholangitis, pancreatitis or vascular complica- atrophied, usually contaminated cholestatic liver lobe can be tions. Until evidence has been presented, EBD remains the accompanied with adverse effects such as liver abscess, com- reference method in most Western countries [107, 109]. plicating further palliative treatment of the patient who typi- The balance between the benefits and risks of biliary drain- cally will need repeated treatment with biliary stents [124]. age is fragile and drainage strategies should be optimized in Additional embolization of segment 4 in preparation of ex- order to minimalize the risk of intrinsic complications. Due to tended right hemihepatectomy is an option depending on the these risks, it may be advisable to undertake surgery without target increase of FRL volume that needs to be attained. To this prior drainage provided there is a surplus of remnant liver end, the left portal vein branches to segment 4 are occluded volume. Wiggers et al. showed that with a FLR > 50% preop- along with embolization of the right portal venous system. The erative biliary drainage was of no added value [88]. technique is challenging and requires an experienced interven- Hence, in patients requiring extended resection, we now tional radiologist since access to the left portal venous system use selective preoperative biliary drainage of only the future can give additional risk of injury. Backflow of embolization remnant liver when FRL is small (< 50%) whereas complete material into the left portal venous system can lead to inadver- preoperative biliary drainage is mandatory in the event of tent embolization and thrombosis of the portal veins supplying (recent) cholangitis. the FRL. Alternatively, to decrease these risks, partial emboli- zation of only segment 4a can be performed [120, 156]. Portal vein embolization Preoperative radiotherapy If the FRL has not sufficient volume or function to undergo a safe resection, portal vein embolization (PVE) is the standard There is no general consensus regarding the use of neoadju- intervention to increase the functional capacity of the FRL. vant therapy for PHC. Low-dose preoperative radiotherapy The local hemodynamic changes proposedly result in a release (3 × 3.5 Gy in 3 days prior to resection) was instituted in our 298 Langenbecks Arch Surg (2018) 403:289–307 center in patients with resectable PHC who received preoper- sided approach comprising an anatomical left ative drainage to prevent seeding metastases. The increased hemihepatectomy, whereas coming from the right, an extended risk of seeding metastases after biliary drainage is an area of right hemihepatectomy is required to include the central sector. debate [114, 125] with various outcomes reported in literature Intraoperative frozen-section pathological examination of [86, 126]. In our cohort, seeding metastasis in up to 20% of the resection margins of the biliary ducts is performed to con- patients after endoscopic stenting has been observed in the firm radicality at the ductal level. In case of residual tumor in laparotomy scar or drain tract [86] . This complication was the resection margin, the level of biliary resection is extended associated with tumor cells contained in the bile that inevita- although in our series, survival was worse in these patients bly contaminated the operative field after bile duct transection compared to patients who had an initial free margin [108]. in the course of resection. After using preoperative radiation, Some surgeons routinely drain the biliary ducts of the remnant no catheter tract recurrences after drainage have been reported liver after reconstruction using trans-anastomotic tubes. We [125]. However, there is no evidence for this concept that is usually do not internally drain the hepaticojejunostomy but uniquely applied in our center [127]. A recent study conducted when there are PTC drains in place, these are positioned in two Western specialized centers, did not show an associa- across the anastomoses allowing access for possible postoper- tion of seeding metastases with center or mode of preoperative ative cholangiography. As recently reported, leaving the drainage, i.e. endoscopic or percutaneous drainage [125]. drains open is not advised since the loss of bile negatively New prospective studies are needed to develop guidelines on influences postoperative regeneration of the liver remnant this topic. [130]. Surgical aspects Resection of the portal vein bifurcation: unconditional or on demand General considerations Controversy exists regarding the unconditional, simultaneous The goal of surgical treatment is to achieve an R0 resection of en bloc resection of the portal vein bifurcation with the tumor. the tumor along with clearance of the regional lymph nodes. The hilar vessels run adjacent to the tumor with less than Because of the central location of the tumor at the liver hilum 1 mm between tumor and portal vein while perineural infiltra- and its proximal extension into the segmental bile ducts, com- tion of the tumor along the bile ducts is a common feature of plete resection requires excision of the extrahepatic biliary duct PHC. Because of this anatomical proximity, a no-touch tech- in combination with extended liver resection. The close relation nique was proposed by the Berlin group in which resection of the tumor with the right and left portal vein and the hepatic includes unconditional excision of the portal vein bifurcation artery branches often demands concomitant vascular resections en bloc with tumor excision and hepatectomy. Using this on- and reconstruction. The Japanese surgeons were the first to cological strategy, dissection of the hepatic hilum is avoided show in the nineties of the previous century, that this aggressive and the risk of dissemination of tumor cells minimized. This approach resulted in improved long-term survival [128]. Still technique however is less feasible in patients requiring left following these lines, radical resection entails excision of the hepatectomies due to the fact that the right hepatic artery usu- liver hilum with (extended) hemihepatectomy including seg- ally crosses the hepatic hilum directly anterior or posterior of ment IV and the caudate lobe, complete lymphadenectomy of the tumor, unless there is a displaced right hepatic artery orig- the hepatoduodenal ligament and excision of the portal vein inating from the superior mesenteric artery that runs along the bifurcation when involved [129]. Additionally, arterial resec- right-lateral margin of the hepatoduodenal ligament. Also, re- tions are undertaken in order to achieve an R0 resection (Fig. 5). construction of the right portal branches with the main stem is The type of resection depends on location of the tumor and technically more demanding. biliary anatomy at the hepatic duct confluence, the radial and Multivariate analysis of the Berlin series of resected PHC longitudinal extent of tumor into the intrahepatic bile ducts and showed that portal vein resection was the only significant its association with adjacent periductal structures, portal vein factor to influence patient survival after confirmed R0 resec- and hepatic arteries. When viewing cross-sectional imaging tion. The 5-year survival rate of curative liver resection in their studies, it is important to perceive the tumor in a three- series was 65% with portal vein resection as compared to 28% dimensional fashion. The tumor extends along the right and left without [131]. Other authors advise to only perform portal biliary ducts into the liver and at the same time, in anterior and vein resection when during exploration, the portal venous posterior direction into the ducts of segment 4 and segment 1, bifurcation/contralateral portal venous branch is found to be respectively. Complete resection therefore should include the invaded by tumor. central sector of the liver along the antero-posterior axis includ- In our center, we do not advocate the unconditional exci- ing the segments 1 and 4 [62]. As the central sector anatomi- sion of the portal vein bifurcation, also because portal venous cally is part of the left hemi-liver, there is an advantage of a left- reconstruction has been associated with an increased Langenbecks Arch Surg (2018) 403:289–307 299 Fig. 5 Extended resection for PHC should include the central sector (segment 4) with segment 1 along the antero-posterior axis of the liver. Depending on the predominant side of the tumor, a left (extended) or right extended hemihepatectomy is chosen for en bloc resection of the hilar area postoperative morbidity rate [88]. Modern preoperative imag- tumor involvement. The right hepatic artery with its right an- ing techniques now accurately demonstrate portal venous in- terior and posterior branches however, is frequently infiltrated volvement. Relying on this information, we only resect the by tumor. The choice of performing a right or left resection is portal vein bifurcation en bloc with the tumor right away when often dictated by the side of the liver in which the hepatic there is evidence of vascular involvement on CT-scan. artery branches are free. In PHC predominantly involving Otherwise, the decision to resect the portal vein bifurcation the left liver, (extended) left hemihepatectomy with concom- is made intraoperatively. This policy has led to portal vein itant resection of the right hepatic artery is hampered by pre- resection in approximately 20% of our cases (Table 2). serving a tumor free, intrahepatic distal stump for arterial re- Survival analysis of our series showed an overall 5-year sur- construction especially when tumor mass is substantial. vival of 44.3%, which is comparable with the 5-year survival Microsurgical techniques are usually applied to create a safe of 43% reported by the Berlin group in a series of patients that anastomosis with the right posterior branch of the right hepatic underwent R0 resection using unconditional en bloc portal artery in these cases. Combined arterial resection and recon- vein resection [132]. struction with portal vein resection is controversial. In the Concomitant resection of the portal vein bifurcation with Nagoya cohort of resected patients, this challenging technique extended hemihepatectomy is followed by end-to-end anasto- was associated with a mortality of 2% and a 5-year survival of mosis of the left portal vein with the main portal venous stem. 30% for patients with advanced cholangiocarcinoma [133]. Complete mobilization of the left portal vein by detaching all Preoperative embolization of the branch of the hepatic ar- side-branches to segments 4 and 1 facilitates reconstruction. tery feeding the future remnant liver and subsequent excision As the caliber of the left portal vein usually is much smaller, it is not attractive in our view, because the biliary anastomosis is important to bevel the anastomosis after oblique clamping depends on arterial periductal arterial perfusion and loss of in order to prevent stenosis of the anastomosis. With arterial blood supply will lead to failure of the (extended) left hepatectomy, reconstruction of the right portal hepaticojejunostomies. Arterialization of the portal vein can branches with the main stem is technically more demanding. be used as last resort when arterial perfusion of the remnant liver is sacrificed, however is preferably applied for salvage of the remnant liver [134]. Concomitant resection of the hepatic artery branch feeding the remnant liver Concomitant resection of segment 1 Adequate arterial perfusion is crucial to function of the rem- nant liver. The left hepatic artery runs along the medial side of As pointed out above, the caudate lobe is part of the central the hepatoduodenal ligament and is therefore less at risk for antero-posterior axis of the liver and is preferably resected 300 Langenbecks Arch Surg (2018) 403:289–307 en bloc with the tumor and liver hilum. Although the seg- Parenchyma sparing liver resection ment 1 bile ducts often drain into the left ductal system, they may drain into any part of the hepatic duct confluence and As volume and function of the remnant liver are the most these ducts are frequently involved by tumor as well. critical factors for postoperative outcomes, parenchyma pre- Routine S1 resection en bloc with (extended) serving techniques can be applied in selected cases. These hemihepatectomy has therefore been implemented at our techniques can be used as an alternative to PVE or in addition institution since 1998 and has increased the rate of R0 re- to PVE, in order to spare as much functional liver tissue as sections and has resulted in improved survival [135]. En possible. bloc excision of segment 1 is therefore recommended with In right-sided tumors that require an extended right resection of PHC. hemihepatectomy, the cranial part of segment 4 (i.e. 4a) may be preserved depending on the level of involvement of the segment 4 bile duct. Free margin of the cut segment 4 bile duct Right or left (extended) hepatectomy is checked using frozen-section pathological examination. In case of a left-sided tumor, a modified extended left The decision to perform a right or left hepatectomy depends hemihepatectomy may be undertaken. Extending left resec- on local tumor extension, portal venous and hepatic arterial tion to include segment 5, the adjacent part of segment 8 involvement and the FRL volume and function. For optimal may be preserved. Whether this can be performed depends preoperative preparation of the patient, such as biliary drain- on the proximal extent of the tumor into the right segmental age or the need for PVE, it is important to preoperatively ducts, and the anatomy of the right sectoral ducts (B5/8 and determine the side of the liver to be resected. B6/7) in relation to the right hepatic duct and hepatic duct If the tumor extent and FRL volume allow both options, a confluence. right-sided (extended) hepatectomy is often preferred because A pitfall of sparing portions of the central sectoral segments it is more likely to achieve oncological radicality and is more a 4 and 8 is cutting off their portal venous and arterial blood straightforward procedure for several reasons [136]. Firstly, supply by resection of the tumor. The central position of the because the biliary confluence is located on the right side of tumor often requires sacrifice of the middle hepatic artery to the hepatoduodenal ligament, a right hepatectomy allows segment 4 or the right-anterior portal vein and hepatic artery more complete resection of the tumor. Also, the right hepatic branches to segment 8 leading to parenchymal infarction. duct is often short (< 1 cm) or even absent in case of a triple Another possibility for parenchymal preservation is hepatic confluence, while the left hepatic duct has a relatively performing a central liver resection (mesohepatectomy) when long and straight course until reaching the border of the left the bile ducts of segments 6 and 7 and the left lateral segments 2 and 3 are not infiltrated by the tumor. This complex proce- portal vein and branching off into the ducts of segments 2 and 3[137]. Therefore, tumors that invade the right sectoral ducts dure includes resection of the central sectors of the liver in- and even the segmental ducts to segment 4 (Bismuth-Corlette cluding segments 4, 5 and 8. In these cases, multiple jejunal type IIIa-IV) can be radically resected by extended right anastomoses with the remaining intrahepatic segmental bile hemihepatectomy. A disadvantage of a right-sided approach ducts are required [129]. A formal central resection is only is that segments 2 and 3 are small and that in many patients possible when the vascular structures supplying the left lateral preoperative right PVE is necessary before undertaking ex- segments as well as the right-posterior segments 6 and 7 are tended right hemihepatectomy. free of the tumor and can be preserved. Tumors predominantly involving the left ductal system (Bismuth-Corlette type IIIb-IV) require a left-sided ap- ALPPS proach. The advantage of a left-sided resection is that the remnant liver, i.e. the right liver segments, usually has more In situ split of the liver in combination with portal vein ligation volume and resection can be extended farther into the right (ALPPS) has been introduced as a method to induce rapid liver. Of note, the volume of segments 6 + 7 usually exceeds hypertrophy of the FRL. Because of the higher mortality and that of segments 2 + 3 which may direct the choice of a right morbidity reported in the initial series of ALPPS, this method or left-sided approach. A formal extended left along with great interest has generated a heated discussion in hemihepatectomy following the medial margin of the right the surgical community [138, 139]. The advantage of ALPPS hepatic vein is technically more difficult and depending on is debated in extended right hemihepatectomy as compared to involvement of the segment 8 ducts, part of the anterior right complete embolization of the right portal venous system in- sector may be preserved (see below). A down-side of a left cluding segment 4 as described above. Several authors have resection is that construction of the (often multiple) biliary reported their results of ALPPS in patients with PHC. Due to anastomoses may be more complex with a higher risk of stenting of the biliary system and ensuing cholangitis, patients biliary complications. were at increased risk of interstage morbidity and mortality Langenbecks Arch Surg (2018) 403:289–307 301 [140, 141]. ALPPS for PHC demonstrated poor outcomes Liver failure is a dreaded complication after extensive hepa- with 48% perioperative mortality in the ALPPS registry tectomy and is a major cause of mortality in patients with PHC [140]. We therefore for now, do not recommend ALPPS for [21, 148]. The risk of postoperative liver failure is increased resection of PHC and rather consider PVE with selective em- due to the combination of intra-operative blood loss, a small bolization of the left portal vein branches to segment 4 for FLR and cholestasis [21, 88, 121]. Reported liver failure augmentation of FRL volume in patients requiring extended ranges from 3 to 25% [31, 88, 143, 144]. Biliary leakage from right hemihepatectomy. either the hepaticojejunal anastomosis or the parenchymal dis- section surface ranges from 6 to 29% [31, 88, 143]. Infections range from 23 to 66% and bleeding complications from 4 to The extent of lymphadenectomy 9% [15, 31, 88, 145, 146](Table 5). Standard lymphadenectomy includes resection of lymph nodes around the extrahepatic bile duct, the portal vein Discussion and future perspectives and hepatic artery, as well as the lymphatic channels and nerves contained in the hepatoduodenal ligament. The The management of perihilar cholangiocarcinoma is complex number of lymph nodes resected is also relevant as less and requires close multidisciplinary collaboration between than 4 lymph nodes evaluated in the specimen was iden- hepatobiliary surgeons, endoscopists, radiologists, medical tified as a poor prognostic factor for time to recurrence oncologists and pathologists. In this review, we provide a [142]. Lymphnodemetastasesthatare limitedtothe he- summary of the current diagnosis and work-up in the light patic pedicle or the hepatoduodenal ligament (N1) are of extended resection and elaborate on future perspectives. included in the field of resection, but those along the Establishing the diagnosis of PHC is still one of the most common hepatic artery and/or celiac axis (N2) are consid- challenging aspects of the diagnostic work-up. New diagnos- ered distant metastases. N2-disease has a poor prognosis tic endoscopic techniques such as SpyGlass and endoscopic and disease specific survival of patients with para-aortic ultrasound enable more precise biopsies, resulting in increased lymph node metastasis was similar to M1 patients, sug- sensitivity and specificity in diagnosing biliary strictures [33]. gesting that survival is not influenced by the extent of These techniques will likely decrease the rate of misdiagnoses lymph node dissection, but rather by the presence of N2 and bring down the number of futile resections performed for disease [12, 78]. Therefore, we do not recommend routine benign disease. The currently used conventional tumor marker lymphadenectomy beyond the hepatoduodenal ligament. CA19–9 is not particularly sensitive or specific. A combina- tion of different markers seems more useful in the diagnosis Complications and follow-up of PHC. Biomarkers such as CYFRA21-1 and MUC-5 need to be evaluated in larger cohorts to assess its Postoperative morbidity and mortality of patients with PHC is clinical value. Molecular techniques such as circulating significant. Reported mortality ranges from 5 to 18% even in miRNA’s and Tumor Educated Platelets (TEP) represent an high volume centers, and morbidity as high as 60–70%, with exciting area with great promise in this field [149, 150]. For around 50% severe complications (Clavien-Dindo grade III or now, approximately 50% of patients with suspicion on PHC higher) [21]. Table 5 shows the incidence of the most common undergo resection without a confirmed tissue diagnosis. complications as reported in literature together with the com- CT-volumetry has traditionally been the golden standard plications recorded in our own series. Risks are particularly for assessment of a sufficient FRL. However, not only the high in patients with tumors requiring an extended resection. quantity but also the quality of the FRL is important while liver volume does not correlate with liver function. In our cohort, total and regional (segmental) liver function was pre- Table 5 Complications and reported incidence in a selection of operatively evaluated using HBS. This quantitative method literature reports including the AMC series allows measurement of FRL-function and can be used in pa- Complication type Incidence literature Incidence AMC tients with impaired quality of liver parenchyma using the same cutoff value. A limitation of using HBS in patients with Liver failure 3–25% [31, 88, 143, 144] 19% (29/156*) PHC is that bilirubin induces competitive uptake with Biliary leakage 6–29% [31, 88, 143] 30% (47/156) mebrofenin as both are taken up by the same hepatocyte trans- Bleeding 4–9% [15, 31, 88, 145, 146] 8% (13/156) porters. In patients with obstructive cholestasis, HBS may Multi organ failure 1–3% [32, 87, 135] 2% (3/156) underestimate liver function when the biliary system is not Infections 23–66% [15, 31, 88, 145, 146] 22% (35/156) completely drained. Mortality 5–17% [88, 143, 147, 157] 9% (18/201) The additional value of staging laparoscopy in the future is *Total cohort: n = 156, missing n =45 questionable, considering the low yield and further 302 Langenbecks Arch Surg (2018) 403:289–307 improvements in accuracy of preoperative imaging tech- in a highly selected group of patients [153, 154]. The idea of a niques. Using a risk score allows to predict unresectable short course of radiation preceding resection was to eradicate PHC at staging laparoscopy in order to make a more selective free floating tumor cells in the bile, thus preventing viable approach to staging laparoscopy. tumor cells of contaminating the peritoneal surface. There is Since patients with PHC typically present with obstructive however no evidence for this concept. jaundice, decompression of the biliary tract is a much-debated The challenge in the coming years is to reduce morbidity topic. For the past, few years it has become clear that drainage and mortality associated with extended resections for PHC. of the biliary tract comes with a serious risk of drainage- Optimizing preoperative workup is key to achieve improved related complications. Since obstructive jaundice impairs liver outcomes after extended resections. regeneration, biliary drainage is still advised in case of a small FLR. The optimal drainage method has yet to be established. In The Netherlands the DRAINAGE trial is underway to eval- Conclusion uate outcomes of PTBD vs. EBD in resectable PHC [108, 151]. In this multi-center trial with an all-comers design, all The field of work-up in PHC is changing with the introduction patients with a presumed PHC and cholestasis are randomized of newer modalities that have emerged over the past few to undergo PTBD or EBD. The study is powered for drainage- years. Upcoming diagnostic modalities and molecular tech- related complications and postoperative outcomes. For now, niques might help to decrease the rate of misdiagnosis of be- we use selective preoperative, endoscopic biliary drainage of nign, inflammatory disease. Assessment of liver function with only the future remnant liver when FRL is small (< 50%) hepatobiliary scintigraphy provides better information on the unless mandated by cholangitis. FRL than volume alone. The selective use of staging laparos- The most important prognostic factor for long-term surviv- copy is advisable to avoid futile laparotomies. In patients re- al of PHC is a margin negative resection of the hilar tumor. In quiring extended resection, selective preoperative biliary experienced hands, even Bismuth-Corlette type IV tumors can drainage is mandatory in cholangitis and when FRL is small be resected with curative intent. R0 resection requires an ag- (< 50%). Preoperative PVE is used when FRL volume is less gressive surgical approach encompassing hilar resection in than 40% and optionally includes the left portal vein branches combination with extended liver resection, frequently accom- to segment 4. ALPPS as alternative to PVE is not recommend- panied with vascular reconstructions. These extended resec- ed in PHC. N2 positive lymph nodes preclude long-term sur- tions are associated with higher morbidity and mortality rates vival. The benefit of unconditional en bloc resection of the than experienced in liver resections without bile duct resec- portal vein bifurcation is uncertain. Although still associated tion, probably because of the sequelae of obstructive jaundice. with considerable morbidity and mortality, an aggressive sur- Survival after resection is however favorable, with 5-year gical approach encompassing extended liver resection includ- overall survival rates comparable with survival after extended ing segment 1, regional lymphadenectomy and conditional liver resection for colorectal liver metastases. portal venous resection offers the only chance for long-term PVE is a widely accepted interventional procedure to in- survival. crease FRL volume and function before undertaking major liver resection. This method of liver augmentation is especial- Authors’ contributions F. Rassam: study conception and design, acquisi- ly of benefit in patients with PHC who require extended liver tion of data, analysis and interpretation of data, drafting of manuscript. E. Roos: study conception and design, acquisition of data, analysis and resection in predamaged livers. We therefore advocate the interpretation of data, drafting of manuscript. K.P. van Lienden: critical liberal use of PVE in patients with PHC in whom the FRL is revision of manuscript. J.E. van Hooft: critical revision of manuscript. below 40% of total liver volume. It is important to note that to H.J. Klümpen: critical revision of manuscript. G. van Tienhoven: critical obtain the maximum hypertrophy effect of segments 2 and 3, revision of manuscript. R.J. Bennink: critical revision of manuscript. M.R. Engelbrecht critical revision of manuscript. A. Schoorlemmer: crit- the side-branches of the left portal vein to segment 4 can be ical revision of manuscript. U.H.W. Beuers: critical revision of manu- embolized as well. Obviously, selective embolization of the script. J. Verheij: critical revision of manuscript. M.G. Besselink: critical segment 4 branches requires expertise of the interventional revision of manuscript. O.R.C. Busch: critical revision of manuscript. radiologist as available in specialized centers. Although the T.M. van Gulik: Study conception and design, critical revision of first successful case of ALPPS was reported by Schlitt in a manuscript. patient with PHC, the use of ALPPS in PHC as alternative to Compliance with ethical standards PVE is not recommended because of the reported high mor- bidity and mortality of the procedure in this category of pa- Conflict of interest The authors declare that they have no conflict of tients [152]. interest. There are no established strategies regarding the use of neo-adjuvant therapies in PHC. The only exception is neo- Ethical approval This article does not contain any studies with human adjuvant chemo-radiation therapy prior to liver transplantation participants or animals performed by any of the authors. Langenbecks Arch Surg (2018) 403:289–307 303 Open Access This article is distributed under the terms of the Creative resectability and outcomes in 380 patients. J Am Coll Surg 215: Commons Attribution 4.0 International License (http:// 343–355. https://doi.org/10.1016/j.jamcollsurg.2012.05.025 creativecommons.org/licenses/by/4.0/), which permits unrestricted use, 16. Coelen RJS, Ruys AT, Wiggers JK, Nio CY, Verheij J, Gouma DJ, distribution, and reproduction in any medium, provided you give appro- Besselink MGH, Busch ORC, van Gulik TM (2016) Development priate credit to the original author(s) and the source, provide a link to the of a risk score to predict detection of metastasized or locally ad- Creative Commons license, and indicate if changes were made. vanced perihilar cholangiocarcinoma at staging laparoscopy. Ann Surg Oncol 23(S5):904–910. https://doi.org/10.1245/s10434-016- 5531-6 17. Valle J, Wasan H, Palmer DH et al (2010) Cisplatin plus gemcitabine versus gemcitabine for biliary tract cancer. N Engl J References Med 4:395–397. https://doi.org/10.1586/egh.10.45 18. Wyluda E, Yee NS (2015) Systemic treatment of advanced biliary tract carcinoma: emerging roles of targeted therapy and molecular 1. Aljiffry M, Abdulelah A, Walsh M et al (2009) Evidence-based profiling. Clin Cancer Drugs 2:80–86 approach to cholangiocarcinoma: a systematic review of the cur- 19. Ito F, Cho CS, Rikkers LF, Weber SM (2009) Hilar cholangiocar- rent literature. J Am Coll Surg 208:134–147. https://doi.org/10. cinoma: current management. Ann Surg 250:210–218. https://doi. 1016/j.jamcollsurg.2008.09.007 org/10.1097/SLA.0b013e3181afe0ab 2. Gatto M, Alvaro D (2010) New insights on cholangiocarcinoma. 20. Groot Koerkamp B, Wiggers JK, Gonen M et al (2015) Survival World J Gastrointest Oncol 2:136–145. https://doi.org/10.4251/ after resection of perihilar cholangiocarcinoma-development and wjgo.v2.i3.136 external validation of a prognostic nomogram. Ann Oncol 26: 3. Khan SA, Thomas HC, Davidson BR, Taylor-Robinson SD 1930–1935. https://doi.org/10.1093/annonc/mdv279 (2005) Cholangiocarcinoma. Lancet (London, England) 366: 21. Coelen RJS, Olthof PB, van Dieren S et al (2016) External vali- 1303–1314. https://doi.org/10.1016/S0140-6736(05)67530-7 dation of the estimation of physiologic ability and surgical stress 4. Bismuth H, Nakache R, Diamond T (1992) Management strate- (E-PASS) risk model to predict operative risk in perihilar cholan- gies in resection for hilar cholangiocarcinoma. Ann Surg 215(1): giocarcinoma. JAMA Surg 147:26–34. https://doi.org/10.1001/ 31–38. https://doi.org/10.1097/00000658-199201000-00005 jamasurg.2016.2305 5. Vaeteewoottacharn K, Seubwai W, Bhudhisawasdi V et al (2014) 22. Hemming AW, Reed AI, Fujita S et al (2005) Surgical manage- Potential targeted therapy for liver fluke associated cholangiocar- ment of hilar cholangiocarcinoma. Ann Surg 241:693–699-702. cinoma. J Hepatobiliary Pancreat Sci 21:362–370. https://doi.org/ https://doi.org/10.1097/01.sla.0000160701.38945.82 10.1002/jhbp.65 23. Anderson JE, Hemming AW, Chang DC et al (2012) Surgical 6. Ong CK, Subimerb C, Pairojkul C et al (2012) Exome sequencing management trends for cholangiocarcinoma in the USA 1998- of liver flukeg-associated cholangiocarcinoma. Nat Genet 44: 2009. J Gastrointest Surg 16:2225–2232. https://doi.org/10.1007/ 690–693. https://doi.org/10.1038/ng.2273 s11605-012-1980-9 7. Rizvi S, Gores GJ (2013) Pathogenesis, diagnosis, and manage- 24. Nagino M, Ebata T, Yokoyama Y et al (2013) Evolution of surgi- ment of cholangiocarcinoma. Gastroenterology 145:1215–1229. cal treatment for perihilar cholangiocarcinoma: a single-center 34- https://doi.org/10.1053/j.gastro.2013.10.013 year review of 574 consecutive resections. Ann Surg 258:129– 8. Sarkar S, Bowlus CL (2016) Primary sclerosing cholangitis: mul- 140. https://doi.org/10.1097/SLA.0b013e3182708b57 tiple phenotypes, multiple approaches. Clin Liver Dis 20:67–77. 25. Maillette de Buy Wenniger LJ, Beuers U (2015) Immunoglobulin https://doi.org/10.1016/j.cld.2015.08.005 G4-related cholangiopathy: clinical and experimental insights. 9. Boberg KM, Schrumpf E, Bergquist A et al (2000) Curr Opin Gastroenterol 31:252–257. https://doi.org/10.1097/ Cholangiocarcinoma in primary sclerosing cholangitis: K-ras mu- mog.0000000000000170 tations and Tp53 dysfunction are implicated in the neoplastic 26. Hubers LM, Maillette de Buy Wenniger LJ, Doorenspleet ME et al development. J Hepatol 32:374–380 (2015) IgG4-associated cholangitis: a comprehensive review. Clin 10. Nagino M, Ebata T, Yokoyama Y et al (2013) Evolution of surgi- Rev Allergy Immunol 48:198–206. https://doi.org/10.1007/ cal treatment for perihilar cholangiocarcinoma: a single-center 34- s12016-014-8430-2 year review of 574 consecutive resections. Ann Surg 258:129– 27. Zaydfudim VM, Wang AY, De Lange EE et al (2015) IgG4- 140. https://doi.org/10.1097/SLA.0b013e3182708b57 associated cholangitis can mimic hilar cholangiocarcinoma. Gut 11. Govil S, Reddy MS, Rela M (2014) Surgical resection techniques Liver 9:556–560. https://doi.org/10.5009/gnl14241 for locally advanced hilar cholangiocarcinoma. Langenbeck's 28. Zen Y, Britton D, Mitra V et al (2015) A global proteomic study Arch Surg 399:707–716. https://doi.org/10.1007/s00423-014- identifies distinct pathological features of IgG4-related and prima- 1216-4 ry sclerosing cholangitis. Histopathology. https://doi.org/10.1111/ 12. Regimbeau JM, Fuks D, Le Treut YP et al (2011) Surgery for hilar his.12813 cholangiocarcinoma: a multi-institutional update on practice and 29. Deshpande V, Zen Y, Chan JK et al (2012) Consensus statement outcome by the AFC-HC study group. J Gastrointest Surg 15: on the pathology of IgG4related disease. Mod Pathol 25:11811192 480–488. https://doi.org/10.1007/s11605-011-1414-0 www.modernpathology.org. https://doi.org/10.1038/modpathol. 13. Hartog H, Ijzermans JNM, van Gulik TM, Koerkamp BG (2016) 2012.72 Resection of perihilar cholangiocarcinoma. Surg Clin North Am 30. Kamisawa T, Zen Y, Pillai S, Stone JH (2015) IgG4-related dis- 96:247–267. https://doi.org/10.1016/j.suc.2015.12.008 ease. Lancet 385:1460–1471. https://doi.org/10.1016/S0140- 6736(14)60720-0 14. Coelen RJS, Ruys AT, Besselink MGH, Busch ORC, van Gulik TM (2016) Diagnostic accuracy of staging laparoscopy for detect- 31. Kloek JJ, van Deldein OM, Erdogan D et al (2008) Differentiation ing metastasized or locally advanced perihilar cholangiocarcino- of malignant and benign proximal bile duct strictures: the diag- ma: a systematic review and meta-analysis. Surg Endosc 30(10): nostic dilemma. World J Gastroenterol 14:5032–5038. https://doi. 1–11. https://doi.org/10.1007/s00464-016-4788-y org/10.3748/wjg.14.5032 15. Matsuo K, Rocha FG, Ito K et al (2012) The blumgart preopera- 32. Corvera CU, Blumgart LH, Darvishian F et al (2005) Clinical and tive staging system for hilar cholangiocarcinoma: analysis of pathologic features of proximal biliary strictures masquerading as 304 Langenbecks Arch Surg (2018) 403:289–307 hilar cholangiocarcinoma. J Am Coll Surg 201:862–869. https:// 48. Kurihara T, Yasuda I, Isayama H et al (2016) Diagnostic and therapeutic single-operator cholangiopancreatoscopy in doi.org/10.1016/j.jamcollsurg.2005.07.011 biliopancreatic diseases: prospective multicenter study in Japan. 33. Victor DW, Sherman S, Karakan T, Khashab MA (2012) Current World J Gastroenterol 22:1891–1901. https://doi.org/10.3748/ endoscopic approach to indeterminate biliary strictures. World J wjg.v22.i5.1891 Gastroenterol 18:6197–6205. https://doi.org/10.3748/wjg.v18. 49. Navaneethan U, Hasan M, Lourdusamy V et al (2015) Single- i43.6197 operator cholangioscopy and tarhetted biopsies in the diagnosis 34. Esnaola NF, Meyer JE, Karachristos A, Maranki JL, Camp ER, of inditerminat ebiliary strictures: a systematic review. Denlinger CS (2016) Evaluation and management of intrahepatic Gastrointest Endosc 82:608–614. https://doi.org/10.1007/978-1- and extrahepatic cholangiocarcinoma. Cancer 122(9):1349–1369. 4939-2914-6 https://doi.org/10.1002/cncr.29692 50. Hara K, Yamao K, Mizuno N, Hijioka S, Imaoka H, Tajika M, 35. Navaneethan U, Njei B, Lourdusamy V et al (2016) Comparative Tanaka T, Ishihara M, Okuno N, Hieda N, Yoshida T, Niwa Y effectiveness of biliary brush cytology and intraductal biopsy for (2016) Endoscopic ultrasonography-guided biliary drainage: who, detection of malignant biliary strictures: a systematic review and when, which, and how? World J Gastroenterol 22(3):1297–1303. meta-analysis. Gastrointest Endosc. 1:23–30. https://doi.org/10. https://doi.org/10.3748/wjg.v22.i3.1297 1007/s40778-014-0003-z.Genome 51. Lee JH, Salem R, Aslanian H et al (2004) Endoscopic ultrasound 36. Barr Fritcher EG, Kipp BR, Halling KC, Clayton AC (2014) and fine-needle aspiration of unexplained bile duct strictures. Am FISHing for pancreatobiliary tract malignancy in endoscopic J Gastroenterol 99:1069–1073. https://doi.org/10.1111/j.1572- brushings enhances the sensitivity of routine cytology. 0241.2004.30223.x Cytopathology 25:288–301. https://doi.org/10.1111/cyt.12170 52. Itoi T, Itokawa F, Uraoka T et al (2013) Novel EUS-guided 37. Liew ZH, Loh TJZ, Lim TKH et al (2017) Role of fluorescence in gastrojejunostomy technique using a new double-balloon enteric situ hybridization in diagnosing cholangiocarcinoma in indetermi- tube and lumen-apposing metal stent (with videos). Gastrointest nate biliary strictures. J Gastroenterol Hepatol. https://doi.org/10. Endosc 78:934–939. https://doi.org/10.1016/j.gie.2013.09.025 1111/jgh.13824 53. Masaki Y, Kurose N, Yamamoto M, Takahashi H, Saeki T, Azumi 38. Barr Fritcher EG, Voss JS, Brankley SM et al (2015) An optimized A, Nakada S, Matsui S, Origuchi T, Nishiyama S, Yamada K, set of fluorescence in situ hybridization probes for detection of Kawano M, Hirabayashi A, Fujikawa K, Sugiura T, Horikoshi pancreatobiliary tract cancer in cytology brush samples. M, Umeda N, Minato H, Nakamura T, Iwao H, Nakajima A, Gastroenterology 149:1813–1824e1. https://doi.org/10.1053/j. Miki M, Sakai T, Sawaki T, Kawanami T, Fujita Y, Tanaka M, gastro.2015.08.046 Fukushima T, Eguchi K, Sugai S, Umehara H (2012) Cutoff 39. Gonda TA (2017) Mutation profile and fluorescence in situ hy- values of serum IgG4 and histopathological IgG4+ plasma cells bridization analyses increase detection of malignancies in biliairy for diagnosis of patients with IgG4-related disease. Int J strictures. Clin Gastroenterol Hepatol. https://doi.org/10.1016/j. Rheumatol 2012:0–5. https://doi.org/10.1155/2012/580814 cgh.2016.12.013.This 54. Oseini AM, Chaiteerakij R, Shire AM et al (2011) Utility of serum 40. Fogel EL, DeBellis M, McHenry L et al (2006) Effectiveness of a immunoglobulin G4 in distinguishing immunoglobulin G4- new long cytology brush in the evaluation of malignant biliary associated cholangitis from cholangiocarcinoma. Hepatology 54: obstruction: a prospective study. Gastrointest Endosc 63:71–77. 940–948. https://doi.org/10.1002/hep.24487 https://doi.org/10.1016/j.gie.2005.08.039 55. Doorenspleet ME, Hubers LM, Culver EL et al (2016) IgG4+ B- 41. Coté GA, Sherman S (2011) Biliary stricture and negative cytol- cell receptor clones distinguish IgG4-related disease from primary ogy: what next? Clin Gastroenterol Hepatol 9:739–743. https:// Sclerosing cholangitis and biliary/pancreatic malignancies. doi.org/10.1016/j.cgh.2011.04.011 Hepatology 64(2):1–49. https://doi.org/10.1002/hep.28568 42. De Bellis M, Fogel EL, Sherman S et al (2003) Influence of stric- 56. Charatcharoenwitthaya P, Enders FB, Halling KC, Lindor KD ture dilation and repeat brushing on the cancer detection rate of (2008) Utility of serum tumor markers, imaging, and biliary cy- brush cytology in the evaluation of malignant biliary obstruction. tology for detecting cholangiocarcinoma in primary sclerosing Gastrointest Endosc 58:176–182. https://doi.org/10.1067/mge. cholangitis. Hepatology 48:1106–1117. https://doi.org/10.1002/ 2003.345 hep.22441 43. Fukuda Y, Tsuyuguchi T, Sakai Yet al (2005) Diagnostic utility of 57. Grunnet M, Mau-Sørensen M (2014) Serum tumor markers in bile peroral cholangioscopy for various bile-duct lesions. Gastrointest duct cancer—a review. Biomarkers 19:437–443. https://doi.org/ Endosc 62:374–382. https://doi.org/10.1016/j.gie.2005.04.032 10.3109/1354750X.2014.923048 44. Kawashima H, Itoh A, Ohno E et al (2012) Transpapillary biliary 58. Viterbo D, Gausman V, Gonda T (2016) Diagnostic and therapeu- forceps biopsy to distinguish benign biliary stricture from malig- tic biomarkers in pancreaticobiliary malignancy. World J nancy: how many tissue samples should be obtained? Dig Endosc Gastrointest Endosc 8:128–142. https://doi.org/10.4253/wjge.v8. 24:22–27. https://doi.org/10.1111/j.1443-1661.2012.01253.x i3.128 45. Kitajima Y, Ohara H, Nakazawa T et al (2007) Usefulness of 59. Liu SL, Song ZF, Hu QG et al (2010) Serum carbohydrate antigen transpapillary bile duct brushing cytology and forceps biopsy for (CA) 19-9 as a prognostic factor in cholangiocarcinoma: a meta- improved diagnosis in patients with biliary strictures. J analysis. Front Med China 4:457–462. https://doi.org/10.1007/ Gastroenterol Hepatol 22:1615–1620. https://doi.org/10.1111/j. s11684-010-0240-1 1440-1746.2007.05037.x 60. Patel AH, Harnois DM, Klee GG et al (2000) The utility of CA 19- 46. Lin LF, Siauw CP, Ho KS, Tung JN (2003) Guidewire technique 9 in the diagnoses of cholangiocarcinoma in patients without pri- for endoscopic transpapillary procurement of bile duct biopsy mary sclerosing cholangitis. Am J Gastroenterol 95:204–207. specimens without endoscopic sphincterotomy. Gastrointest https://doi.org/10.1016/S0002-9270(99)00744-3 Endosc 58:272–274. https://doi.org/10.1067/mge.2003.329 61. Chapman MH, Sandanayake NS, Andreola F et al (2011) 47. Chen YK, Pleskow DK (2007) SpyGlass single-operator peroral Circulating CYFRA 21-1 is a specific diagnostic and prognostic cholangiopancreatoscopy system for the diagnosis and therapy of biomarker in biliary tract cancer. J Clin Exp Hepatol 1:6–12. bile-duct disorders: a clinical feasibility study (with video){a fig- https://doi.org/10.1016/S0973-6883(11)60110-2 ure is presented}. Gastrointest Endosc 65:832–841. https://doi. 62. Van Gulik TM, Kloek JJ, Ruys AT et al (2011) Multidisciplinary org/10.1016/j.gie.2007.01.025 management of hilar cholangiocarcinoma (Klatskin tumor): Langenbecks Arch Surg (2018) 403:289–307 305 extended resection is associated with improved survival. Eur J external validation of a prognostic nomogram. Ann Oncol 26: 1930–19 Surg Oncol 37:65–71. https://doi.org/10.1016/j.ejso.2010.11.008 35. https://doi.org/10.1093/annonc/mdv279 63. Chaib E, Kanas AF, Galvão FHF, D’Albuquerque LAC (2014) 79. Buettner S, van Vugt JLA, Gani F et al (2016) A comparison of Bile duct confluence: anatomic variations and its classification. prognostic schemes for perihilar cholangiocarcinoma. J Surg Radiol Anat 36:105–109. https://doi.org/10.1007/s00276- Gastrointest Surg 20:1716–1724. https://doi.org/10.1007/s11605- 013-1157-6 016-3203-2 64. Van Gulik TM, Dinant S, Busch ORC et al (2007) Original article: 80. Groot Koerkamp B, Wiggers JK, Allen PJ et al (2014) American new surgical approaches to the Klatskin tumour. Aliment joint committee on cancer staging for resected perihilar cholangio- Pharmacol Ther 26(Suppl 2):127–132. https://doi.org/10.1111/j. carcinoma: a comparison of the 6th and 7th editions. HPB 16: 1365-2036.2007.03485.x 1074–1082. https://doi.org/10.1111/hpb.12320 65. Wiggers JK, te Riele WW, van Dongen TH et al (2016) Combined 81. Deoliveira ML, Schulick RD, Nimura Y et al (2011) New staging liver and extrahepatic bile duct resection for biliary invasion of system and a registry for perihilar cholangiocarcinoma. colorectal metastasis: a case-cohort analysis and systematic re- Hepatology 53:1363–1371. https://doi.org/10.1002/hep.24227 view. HepatoBiliary Surg Nutr 5(4):350–357. https://doi.org/10. 82. Ding G, Yang Y, Cao L et al (2015) A modified Jarnagin-Blumgart 21037/hbsn.2016.05.01 classification better predicts survival for resectable hilar cholan- 66. Olthof SC, Othman A, Clasen S et al (2016) Imaging of cholan- giocarcinoma. World J Surg Oncol 13:1–9. https://doi.org/10. giocarcinoma. Visc Med 32:402–410. https://doi.org/10.1159/ 1186/s12957-015-0526-5 83. Nagino M (2011) Perihilar cholangiocarcinoma: a much needed 67. Blackbourne L, Earnhardt R, Sistrom C et al (1994) The sensitiv- but imperfect new staging system. Nat Rev Gastroenterol Hepatol ity and role of ultrasound in the evaluation of biliary obstruction. 8:252–253. https://doi.org/10.1038/nrgastro.2011.67 Am Surg 60:683–690 84. Steyerberg EW, Vergouwe Y (2014) Towards better clinical pre- 68. Ruys AT, Busch OR, Rauws EA et al (2013) Prognostic impact of diction models: seven steps for development and an ABCD for preoperative imaging parameters on resectability of hilar cholan- validation. Eur Heart J 35:1925–1931. https://doi.org/10.1093/ giocarcinoma. HPB Surg 2013:657309. https://doi.org/10.1155/ eurheartj/ehu207 2013/657309 85. Iacono C, Ruzzenente A, Campagnaro T, Bortolasi L, 69. Engelbrecht MR, Katz SS, Van Gulik TM et al (2015) Imaging of Valdegamberi A, Guglielmi A (2013) Role of preoperative biliary perihilar cholangiocarcinoma. Am J Roentgenol 204:782–791. drainage in jaundiced patients who are candidates for https://doi.org/10.2214/AJR.14.12830 pancreatoduodenectomy or hepatic role of preoperative biliary 70. Mar WA, Shon AM, Lu Y et al (2016) Imaging spectrum of chol- drainage in jaundiced patients who arecandidates for angiocarcinoma: role in diagnosis, staging, and posttreatment pancreatoduodenectomy or hepatic resection: highlights and evaluation. Abdom Radiol 41:553–567. https://doi.org/10.1007/ drawbacks. Ann Surg 257(2):191–204. https://doi.org/10.1097/ s00261-015-0583-9 SLA.0b013e31826f4b0e 71. Choi YH, Lee JM, Lee JY et al (2008) Biliary malignancy: value 86. Ten Hoopen-Neumann H, Gerhards MF, Van Gulik TM et al of arterial, pancreatic, and hepatic phase imaging with (1999) Occurrence of implantation metastases after resection of multidetector-row computed tomography. J Comput Assist klatskin tumors. Dig Surg 16:209–213. https://doi.org/10.1159/ Tomogr 32:362–368. https://doi.org/10.1097/RCT. 0b013e318126c134 87. Martel G, Cieslak KP, Huang R et al (2015) Comparison of tech- 72. Hee SP, Lee JM, Choi JY et al (2008) Preoperative evaluation of niques for volumetric analysis of the future liver remnant: impli- b i l e du ct ca nc e r : M R I comb ine d w i t h M R cations for major hepatic resections. HPB 17:1051–1057. https:// cholangiopancreatography versus MDCT with direct cholangiog- doi.org/10.1111/hpb.12480 raphy. Am J Roentgenol 190:396–405. https://doi.org/10.2214/ 88. Wiggers JK, Koerkamp BG, Cieslak KP et al (2016) AJR.07.2310 Postoperative mortality after liver resection for perihilar 73. Yeh BM, Liu PS, Soto JA et al (2009) MR imaging and CT of the cholangiocarcinoma: development of a risk score and im- biliary tract. Radiographics 29:1669–1688. https://doi.org/10. portance of biliary drainage of the future liver remnant. J 1148/rg.296095514 Am Coll Surg 223:321–331.e1. https://doi.org/10.1016/j. 74. Ruys AT, Bennink RJ, Van Westreenen HL et al (2011) FDG- jamcollsurg.2016.03.035 positron emission tomography/computed tomography and stan- 89. Hammond JS, Guha IN, Beckingham IJ, Lobo DN (2011) dardized uptake value in the primary diagnosis and staging of hilar Prediction, prevention and management of postresection liver fail- cholangiocarcinoma. HPB 13:256–262. https://doi.org/10.1111/j. ure. Br J Surg 98:1188–1200. https://doi.org/10.1002/bjs.7630 1477-2574.2010.00280.x 90. Van Den Broek E, Dijkstra MJJ, Krijgsman O et al (2015) High 75. Annunziata S, Pizzuto D, Caldarella C et al (2014) Diagnostic prevalence and clinical relevance of genes affected by chromo- accuracy of Fluorine-18-Fluorodeoxyglucose positron emission somal breaks in colorectal cancer. PLoS One 10:1–14. https:// tomography in the evaluation of the primary tumor in patients with doi.org/10.1371/journal.pone.0138141 cholangiocarcinoma: a meta-analysis. Eur J Nucl Med Mol 91. Cieslak KP, Bennink RJ, de Graaf W et al (2016) Measurement of Imaging 41:S488. https://doi.org/10.1007/s00259-014-2901-9 liver function using hepatobiliary scintigraphy improves risk as- 76. Chaiteerakij R, Harmsen WS, Marrero CR et al (2015) A new sessment in patients undergoing major liver resection. HPB 18: clinically based staging system for perihilar cholangiocarcinoma. 773–780. https://doi.org/10.1016/j.hpb.2016.06.006 Am J Gastroenterol 2014:1881–1890. https://doi.org/10.1038/ajg. 92. Cieslak KP, Runge JH, Heger M et al (2014) New perspectives in 2014.327.A the assessment of future remnant liver. Dig Surg 31:255–268. 77. Zaydfudim VM, Clark CJ, Kendrick ML et al (2013) Correlation https://doi.org/10.1159/000364836 of staging systems to survival in patients with resected hilar chol- 93. Shoup M, Gonen M, D’Angelica M et al (2003) Volumetric anal- angiocarcinoma. Am J Surg 206:159–165. https://doi.org/10. ysis predicts hepatic dysfunction in patients undergoing major 1016/j.amjsurg.2012.11.020 liver resection. J Gastrointest Surg 7:325–330. https://doi.org/10. 78. Groot Koerkamp B, Wiggers JK, Gonen M et al (2015) Survival 1016/S1091-255X(02)00370-0 after resection of perihilar cholangiocarcinoma-development and 306 Langenbecks Arch Surg (2018) 403:289–307 94. Vauthey JN, Abdalla EK, Doherty DA et al (2002) Body surface Endosc Other Interv Tech 31:422–429. https://doi.org/10.1007/ area and body weight predict total liver volume in western adults. s00464-016-4993-8 Liver Transpl 8:233–240. https://doi.org/10.1053/jlts.2002.31654 111. Fang Y, Gurusamy KS, Wang Q et al (2013) Meta-analysis of 95. Ribero D, Chun YS, Vauthey JN (2008) Standardized liver randomized clinical trials on safety and efficacy of biliary drainage volumetry for portal vein embolization. Semin Intervent Radiol before surgery for obstructive jaundice. Br J Surg. 100:1589– 25:104–109. https://doi.org/10.1055/s-2008-1076681 1596. https://doi.org/10.1002/bjs.9260 96. Dinant S, de Graaf W, Verwer BJ et al (2007) Risk assessment of 112. Celotti A, Solaini L, Montori G et al (2017) Preoperative biliary posthepatectomy liver failure using hepatobiliary scintigraphy and drainage in hilar cholangiocarcinoma: systematic review and me- CT volumetry. J Nucl Med 48:685–692. https://doi.org/10.2967/ ta-analysis. Eur J Surg Oncol 43:4–11. https://doi.org/10.1016/j. jnumed.106.038430 ejso.2017.04.001 97. de Graaf W, van Lienden KP, Dinant S et al (2010) Assessment of 113. Wiggers JK, Coelen RJS, Rauws EAJ et al (2015) Preoperative future remnant liver function using hepatobiliary scintigraphy in endoscopic versus percutaneous transhepatic biliary drainage in patients undergoing major liver resection. J Gastrointest Surg 14: potentially resectable perihilar cholangiocarcinoma 369–378. https://doi.org/10.1007/s11605-009-1085-2 (DRAINAGE trial): design and rationale of a randomized con- 98. Bennink RJ, Dinant S, Erdogan D et al (2004) Preoperative as- trolled trial. Bmc gastroenterol 14:1–8. https://doi.org/10.1186/ sessment of postoperative remnant liver function using s12876-015-0251-0 hepatobiliary scintigraphy. J Nucl Med 45:965–971 114. Komaya K, Ebata T, Yokoyama Y et al (2017) Verification of the 99. de Graaf W, van Lienden KP, van Gulik TM, Bennink RJ (2010) oncologic inferiority of percutaneous biliary drainage to endo- scopic drainage: a propensity score matching analysis of resectable 99mTc-Mebrofenin hepatobiliary scintigraphy with SPECT for the assessment of hepatic function and liver functional volume perihilar cholangiocarcinoma. Surgery (United States) 161:394– before partial hepatectomy. J Nucl Med 51:229–236. https://doi. 404. https://doi.org/10.1016/j.surg.2016.08.008 org/10.2967/jnumed.109.069724 115. Nimura Y (2008) Preoperative biliary drainage before resection 100. Du Bois D, Du Bois E (1989) A formula to estimate the approx- for cholangiocarcinoma (pro). HPB 10:130–133. https://doi.org/ imate surface area if height and weight be known. Nutrition 5: 10.1080/13651820801992666 303–311 116. Kawakami H, Kuwatani M, Onodera M et al (2011) Endoscopic nasobiliary drainage is the most suitable preoperative biliary drain- 101. Erdogan D, Heijnen BHM, Bennink RJ et al (2004) Preoperative assessment of liver function: a comparison of 99mTc-Mebrofenin age method in the management of patients with hilar cholangio- scintigraphy with indocyanine green clearance test. Liver Int 24: carcinoma. J Gastroenterol 46:242–248. https://doi.org/10.1007/ 117–123. https://doi.org/10.1111/j.1478-3231.2004.0901.x s00535-010-0298-1 102. De Graaf W, Häusler S, Heger M et al (2011) Transporters in- 117. Kawashima H, Itoh A, Ohno E et al (2012) Preoperative endo- scopic nasobiliary drainage in 164 consecutive patients with volved in the hepatic uptake of 99mTc-mebrofenin and indocya- nine green. J Hepatol 54:738–745. https://doi.org/10.1016/j.jhep. suspected perihilar cholangiocarcinoma. Ann Surg 257:1. https:// 2010.07.047 doi.org/10.1097/SLA.0b013e318262b2e9 103. Schaap FG, van der Gaag NA, Gouma DJ, Jansen PLM (2009) 118. Farges O, Regimbeau JM, Fuks D et al (2013) Multicentre High expression of the bile salt-homeostatic hormone fibroblast European study of preoperative biliary drainage for hilar cholan- growth factor 19 in the liver of patients with extrahepatic chole- giocarcinoma. Br J Surg 100:274–283. https://doi.org/10.1002/ stasis. Hepatology 49:1228–1235. https://doi.org/10.1002/hep. bjs.8950 22771 119. De Graaf W, Van Lienden KP, Van Den Esschert JW et al (2011) 104. Keppler D (2014) Special section on transporters in toxicity and Increase in future remnant liver function after preoperative portal disease—minireview the roles of MRP2, MRP3, OATP1B1 and vein embolization. Br J Surg 98:825–834. https://doi.org/10.1002/ OATP1B3 in conjugated hyperbilirubinemia. Drug Metab Dispos bjs.7456 42(4):561–565. https://doi.org/10.1124/dmd.113.055772 120. Van Gulik TM, Van Den Esschert JW, De Graaf W et al (2009) 105. Gaag NA, Van Der KJJ, de Castro SMM (2009) Preoperative Controversies in the use of portal vein embolization. Dig Surg 25: biliary drainage in patients with obstructive jaundice: history and 436–444. https://doi.org/10.1159/000184735 current status. J Gastrointest Surg 13:814–820. https://doi.org/10. 121. Olthof PB, Wiggers JK, Groot Koerkamp B et al (2017) 1007/s11605-008-0618-4 Postoperative liver failure risk score: identifying patients with 106. Kennedy TJ, Yopp A, Qin Y et al (2009) Role of preoperative resectable perihilar cholangiocarcinoma who can benefit from biliary drainage of liver remnant prior to extended liver resection portal vein embolization. J Am Coll Surg 225(3):387–394. for hilar cholangiocarcinoma. HPB (Oxford) 11:445–451. https:// https://doi.org/10.1016/j.jamcollsurg.2017.06.007 doi.org/10.1111/j.1477-2574.2009.00090.x 122. Hayashi S, Baba Y, Ueno K et al (2007) Acceleration of primary 107. Al Mahjoub A, Menahem B, Fohlen A et al (2017) Preoperative liver tumor growth rate in embolized hepatic lobe after portal vein biliary drainage in patients with resectable perihilar cholangiocar- embolization. Acta Radiol 48:721–727. https://doi.org/10.1080/ cinoma: is percutaneous transhepatic biliary drainage safer and 02841850701424514 more effective than endoscopic biliary drainage? A meta-analysis. 123. Glantzounis GK, Tokidis E, Basourakos SP et al (2017) The role J Vasc Interv Radiol 28:576–582. https://doi.org/10.1016/j.jvir. of portal vein embolization in the surgical management of primary 2016.12.1218 hepatobiliary cancers. A systematic review. Eur J Surg Oncol 43: 108. Wiggers JK, Koerkamp BG, Coelen RJ et al (2015) Preoperative 32–41. https://doi.org/10.1016/j.ejso.2016.05.026 biliary drainage in perihilar cholangiocarcinoma: identifying pa- 124. Huisman F, Cieslak KP, Van Lienden KP, Bennink RJ, Van Gulik tients who require percutaneous drainage after failed endoscopic TM (2017) Liver related complications in unresectable disease drainage. Endoscopy 47:1124–1131 after portal vein embolization. Hepatobiliary Surg Nutr 6:379– 109. Farges O, Regimbeau JM, Fuks D et al (2013) Multicentre 386. European study of preoperative biliary drainage for hilar cholan- 125. Wiggers JK, Groot Koerkamp B, Coelen RJ et al (2015) giocarcinoma. Br J Surg. 100:274–283. https://doi.org/10.1002/ Percutaneous preoperative biliary drainage for resectable perihilar bjs.8950 cholangiocarcinoma: no association with survival and no increase 110. Jo JH, Chung MJ, Han DH et al (2017) Best options for preoper- in seeding metastases. Ann Surg Oncol 22:1156–1163. https://doi. ative biliary drainage in patients with Klatskin tumors. Surg org/10.1245/s10434-015-4676-z Langenbecks Arch Surg (2018) 403:289–307 307 126. Heimbach JK, Sanchez W, Rosen CB, Gores GJ (2011) Trans- intent resection. J Am Coll Surg 8:583–592. https://doi.org/10. peritoneal fine needle aspiration biopsy of hilar cholangiocarcino- 1002/aur.1474.Replication ma is associated with disease dissemination. HPB 13:356–360. 143. Van Riel WG, Van Golen RF, Reiniers MJ et al (2016) How much https://doi.org/10.1111/j.1477-2574.2011.00298.x ischemia can the liver tolerate during resection? Hepatobiliary 127. Grendar J, Grendarova P, Sinha R, Dixon E (2014) Neoadjuvant Surg Nutr 5(1):58–71. https://doi.org/10.3978/j.issn.2304-3881. therapy for downstaging of locally advanced hilar cholangiocarci- 2015.07.05 noma: a systematic review. HPB 16:297–303. https://doi.org/10. 144. van de Kerkhove MP, de Jong KP, Rijken AM et al (2003) MARS 1111/hpb.12150 treatment in posthepatectomy liver failure. Liver Int 23:44–51 128. Nishio H, Nagino M, Nimura Y (2005) Surgical management of 145. Rocha FG, Matsuo K, Blumgart LH, Jarnagin WR (2010) Hilar hilar cholangiocarcinoma: the Nagoya experience. HPB 7:259– cholangiocarcinoma: the Memorial Sloan-Kettering Cancer 262. https://doi.org/10.1080/13651820500373010 Center experience. J Hepatobiliary Pancreat Sci 17:490–496. 129. Van Gulik TM, Ruys AT, Busch ORC et al (2011) Extent of liver https://doi.org/10.1007/s00534-009-0205-4 resection for hilar cholangiocarcinoma (klatskin tumor): how 146. Jarnagin WR, Fong Y, DeMatteo RP et al (2001) Staging, resect- much is enough? Dig Surg 28:141–147. https://doi.org/10.1159/ ability, and outcome in 225 patients with hilar cholangiocarcino- ma. Ann Surg 234:507–517–9. https://doi.org/10.1097/00000658- 130. Olthof PB, Coelen RJS, Wiggers JK et al (2016) External biliary 200110000-00010 drainage following major liver resection for perihilar cholangio- 147. De Jong MC, Marques H, Clary BM et al (2012) The impact of carcinoma: impact on development of liver failure and biliary portal vein resection on outcomes for hilar cholangiocarcinoma: a leakage. HPB 18:1–6. https://doi.org/10.1016/j.hpb.2015.11.007 multi-institutional analysis of 305 cases. Cancer 118(19):4737– 131. Neuhaus P, Jonas S, Bechstein WO et al (1999) Extended resec- 4747. https://doi.org/10.1002/cncr.27492 tions for hilar cholangiocarcinoma. Ann Surg 230:808–818 dis- 148. Koch M, Garden OJ, Padbury R et al (2011) Bile leakage after cussion 819 hepatobiliary and pancreatic surgery: a definition and grading of 132. Neuhaus P, Thelen A, Jonas S et al (2012) Oncological superiority severity by the International Study Group of Liver Surgery. of hilar en bloc resection for the treatment of hilar cholangiocar- Surgery 149:680–688. https://doi.org/10.1016/j.surg.2010.12.002 cinoma. Ann Surg Oncol 19:1602–1608. https://doi.org/10.1245/ 149. Correa-Gallego C, Maddalo D, Doussot A et al (2016) Circulating s10434-011-2077-5 plasma levels of MicroRNA-21 and MicroRNA-221 are potential 133. Nagino M, Nimura Y, Nishio H et al (2010) Hepatectomy with diagnostic markers for primary intrahepatic cholangiocarcinoma. simultaneous resection of the portal vein and hepatic artery for PLoS One 11:1–16. https://doi.org/10.1371/journal.pone.0163699 advanced perihilar cholangiocarcinoma: an audit of 50 consecu- 150. Feller SM, Lewitzky M (2016) Hunting for the ultimate liquid tive cases. Ann Surg 252:115–123. https://doi.org/10.1097/SLA. cancer biopsy—let the TEP dance begin. Cell Commun Signal 0b013e3181e463a7 14:24. https://doi.org/10.1186/s12964-016-0147-9 134. Bhangui P, Salloum C, Lim C et al (2014) Portal vein 151. Dekker AM, Wiggers JK, Coelen RJ et al (2016) Perioperative arterialization: a salvage procedure for a totally de-arterialized blood transfusion is not associated with overall survival or time to liver. The Paul Brousse Hospital experience. HPB 16:723–738. recurrence after resection of perihilar cholangiocarcinoma. HPB https://doi.org/10.1111/hpb.12200 18:262–270. https://doi.org/10.1016/j.hpb.2015.08.004 135. Dinant S, Gerhards MF, Busch ORC et al (2005) The importance of complete excision of the caudate lobe in resection of hilar chol- 152. Schnitzbauer AA, Lang SA, Goessmann H et al (2012) Right angiocarcinoma. HPB (Oxford) 7:263–267. https://doi.org/10. portal vein ligation combined with in situ splitting induces rapid 1080/13651820500372376 left lateral liver lobe hypertrophy enabling 2-staged extended right 136. Uesaka K (2012) Left hepatectomy or left trisectionectomy with hepatic resection in small-for-size settings. Ann Surg 255:405– resection of the caudate lobe and extrahepatic bile duct for hilar 414. https://doi.org/10.1097/SLA.0b013e31824856f5 cholangiocarcinoma (with video). J Hepatobiliary Pancreat Sci 19: 153. Gores GJ, Murad SD, Heimbach JK, Rosen CB (2013) Liver 195–202. https://doi.org/10.1007/s00534-011-0474-6 transplantation for perihilar cholangiocarcinoma. Dig Dis 31: 137. Bismuth H (1982) Surgical anatomy and anatomical surgery of the 126–129. https://doi.org/10.1159/000347207 liver. World J Surg 6:3–9. https://doi.org/10.1007/BF01656368 154. Skipworth JR, Olde Damink SWM, Imber C et al (2011) Review 138. Oldhafer KJ, Stavrou GA, van Gulik TM (2016) ALPPS—where article: surgical, neo-adjuvant and adjuvant management strategies do we stand, where do we go? Ann Surg 263:1. https://doi.org/10. in biliary tract cancer. Aliment Pharmacol Ther 34:1063–1078. 1097/SLA.0000000000001633 https://doi.org/10.1111/j.1365-2036.2011.04851.x 139. Lang H, de Santibanes E, Clavien PA (2017) Outcome of ALPPS 155. Bird N, Elmasry M, Jones R et al (2017) Role of staging laparos- for perihilar cholangiocarcinoma: case-control analysis including copy in the stratification of patients with perihilar cholangiocarci- the first series from the international ALPPS registry. HPB 19: noma. Br J Surg 104:418–425. https://doi.org/10.1002/bjs.10399 379–380. https://doi.org/10.1016/j.hpb.2017.01.024 156. Van Lienden KP, Van Den Esschert JW, De Graaf W et al (2013) 140. Olthof PB, Coelen RJS, Wiggers JK et al (2017) High mortality Portal vein embolization before liver resection: a systematic re- after ALPPS for perihilar cholangiocarcinoma: case-control anal- view. Cardiovasc Intervent Radiol 36:25–34. https://doi.org/10. ysis including the first series from the international ALPPS regis- 1007/s00270-012-0440-y try. HPB 19:381–387. https://doi.org/10.1016/j.hpb.2016.10.008 157. Dumitrascu T, Brasoveanu V, Stroescu C, Ionescu M, Popescu I 141. Li J, Ewald F, Gulati A, Nashan B (2016) Associating liver parti- (2015) Major hepatectomies for perihilar cholangiocarcinoma: tion and portal vein ligation for staged hepatectomy: from techni- predictors for clinically relevant postoperative complications cal evolution to oncological benefit. World J Gastrointest Surg 8: using the international study Group of Liver Surgery definitions. 124–133. https://doi.org/10.4240/wjgs.v8.i2.124 Asian J Surg 39(2):1–9. https://doi.org/10.1016/j.asjsur.2015.04. 142. Groot Koerkamp B, Wiggers JK, Allen PJ et al (2016) Recurrence rate and pattern of perihilar cholangiocarcinoma after curative

Journal

Langenbeck's Archives of SurgerySpringer Journals

Published: Jan 19, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off