NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Holzheimer RG, Mannick JA, editors. Surgical Treatment: Evidence-Based and Problem-Oriented. Munich: Zuckschwerdt; 2001.

Cover of Surgical Treatment

Surgical Treatment: Evidence-Based and Problem-Oriented.

Show details

Klatskin tumors

, M.D.

Author Information

Carcinoma of the hepatic duct bifurcation was first described by Altemeir in 1957. Subsequently, a series of 13 patients with this diagnosis was reported by Klatskin in 1965. Since that time, cholangiocarcinomas at this location have carried the eponym of Klatskin tumors.

Definition, pathogenesis and epidemiology

Cancer of the bile duct, cholangiocarcinoma, can occur anywhere along the biliary tract from the ampulla of Vater to the intrahepatic biliary radicals. The hepatic duct bifurcation is the most frequently involved site accounting for approximately 60–80% of cholangiocarcinomas.

The etiology of cholangiocarcinoma has not been clearly defined. A number of pathologic conditions, however, resulting in either acute or chronic biliary tract epithelial injury may predispose to malignant change. Primary sclerosing cholangitis, an idiopathic inflammatory condition of the biliary tree, has been clearly associated with the development of cholangiocarcinoma in up to 40% of patients. Congenital biliary cystic disease, such as choledochal cysts or Caroli's disease, has also been associated with malignant transformation in up to 25% of cases. These conditions appear to be related to an anomalous pancreatico-biliary duct junction and, perhaps, are related to the reflux of pancreatic secretions into the bile duct. Chronic biliary tract parasitic infection, seen commonly in Southeast Asia due to Clonorchis sinensis and Opisthorchis viverrini, has also been identified as a risk factor. Although gallstones and cholecystectomy are not thought to be associated with an increased incidence of cholangiocarcinoma, hepatolithiasis and choledocholithiasis may predispose to malignant change. Finally, industrial exposure to asbestos and nitrosamines, and the use of the radiologic contrast agent, Thorotrast (thorium dioxide), are considered to be risk factors for the development of cholangiocarcinoma.

Approximately 15,000 new cases of liver and biliary tract carcinoma are diagnosed annually in the United States, with roughly 10% of these cases being Klatskin tumors. Cholangiocarcinoma accounts for approximately 2% of all cancer diagnoses, with an overall incidence of 1.2/100,000 individuals. Two-thirds of cases occur in patients over the age of 65, with a near ten-fold increase in patients over 80 years of age. The incidence is similar in both men and women.

Diagnosis and Staging

Jaundice is a presenting complaint in almost all cases of Klatskin tumors. Non-specific complaints, such as weight loss and abdominal pain, may also occur. Cholangitis is an uncommon symptom at the time of presentation of cholangiocarcinoma, but frequently occurs after endoscopic or percutaneous biliary tract instrumentation.

The evaluation of the jaundiced patient should begin with either an abdominal ultrasound or computed tomography (CT) scan. Findings on either study suggestive of Klatskin tumor include a dilated intrahepatic biliary tree, a normal or collapsed gallbladder and extrahepatic biliary tree, and a normal pancreas. Contrast enhanced spiral CT scanning provides highly accurate technique for imaging small lesions of biliary tree and clarifying the relationship of the tumor and portal venous and arterial structures. More recently, the use of magnetic resonance cholangiopancreatography (MRCP) has been able to clearly identify the site and nature of biliary obstruction in a noninvasive manner.

After biliary ductal demonstration has been demonstrated by imaging studies, it is imperative for proximal biliary tumors that cholangiography be performed. Cholangiography should define not only the location, but also visualize the uppermost extent of the tumor to determine the limits of resectability. Percutaneous transhepatic cholangiography (PTC) is favored over endoscopic retrograde cholangiography (ERC) in that it better defines the proximal extent of the tumor involvement and may allow placement of percutaneous transhepatic biliary catheters. In most patients with a recent Klatskin tumor, both the right and left hepatic ducts must be intubated to establish adequate biliary drainage.

The use of biliary catheters in patients with Klatskin tumors is somewhat controversial. The routine use of preoperative biliary stenting has been shown by randomized prospective studies to not improve surgical outcome in jaundiced patients prior to biliary surgery. Patients with significant malnutrition, biliary sepsis, or medical problems such as renal failure or in patients who may encounter a delay prior to surgical intervention, however, may benefit from biliary drainage to allow relief of biliary obstruction prior to the planned surgical procedure. Furthermore, the preoperatively-placed biliary catheters are used primarily to assist in the identification of dissection of the hepatic duct bifurcation and the individual right and left hepatic ducts. In addition, in those operative techniques in which large bore silastic stents are used for reconstruction, the biliary catheter greatly facilitate their placement. These catheters, however, may be associated with direct complications such as bleeding and infection. Furthermore, a recent report from the Memorial Sloan-Kettering Cancer Center, has shown that the use of biliary stents in patients undergoing surgical procedures was associated with a significant increase in postoperative infectious complications.

A preoperative histologic diagnosis of a Klatskin tumor can be extremely difficult to obtain. Although percutaneous biopsies and brushings for cytology are valuable if positive for malignancy, they are generally associated with a relatively low sensitivity (< 50%). In general, a tissue diagnosis is not necessary prior to proceeding with surgical intervention. However, a tissue diagnosis is important to differentiate a stricture due to primary sclerosing cholangitis from hilar cholangiocarcinoma in patients who might otherwise be candidates for liver transplantation.

The preoperative staging, prior to consideration of surgical resection, includes the use of contrast-enhanced CT scanning and cholangiography. Findings on CT scan, such as extra hepatic disease or bilobar liver metastases, preclude curative resection. Cholangiographic appearance of extensive bilobar proximal extension into the hepatic parenchyma also indicates unresectability. Finally, involvement of the major hepatic portal vessels (hepatic artery and portal vein) with tumor encasement or occlusion will preclude resection. Spiral CT scanning with bolus contrast, duplex ultrasonography or magnetic resonance imaging (MRI) are non-invasive tools for assessing tumor extent and invasion of the major portal vascular structures. If the status of major portal vascular structures cannot be determined, hepatic angiography with venous phase studies remains the gold standard for determining major vessel involvement. More recently, the routine use of staging laparoscopy prior to laparotomy has been suggested in order to minimally invasively assess patients for evidence of liver metastases or peritoneal implants.

Surgical Treatment

At completion of preoperative staging, approximately two-thirds of patients with Klatskin tumors will be considered operable. Resectability is further assessed at operation by careful exploration for peritoneal implants, hepatic metastasis and regional lymph node involvement. The goal of any surgical procedure is to remove all gross and microscopic disease and to relieve biliary obstruction. The standard surgical procedure includes resection of the extrahepatic bile duct from the duodenum to above the level of the hepatic confluence, cholecystectomy, portal lymphadenectomy and biliary-enteric anastomosis. Controversy, however, exists with respect to the performance of routine, hepatic lobectomy and caudate lobectomy. Unfortunately, there have been no prospective, randomized trials to address this controversy.

The standard operative approach at the Johns Hopkins Hospital for bifurcation tumors is limited resection to the hepatic duct bifurcation with reconstruction as a bilateral hepaticojejunostomy with the routine use of transhepatic silastic stents. At our Institution, hepatic lobectomy has been added to a hilar resection, in cases where proximal tumor involvement extended into the hepatic parenchyma. In the most recent review of the Hopkins experience in 1996, 196 patients underwent surgical exploration for Klatskin tumors. 109 patients (56%) underwent resection, consisting of the excision of the bile duct bifurcation, placement of silastic transhepatic biliary stents and reconstruction with a Roux-en-Y hepaticojejunostomy. Fifteen patients (14%) also had an hepatic lobectomy in addition to the bile duct resection, in an attempt to remove all tumor. In 73 of the resected patients (67%), all gross tumor was removed, while in 36 patients (33%) gross tumor was left behind after resection.

Of the 196 patients with Klatskin tumors undergoing operative exploration, 8 (4.1%) died after surgery. Of the 109 resected patients, 4 patients (3.6%) died. Only one (6.6%) of the 15 patients undergoing hepatic resection for perihilar tumor died after operation. Of the 87 patients with perihilar tumors undergoing palliative procedures, 4 (4.6%) died. The morbidity rates for patients undergoing operative management of hilar cholangiocarcinoma was 47%. The mean length of hospital stay was 23 ± 1.3 days.

In the 109 patients with resected perihilar Klatskin tumors, the one-, three- and five-year survival was 68%, 30% and 11%, respectively. The median survival was 19 months. The addition of a hepatic lobectomy to resection of bile duct did not alter the survival rate (figure 1). The one-, three- and five-year survival rates of the 15 patients who underwent hepatic lobectomy, as well as extra hepatic bile duct resection, were 64%, 50% and 10%, respectively. The median survival was 18 months. In the 94 patients treated with bile duct resection alone, the one-, three- and five-year survival rates were 68%, 26% and 11%, respectively. The median survival was 19 months. However, if a negative microscopic margin was obtained after resection, survival was significantly prolonged (p < 0.05). In the 28 patients with negative microscopic margins, the five-year actuarial survival was 19% as compared to the 9% five-year survival rate for 81 patients with positive margins (figure 2). In addition, the median survival increased to 41 from 18 months. None of the six patients with perihilar lymph node metastasis, who underwent resection of the extra hepatic bile duct, survived past 18 months. In the remaining 39 patients without evidence of lymph node involvement, the five-year survival was 15%. Negative lymph nodes resulted in an increased median survival of 5 to 12 months.

Figure 1. Influence of hepatic resection on survival of patients with resected Klatskin tumors (from Nakeeb et al, 1996).

Figure 1

Influence of hepatic resection on survival of patients with resected Klatskin tumors (from Nakeeb et al, 1996).

Figure 2. Influence of microscopically negative margins on survival of patients with resected Klatskin tumors (from Nakeeb et al, 1996).

Figure 2

Influence of microscopically negative margins on survival of patients with resected Klatskin tumors (from Nakeeb et al, 1996).

The Hopkins’ results, however, appear to compare unfavorably to other retrospective series. In most of these series, liver resection is routinely performed. Utilizing this approach, a negative histologic margin, is generally achieved in over 50% of patients. Typical of this experience is the recent report from the Memorial Sloan-Kettering Cancer Center by Burke and colleagues. In that series, 90 patients with Klatskin tumors were evaluated and 33% underwent potentially curative surgical resection. Twenty-two patients (73%) underwent a combined bile duct resection with partial hepatectomy. The 30-day mortality was 7%. An overall negative histologic margin was obtained in 83% of the patients. However, in patients in whom a hepatic resection was included the incidence of a negative margin was 88%, while only 50% of resections had a negative margin in which a hepatic resection was not performed. The overall median survival was 40 months. The median survival for those patients with a negative histologic margin has not yet been obtained with a mean follow-up of 26 months. A median survival of 22 months was seen in those patients with positive margins. This 5-year survival for the entire group was 45%, but increased to 56% if a negative histologic margin was obtained. Five-year survival was 0% if the margin was positive. Unfortunately, these excellent results have not been universally obtained. This is particularly true with respect to the operative mortality associated with hepatectomy. In a multi-center review by the French Surgical Association (Redding, 1991), the 30-day mortality was 17%. Similarly, Pichlmayr, in a single center experience, reported a 10.5% operative mortality following hepatectomy. It would appear that although hepatectomy increases the probability of negative surgical margin, and therefore a greater chance of cure; an increase in postoperative mortality would be expected.

In addition to partial hepatectomy in the management of Klatskin tumors, some authors have stressed the importance of performing a caudate lobe resection for Klatskin tumors. These authors argue that the ducts draining the caudate lobe entered directly into the bifurcation and are frequently involved with Klatskin tumors.

The final surgical approach to hilar cholangiocarcinoma is even more controversial and involves liver transplantation. At present, these results are confined to anecdotal reports and small series with results no better than for standardized radical resection. Therefore, at present, transplantation should not be considered standard management for this disease, but rather only considered as part of an investigational protocol.

Palliative therapy in patients with Klatskin tumors can be accomplished either operatively or by surgical biliary bypass. If preoperative staging suggests unresectable disease, or if the patient is not a suitable surgical candidate, then a nonoperative approach is indicated. Although endoscopically-placed biliary catheters are preferred for mid- and distal bile duct malignant obstruction, Klatskin tumors usually require intubation of both the left and right ducts to avoid complications of recurrent cholangitis and/or liver abscess. Therefore, percutaneous transhepatic catheters provide the best access for palliation of inoperable hilar cholangiocarcinoma. These catheters can be used for long-term palliation and are readily exchangeable. In the event of catheter occlusion or sludge and associated biliary sepsis, the catheters can be easily externalized and changed. In selective patients, percutaneous catheters can be used for access for placement of internal metallic Wallstents to eliminate the external catheter.

In patients who undergo abdominal exploration for attempted surgical resection but are found to have unresectable tumors, there are two options. In patients with wide-spread intraperitoneal tumor, minimal operative manipulation should be performed. Transhepatic biliary catheters placed preoperatively are left in position for palliation of biliary obstruction. A cholecystectomy can be performed to prevent the development of acute cholecystitis from cystic duct obstruction, either from tumor or the indwelling catheters. Postoperatively, the transhepatic catheter can be exchanged fluoroscopically for larger diameter soft silastic catheter or a Wallstent. In selected patients with locally advanced unresectable cancer at the porta, there are two options: a cholecystectomy could be performed and the bile duct distal to the tumors divided. The malignant stricture can be dilated to place a progressively larger silastic catheters into the right and left ductal systems. A choleduojejunostomy to a Roux-en-Y jejunal limb is completed. Another palliative approach is a biliary enteric bypass to segment III branch of the left hepatic duct. This technique involves identifying the ducts in the superficial location of the umbilical fissure and constructing a long side-to-side anastomosis to a Roux limb. This technique will usually decompress both the right and left systems and can be accomplished without the need for long-term stents.

Adjuvant therapy

At present, there is no effective nonsurgical therapy for cholangiocarcinoma, as either an adjuvant therapy or for unresectable disease. No prospective, randomized trials demonstrating a therapeutic benefit for radiation, chemotherapy or combination therapy have yet been reported. At our Institution, a prospective, non-randomized study was completed in which appropriate patients undergoing surgical resection and/or palliation were treated with both external beam radiation plus the use of Iridium implants placed via percutaneous silastic stents. In that series, postoperative radiation therapy did not improve survival in either resected or palliated patients. The mean survival in the 23 radiated patients was 18.4 ± 2.9 months versus 20.1 ± 2.4 months for patients not receiving radiation therapy. Despite the lack of any apparent benefit, the overall poor prognosis for Klatskin tumors supports continued clinical investigations in the use of adjuvant therapy in both resectable and unresectable patients.


Cholangiocarcinoma represents a rare malignancy. Without intervention, death due to progressive jaundice is inevitable. The goals of therapy, therefore, are resection of all disease and relief of biliary obstruction. A complete resection with a negative histologic margin appears to be associated with improved survival. Although no prospective, randomized studies are available, most recent series strongly support appropriate hepatic resection to achieve a negative margin.


Bismuth H, Nakache R, Diamond T. Management strategies in resection for hilar cholangiocarcinoma. Ann Surg. (1992);215:31–38. [PMC free article: PMC1242367] [PubMed: 1309988]
Burke E, Jarnigan W R, Hochwald S N. et al. Hilar cholangiocarcinoma patterns of spread, the importance of hepatic resection for curative operation, and a presurgical clinical staging system. Ann Surg. (1998);228:385–394. [PMC free article: PMC1191497] [PubMed: 9742921]
Hochwald S N, Burke E, Jarnagin W R, Fong Y, Blumgart L H. Association of preoperative biliary stenting with increased postoperative infectious complications in proximal cholangiocarcinoma. Arch Surg. (1999);134:261–266. [PubMed: 10088565]
Iwatsuki S, Todo S, Marsh J W. et al. Treatment of hilar cholangiocarcinoma (Klatskin tumors) with hepatic resection or transplantation. J Am Coll Surg. (1998);187:358–364. [PMC free article: PMC2991118] [PubMed: 9783781]
Kuvshinoff B W, Armstrong J G, Fong et al. Palliation of irresectable hilar cholangiocarcinoma with biliary drainage and radiotherapy. Br J Surg. (1995);82:1522–1555. [PubMed: 8535808]
Miyazaki M, Ito H, Nakagawa K. et al. Aggressive surgical approaches to hilar cholangiocarcinoma: hepatic or local resection? Surgery. (1998);123:131–136. [PubMed: 9481397]
Nagino M, Nimura Y, Kamiya J. et al. Segmental liver resection for hilar cholangiocarcinoma. Hepatogastroenterology. (1998);45:7–13. [PubMed: 9496478]
Nagorney D M, Donohue J H, Farnell M, Sckleck L D, Ilstrup D M. Outcomes after curative resection of cholangiocarcinoma. Arch Surg. (1993);128:871–879. [PubMed: 8393652]
Nakeeb A, Pitt, HA, Sohn T A. et al. Cholangiocarcinoma: a spectrum of intrahepatic, perihilar, and distal tumors. Ann Surg. (1996);224:463–475. [PMC free article: PMC1235406] [PubMed: 8857851]
Pichlmayr R, Weimann A, Klempnauer J. et al. Surgical treatment in proximal bile duct cancer. A single-center experience. Ann Surg. (1996);224:628–638. [PMC free article: PMC1235440] [PubMed: 8916878]
Pitt H A, Nakeeb A, Ross R A, Coleman et al. Perihilar cholangiocarcinoma. Postoperative radiotherapy does not improve survival. Ann Surg. (1995);221:788–798. [PMC free article: PMC1234714] [PubMed: 7794082]
Copyright © 2001, W. Zuckschwerdt Verlag GmbH.
Bookshelf ID: NBK6906


  • PubReader
  • Print View
  • Cite this Page

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...