Diseases of the Liver and Biliary System - part 10 docx

found in the pancreatic juice of patients with chronic
pancreatitis. This may relate to K-ras mutations in areas
of duct hyperplasia [41]. Over a 2-year follow-up a
minority develop pancreatic carcinoma [40].
Pathology
Histologically, the tumour is an adenocarcinoma,
whether arising from pancreatic duct, acinus or
bile duct. The ampullary tumours have a papillary
arrangement and are often of low-grade malignancy;
fibrosis is prominent. They tend to be polypoid and
soft, whereas the acinar tumours are infiltrative, large
and firm.
Obstruction of common bile duct
This results from direct invasion causing a scirrhous
reaction, from annular stenosis, and from tumour tissue
filling the lumen. The duct may also be compressed by
the tumour mass.
The bile ducts dilate and the gallbladder enlarges. An
ascending cholangitis in the obstructed duct is rare. The
liver shows the changes of cholestasis.
Pancreatic changes
The main pancreatic duct may be obstructed as it enters
the ampulla. The ducts and acini distal to the obstruction
dilate and later rupture, causing focal areas of pancreati-
tis and fat necrosis. Later all the acinar tissue is replaced
by fibrous tissue. Occasionally, particularly in the acinar
type, fat necrosis and suppuration may occur in and
around the pancreas.
Diabetes mellitus or impaired glucose tolerance is
found in 60–80% of patients. Apart from destruction of
insulin-producing cells by the tumour, this may be due

to production of islet amyloid polypeptide (IAPP) by
islet cells adjacent to the tumour [36].
Spread of the tumour
Direct extension in the wall of the bile duct and infiltra-
tion through the head of the pancreas is common with
the acinar although not with the ampullary type. The
second part of the duodenum may be invaded, with
ulceration of the mucosa and secondary haemorrhage.
The splenic and portal veins may be invaded and may
thrombose with resultant splenomegaly.
Involvement of regional nodes is found in approxi-
mately a third of operated cases. Perineural lymphatic
spread is common. Blood-borne metastases, with secon-
daries in liver and lungs, follow invasion of the splenic
or portal veins. There may also be peritoneal and
omental metastases.
Clinical features
Both sexes are affected, but males more frequently than
females in a ratio of 2:1. The sufferer is usually between
50 and 69 years old.
The clinical picture is a composite one of cholestasis
with pancreatic insufficiency, and the general and local
effects of a malignant tumour (fig. 36.1).
Jaundice is of gradual onset, progressively deepening,
but ampullary neoplasms can cause mild and intermit-
tent jaundice. Itching is a common but not invariable
feature, and when present comes after jaundice. Cholan-
gitis is unusual.
Cancer of the head of the pancreas is not always pain-
less. Pain may be experienced in the back, the epigas-

trium and right upper quadrant, usually as a continuous
distress worse at night and sometimes ameliorated by
crouching. It may be aggravated by eating.
Weakness and weight loss are progressive and have
usually continued for at least 3 months before jaundice
develops.
Although frank steatorrhoea is rare, the patient often
complains of a change in bowel habit, usually diarrhoea.
Vomiting and intestinal obstruction follows invasion
of the second part of the duodenum in 15–20% of
patients. Ulceration of the duodenum can erode a vessel
with haematemesis or, more commonly, occult bleeding.
Difficulty in making a diagnosis may make the patient
depressed. It then becomes easy to believe mistakenly,
that the patient is psychoneurotic.
Examination. The patient is jaundiced and shows evi-
dence of recent weight loss. Theoretically, the gallblad-
der should be enlarged and palpable (Courvoisier’s law).
In practice, the gallbladder is only felt in about half the
patients, although at subsequent laparotomy a dilated
gallbladder is found in three-quarters. The liver is
enlarged with a sharp, smooth, firm edge. The pancreatic
tumour is usually impalpable.
The spleen is palpable if involvement of the splenic
vein has caused thrombosis. Peritoneal invasion is fol-
lowed by ascites.
Lymphatic metastases are more usual with cancer
of the body rather than head of the pancreas [51]. Occa-
sionally, however, axillary, cervical and inguinal glands
may be enlarged and Virchow’s gland in the left supra-

clavicular fossa may be palpable.
Sometimes, widespread venous thromboses simulate
thrombophlebitis migrans.
Investigations
Glycosuria occurs in 60–80% and with it there is an
impaired oral glucose tolerance test.
Blood biochemistry. The serum alkaline phosphatase
level is raised. The serum amylase and lipase concentra-
640 Chapter 36
tions are sometimes persistently elevated in carcinoma
of the ampullary region. Hypoproteinaemia with, later,
peripheral oedema may be found.
There is no reliable serum tumour marker with suffi-
cient specificity or sensitivity. Using a cut-off of 70U/ml
(almost twice the upper limit of normal), CA 19-9 has
the greatest sensitivity (around 70%) and specificity
(90%) of current markers for carcinoma of the pancreas
[14]. However its sensitivity in the detection of early
tumour is lower [22]. It can be elevated in benign biliary
obstruction.
Haematology. Anaemia is mild or absent. The leucocyte
count may be normal or raised with a relative increase
in neutrophils. The erythrocyte sedimentation rate is
usually raised.
Differential diagnosis
The diagnosis must be considered in any patient over
40 years with progressive or even intermittent cholesta-
sis. The suspicion is strengthened by persistent or
unexplained abdominal pain, weakness and weight
loss, diarrhoea, glycosuria, positive faecal occult blood,

hepatomegaly, a palpable spleen or thrombophlebitis
migrans.
Radiology
Most scanning techniques can detect pancreatic masses
with high accuracy in specialist units. There are data
supporting transabdominal ultrasound, spiral (helical)
CT, dual-phase (arterial and venous) spiral CT, MRI,
endoscopic ultrasound and PET scanning. The most
effective initial technique is helical, dual-phase CT (see
fig. 32.5) which has an accuracy of over 90% for tumours
greater than 1.5cm and 70% for smaller tumours [24].
This technique is widely available with expertise in
interpretation [19]. MRI with angiography (gadolinium
enhancement) and endoscopic ultrasound have a similar
accuracy in some centres [2, 4, 16], but these techniques
and expertise are not as widely available as spiral CT.
Scanning may also show dilatation of bile ducts and
pancreatic duct, hepatic metastases and local spread of
the primary lesion.
Percutaneous ultrasound or CT-guided fine-needle
aspiration of the pancreatic mass is safe and has a sensi-
tivity of over 90% in some units. There is a small risk of
seeding of tumour cells along the needle track.
ERCP can usually demonstrate the pancreatic and bile
ducts, allow biopsy of any ampullary lesion (fig. 36.2),
and provide bile or pancreatic juice or brushings from
the stricture for cytological examination (fig. 36.3). The
appearance of the bile duct and/or pancreatic duct stric-
ture (double duct sign) gives a good indication of the
underlying malignant cause of the stricture (see fig.

32.12a) but occasionally appearances are deceptive [31]
and tissue diagnosis should be sought. Biliary brush
cytology has a sensitivity of around 60% for pancreatic
carcinoma [23, 46]. Unusual tumours such as lymphoma
need to be identified since they may respond to specific
therapy.
In the patient with vomiting, barium meal and/or
endoscopy will show the extent of duodenal invasion
and obstruction.
Tumour staging
This gives an indication of whether the tumour is
resectable or not. Clinical evidence, chest X ray, ultra-
sound or CT will show obvious metastatic disease.
Dual or triple-phase spiral CT is highly accurate in
predicting irresectability (approximately 90%) [24] but
Diseases of the Ampulla of Vater and Pancreas 641
Fig. 36.2. Abnormal ampulla at ERCP. Note irregular surface
with nodularity. Biopsy showed adenocarcinoma.
Fig. 36.3. Brush cytology taken from a low common bile duct
stricture. There is a sheet of benign biliary epithelial cells and
above this a small group of large polymorphic cells
characteristic of adenocarcinoma.
is less accurate in predicting resectability. Features
suggesting irresectability are local extension of tumour,
encasement of extra-pancreatic arteries or veins, inva-
sion of adjacent organs and lymph node metastases
(more than an isolated solitary node). Most irresectable
lesions (70%) have three or four of these features
—
the

minority having only one or two. Spiral CT may also
show hepatic metastases but the detection rate is higher
when combined with arterioportography.
Endoscopic ultrasound has a similar accuracy as dual
phase helical CT in assessing irresectability, but expertise
is not as widely available [24].
Experience with MRI is growing but it remains second
choice to helical dual-phase CT.
For vascular involvement these techniques can give
the same information as digital subtraction angiography
(DSA) which is now used only when the data from scan-
ning is inconclusive or conflicting.
Laparoscopy is valuable to show and biopsy minute
hepatic metastases and peritoneal and omental seedings.
Negative results from laparoscopy, CT and angiography
correspond to a 78% resectability rate [51].
Prognosis
The prognosis of pancreatic carcinoma is grave. After
biliary bypass surgery the mean survival is about 6
months. The acinar type carries a worse prognosis than
the ductal type because regional lymph glands are
involved earlier. Only the minority of tumours, between
5 and 20%, are resectable.
Resection has had an operative mortality of approxi-
mately 15–20%, but recent reports have shown this to fall
to 5% and less in specialist centres with a few expert
surgeons performing more operations [14]. A recent
report from a superspecialist unit of zero mortality after
145 consecutive pancreatico-duodenal resections is
exceptional [10].

Coincident with reduced operative mortality has been
a rise in reported 5-year survival to around 20%. This
may reflect earlier detection of disease through the
newer scanning techniques, or selection of patients with
less extensive spread of disease. Disease recurrence,
however, remains a problem [14]. Total pancreatectomy
does not lead to longer survival than the less extensive
Whipple’s procedure and produces exocrine insuffi-
ciency and brittle diabetes.
The overall outlook for carcinoma of the pancreas
however, is grim with only 23 of 912 patients with carci-
noma of the pancreas in one series surviving 3 years and
only two of these being considered cures [11].
Prognosis for carcinoma of the ampulla is better. 85%
or more of patients survive 5 years after resection if the
tumour has not spread beyond the margins of the
sphincter of Oddi. With more extensive tumour the
5-year survival falls to 11–35% [15, 52].
Treatment
Resection
A decision to attempt resection depends on the clinical
state of the patient and the staging of the tumour derived
from radiological imaging. Difficulties in removal arise
because of the inaccessibility of the pancreas on the pos-
terior wall of the abdomen in the vicinity of vital struc-
tures. The resectability rate is therefore low.
The classical procedure is pancreatico-duodenectomy
(Whipple’s operation) which is performed in one stage
with removal of related regional lymph nodes, the entire
duodenum and the distal third of the stomach. This

operation was modified in 1978 [49] to preserve antral
and pyloric function (pylorus-preserving pancreatico-
duodenectomy). This reduces post-gastrectomy symp-
toms and marginal ulceration, and improves nutrition.
Survival is the same as those having the classical proce-
dure. The continuity of the biliary passages is restored by
anastomosis of the common bile duct with the jejunum.
Pancreatico-jejunostomy drains the duct of the remain-
ing pancreas. The continuity of the intestinal tract is
restored by duodeno-jejunostomy.
Frozen section examination of the resection margins is
mandatory.
Prognostic factors are tumour size, resection margin
and lymph node status [14]. The best indicator is lymph
node histology. If negative at resection, 5-year survival is
40–50%, compared with 8% in those with nodes positive
for metastasis [9]. Prognosis is also related to whether or
not there is histological evidence of vascular invasion
(median survival 11 vs. 39 months).
Carcinoma of the ampulla is also treated by pancre-
atico-duodenectomy. Local resection (ampullectomy) is
an alternative in selected patients with premalignant or
malignant ampullary lesions [5]. Endoscopic photody-
namic therapy has produced remission or reduced
tumour bulk in a series of patients with ampullary carci-
noma unsuitable for surgery [1]. This technique uses
endoscopic delivery of red light (630nm) to tumour sen-
sitized with haematoporphyrin given intravenously.
Palliative procedures
The choice lies between surgical bypass and endoscopic

or percutaneous trans-hepatic insertion of an endopros-
thesis (stent).
Palliative surgical biliary bypass is an option in irre-
sectable patients with a predicted longer survival. For
the jaundiced patient with vomiting due to duodenal
642 Chapter 36
obstruction, choledocho-jejunostomy with gastroen-
terostomy is necessary. Gastric outlet and duodenal
stenosis can be treated by endoscopically or radiologi-
cally placed expandable mesh metal stents [33, 43]
although experience with this technique is limited. For
the patient with bile duct obstruction alone, some argue
for prophylactic gastric bypass surgery at the time of
biliary bypass but most would make this decision at the
time of operation, according to the size of the tumour.
Endoscopic stent insertion (fig. 36.4) is successful in up
to 95% of patients (60% after the first session) and has a
lower 30-day mortality than surgical bypass [42]. When
the endoscopic approach fails, the percutaneous, or com-
bined percutaneous/endoscopic approach can be used
(Chapter 32).
Percutaneous stent insertion (see fig. 32.21) has a similar
mortality, early morbidity and mean survival time (19
vs. 15 weeks) to palliative surgery partly due to the com-
plications of the trans-hepatic approach (haemorrhage,
bile leakage) [7]. Endoscopic stent insertion has a lower
complication rate and mortality than the percutaneous
route [44].
Within 3 months of insertion 20–30% of plastic stents
need to be replaced because of obstruction by biliary

sludge. Metal mesh expandable stents can be inserted
endoscopically or percutaneously (see fig. 32.17). They
remain patent significantly longer than plastic stents
(mean 273 vs. 126 days) [12]. However, because of their
cost, they are best restricted to those patients with irre-
sectable peri-ampullary carcinoma who at the time of
first stent exchange because of blockage are judged likely
to have slower progression and a longer survival [35].
This may be predicted to some extent by tumour size
and weight loss [39]. If plastic stents are used, elective
exchange of the stent every 3 months gives patients a
longer complication-free interval than those having stent
exchange only once blockage occurs [38].
The choice between surgical and non-surgical relief of
biliary obstruction depends upon the expertise available
and the clinical status of the patient.
The non-surgical insertion of a stent is particularly
applicable to older, poor-risk patients especially when a
large, clearly inoperable pancreatic mass has been
imaged or where there is extensive metastatic disease.
For the younger patient with irresectable disease, surgi-
cal bypass should still be considered if longer than
average survival is expected.
With all the approaches now available, no patient with
carcinoma of the pancreas should die jaundiced or with
intolerable itching.
Chemotherapy: radiotherapy
Pre-operative adjuvant chemotherapy and radiotherapy
have produced disappointing results. Selected patients
may benefit from adjuvant combined radiotherapy and

chemotherapy after radical resection [20]. For patients
with irresectable tumour many chemotherapeutic
regimes have been studied in randomized trials [14]. 5-
fluorouracil (5-FU) has been used widely. Recent data
suggest that treatment with gemcitabine may have some
benefit on survival and the alleviation of disease-related
symptoms [8]. Ideally all patients should participate in
comparative studies. Otherwise gemcitabine may be
considered for patients with advanced pancreatic car-
cinoma with or without metastases, who are still self-
caring [14, 32]. For localized disease, radiotherapy and
concomitant 5-FU is an alternative [14]. In patients who
are no longer self-caring, management should focus on
palliative measures rather than aggressive chemoradia-
tion or chemotherapy.
For patients with pain, coeliac plexus block may be
Diseases of the Ampulla of Vater and Pancreas 643
Fig. 36.4. Polyethylene 10 French endoprosthesis inserted
across low common bile duct stricture by the endoscopic route.
Note good flow of contrast into duodenum and decompressed
biliary system.
more effective than oral or parenteral analgesia. Benefit
is more immediate [37]. However, although coeliac
plexus block either done percutaneously under X-ray
screening or at operation can reduce pain for a few
months, pain may return in over half of patients [27].
Benign villous adenoma of the ampulla of Vater
This leads to biliary colic and obstructive jaundice. The
ampullary tumour is seen and biopsied at ERCP.
Dysplasia may be present on biopsy. Carcinoembry-

onic antigen (CEA) and CA19-9 are found on immuno-
histochemistry [53]. These lesions should be regarded as
potentially premalignant. Local resection or pancreatico-
duodenectomy is indicated [5]. In patients unfit for
surgery, stenting is palliative. Endoscopic ablation may
be successful using laser, monopolar or bipolar coagula-
tion, or photodynamic therapy [1, 34].
Cystic tumours of the pancreas [18]
These may be benign or malignant and include cystic
adenocarcinoma, cystic adenoma (serous and mucinous)
and papillary cystic tumours. They may be misdiag-
nosed as pseudocysts. 40% of patients are asymptomatic.
Work-up is by CT, endoscopic ultrasound, angiography
and ERCP. Cyst fluid analysis (cytology, tumour
markers) may be valuable in differentiating the type of
tumour [26]. Around 40% of lesions are malignant. In
general, resection should be attempted. Frozen and even
routine histology may be misleading. Mucinous cystic
neoplasm should be considered potentially malignant.
Endocrine tumours of the pancreas [3, 47]
These include insulinoma (70%), gastrinoma (20%),
vaso-active intestinal polypeptide-secreting tumour
(VIPoma), glucagonoma, polypeptide-secreting tumour
(PPoma) and somatostatinoma. Some may be non-func-
tioning [6]. They present with either the systemic effects
of the hormone released or a mass effect with pain or
jaundice as with pancreatic carcinoma. A variable pro-
portion are malignant, depending on the endocrine type.
Treatment is by surgical resection or debulking, and
medical measures to counter the effect of any hormone

released. Survival depends on the tumour type and
stage.
Chronic pancreatitis
Pancreatitis, usually of alcoholic aetiology, can cause
narrowing of the intra-pancreatic portion of the common
bile duct. The resultant cholestasis may be transient
during exacerbations of acute pancreatitis. It is presum-
ably related to oedema and swelling of the pancreas.
More persistent cholestasis follows encasement of the
intra- and peri-pancreatic bile duct in a progressively
fibrotic pancreatitis. Pseudocysts of the head of the pan-
creas and abscesses can also cause biliary obstruction
and persistent cholestasis.
Bile duct stenosis affects about 8% of patients with
chronic alcoholic pancreatitis and this figure would be
higher if more cholangiograms were done. It should be
suspected if the serum alkaline phosphatase is more than
twice elevated for longer than 1 month. ERCP shows a
smooth narrowing of the lower end of the common bile
duct, sometimes adopting a ‘rat tail’ configuration (fig.
36.5). The main pancreatic duct may be tortuous, irregu-
lar and dilated. Pancreatic calcification may be present.
Liver biopsy shows portal fibrosis, features of biliary
obstruction, and sometimes biliary cirrhosis. Features of
alcoholic liver disease are unusual. Hepatic fibrosis
regresses after biliary decompression [21].
Splenic vein thrombosis is a complication of chronic
pancreatitis.
Management
Early diagnosis of a biliary stricture due to pancreatitis is

essential as biliary cirrhosis and acute cholangitis can
develop in the absence of clinical jaundice.
If alcohol is responsible for the pancreatitis the patient
must abstain completely.
The place of surgery is controversial. Clinical, labora-
tory and imaging data do not necessarily distinguish
644 Chapter 36
Fig. 36.5. ERCP in a patient with alcoholic chronic
pancreatitis. Note the ‘rat tail’ narrowing of the distal common
bile duct (arrow).
those patients with significant bile duct obstruction from
those with alcoholic liver disease or normal liver histol-
ogy [25]. Liver biopsy is valuable in deciding whether
surgical decompression of the bile duct is necessary.
Plastic and metal mesh stents successfully relieve bile
duct obstruction due to chronic pancreatitis [13], but
longer-term data are needed to judge whether this is an
appropriate method of treatment. Acute cholangitis,
biliary cirrhosis and protracted jaundice are strong indi-
cations for surgery [45]. Choledocho-enterostomy is the
usual procedure.
Obstruction of the common bile duct by
enlarged lymph glands
This is rare. The enlarged glands are nearly always
metastatic, frequently from a primary in the alimentary
tract, lung or breast, or from a hepato-cellular carcinoma.
Malignant glands in the porta hepatis are associated
with deep jaundice, the main bile ducts usually being
invaded rather than compressed. Secondary deposits in
the hepatic parenchyma may also invade the bile ducts,

causing segmental obstruction.
Glands along the common duct may be enlarged in
non-malignant conditions, but the bile duct usually
escapes compression. Jaundice in infections such as
tuberculosis, sarcoidosis or infectious mononucleosis is
not obstructive but due to direct hepatic involvement or
haemolysis.
Glandular enlargement in the reticuloses does, very
rarely, cause obstruction to the common bile duct, but
jaundice complicating these diseases is more often
due to hepatic parenchymal involvement, to increased
haemolysis or loss of intra-hepatic bile ducts (Chapter
13).
Other causes of extrinsic pressure on the
common bile duct
Duodenal ulceration
This is an extremely rare cause of obstructive jaundice.
Perforation, so that the ulcer impinges against the bile
duct or causes adhesive peritonitis, may rarely result in
biliary obstruction. This can also follow scarring as the
ulcer heals or endoscopic sclerosis for bleeding.
Duodenal diverticulum
Diverticula of the duodenum are often found near the
ampulla of Vater, but rarely cause obstruction of the bile
duct. They are associated with an increased rate of gall-
bladder and common bile duct stones, and the recur-
rence of duct stones [54].
References
1 Abulafi AM, Allardice JT, Williams NS et al. Photodynamic
therapy for malignant tumours of the ampulla of Vater. Gut

1995; 36: 853.
2 Adamek HE, Albert J, Breer H et al. Pancreatic cancer detec-
tion with magnetic resonance cholangiopancreatography
and endoscopic retrograde cholangiopancreatography: a
prospective controlled study. Lancet 2000: 356; 190.
3 Aldridge MC. Islet cell tumours: surgical treatment. Hosp.
Med. 2000; 61: 830.
4 Arslan A, Buanes T, Geitung JT. Pancreatic carcinoma: MR,
MR angiography and dynamic helical CT in the evaluation
of vascular invasion. Eur. J. Radiol. 2001; 38: 151.
5 Asbun HJ, Rossi RL, Munson JL. Local resection for
ampullary tumours: is there a place for it? Arch. Surg. 1993;
128: 515.
6 Bartsch DK, Schilling T, Ramaswamy A et al. Management
of nonfunctioning islet cell carcinomas. World J. Surg. 2000;
24: 1418.
7 Bornman PC, Harries-Jones EP, Tobias R et al. Prospective
controlled trial of transhepatic biliary endoprosthesis vs.
bypass surgery for incurable carcinoma of head of pancreas.
Lancet 1986; i: 61.
8 Burris III HA, Moore MJ, Andersen J et al. Improvements in
survival and clinical benefit with gemcitabine as first-line
therapy for patients with advanced pancreas cancer: a ran-
domised trial. J. Clin. Oncol. 1997: 15; 2403.
9 Cameron JL, Crist DW, Sitzman JV et al. Factors influencing
survival following pancreaticoduodenectomy for pancre-
atic cancer. Am. J. Surg. 1991; 161: 120.
10 Cameron JL, Pitt HA, Yeo CJ et al. One hundred and forty-
five consecutive pancreatico-duodenectomies without mor-
tality. Ann. Surg. 1993; 217: 430.

11 Connolly MM, Dawson PJ, Michelassi F et al. Survival in
1001 patients with carcinoma of the pancreas. Ann. Surg.
1987; 206: 366.
12 Davids PHP, Groen AK, Rauws EAJ et al. Randomized trial
of self-expanding metal stents vs. polyethylene stents for
distal malignant biliary obstruction. Lancet 1992; 340: 1488.
13 Deviere J, Cremer M, Baize M et al. Management of common
bile duct stricture caused by chronic pancreatitis with metal
mesh self expandable stents. Gut 1994; 35: 122.
14 DiMagno EP, Reber HA, Tempero MA. AGA technical
review on the epidemiology, diagnosis and treatment of
pancreatic ductal adenocarcinoma. Gastroenterology
1999:
117; 1464.
15 Dorandeu A, Raoul J-L, Siriser F et al. Carcinoma of the
ampulla of Vater: prognostic factors after curative surgery: a
series of 45 cases. Gut 1997; 40: 350.
16 Erickson RA, Garza AA. Impact of endoscopic ultrasound
on the management and outcome of pancreatic carcinoma.
Am. J. Gastroenterol. 2000: 95; 2248.
17 Everhart J, Wright D. Diabetes mellitus as a risk factor for
pancreatic cancer. JAMA 1995; 273: 1605.
18 Fernandez-del Castillo C, Warshaw AL. Current manage-
ment of cystic neoplasms of the pancreas. Adv. Surg. 2000;
34: 237.
19 Freeny PC. Pancreatic carcinoma: imaging update 2001. Dig.
Dis. 2001; 19: 37.
20 Gastrointestinal Tumor Study Group, Kalser MH, Ellenberg
SS. Further evidence of effective adjuvant combined radia-
Diseases of the Ampulla of Vater and Pancreas 645

tion and chemotherapy following curative resection of the
pancreatic head. Cancer 1987; 59: 2006.
21 Hammel P, Couvelard A, O’Toole D et al. Regression of liver
fibrosis after biliary drainage in patients with chronic pan-
creatitis and stenosis of the common bile duct. N. Engl. J.
Med. 2001; 344: 418.
22 Kawa S, Tokoo M, Hasebe O et al. Comparative study of CA
242 and CA 19-9 for the diagnosis of pancreatic cancer. Br. J.
Cancer 1994; 70; 481.
23 Kurzawinski TR, Deery A, Dooley JS et al. A prospective
study of biliary cytology in 100 patients with bile duct stric-
tures. Hepatology 1993; 18: 1399.
24 Legman P, Vignaus O, Dousser B et al. Pancreatic tumours:
comparison of dual-phase helical CT and endoscopic
sonography. Am. J. Radiol. 1998; 170: 1315.
25 Lesur G, Levy P, Flejou J-F et al. Factors predictive of liver
histopathological appearance in chronic alcoholic pancre-
atitis with common bile duct stenosis and increased serum
alkaline phosphatase. Hepatology 1993; 18: 1078.
26 Lewandrowski K, Lee J, Southern J et al. Cyst fluid analysis
in the differential diagnosis of pancreatic cysts: a new
approach to the preoperative assessment of pancreatic
cystic lesions. Am. J. Roentgenol. 1995; 164: 815.
27 Lillemoe KD, Cameron JL, Kaufman HS et al. Chemical
splanchnicectomy in patients with unresectable pancreatic
cancer: a prospective randomized trial. Ann. Surg. 1993; 217:
447.
28 Lowenfels AB, Maisonneuve P, Cavallini G et al. Pancreatitis
and the risk of pancreatic cancer. N. Engl. J. Med. 1993; 328:
1433.

29 Lowenfels AB, Maisonneuve P, DiMagno EP et al. Heredi-
tary pancreatitis and the risk of pancreatic cancer. J. Natl.
Cancer Inst. 1997; 89: 442.
30 McDowell RK, Gazelle GS, Murphy BL et al. Mucinous
ductal ectasia of the pancreas. J. Comput. Assist. Tomogr.
1997; 21: 383.
31 Menges M, Lerch MM, Zeitz M. The double duct sign in
patients with malignant and benign pancreatic lesions. Gas-
trointest. Endosc. 2000; 52: 74.
32 National Institute for Clinical Excellence (NICE) (2001).
Guidance on gemcitabine for pancreatic cancer.
HTTP://www.nice.org.uk. May 2001.
33 Nevitt AW, Vida F, Kozarek RA
et al. Expandable metallic
prostheses for malignant obstructions of gastric outlet and
proximal small bowel. Gastrointest. Endosc. 1998; 47: 271.
34 Norton ID, Geller A, Petersen BT et al. Endoscopic surveil-
lance and ablative therapy for periampullary adenomas.
Am. J. Gastroenterol. 2001; 96: 101.
35 O’Brien S, Hatfield ARW, Craig PI et al. A three-year follow-
up of self expanding metal stents in the endoscopic pallia-
tion of longterm survivors with malignant biliary
obstruction. Gut 1995; 36: 618.
36 Pemert J, Marsson J, Westermark GT et al. Islet amyloid
polypeptide in patients with pancreatic cancer and diabetes.
N. Engl. J. Med. 1994; 360: 313.
37 Polati E, Finco G, Gottin L et al. Prospective randomized
double-blind trial of neurolytic coeliac plexus block in
patients with pancreatic cancer. Br. J. Surg. 1998; 85: 199.
38 Prat F, Chapat O, Ducot B et al. A randomized trial of

endoscopic drainage methods for inoperable malignant
strictures of the common bile duct. Gastrointest. Endosc.
1998; 47: 1.
39 Prat F, Chapat O, Ducot B et al. Predictive factors for sur-
vival of patients with inoperable malignant distal biliary
strictures: a practical management guide. Gut 1998; 42: 76.
40 Queneau P-E, Adessi G-L, Thibault P et al. Early detection of
pancreatic cancer in patients with chronic pancreatitis: diag-
nostic utility of a K-ras point mutation in the pancreatic
juice. Am. J. Gastroenterol. 2001: 96; 700.
41 Rivera JA, Fernandez-del Castillo C, Rall CJN et al. Analysis
of K-ras oncogene mutations in chronic pancreatitis with
ductal hyperplasia. Surgery 1997; 121: 42.
42 Smith AC, Dowsett JF, Russell RCG et al. Randomised trial
of endoscopic stenting vs. surgical bypass in malignant low
bileduct obstruction. Lancet 1994; 344: 1655.
43 Soetikno RM, Lichtenstein DR, Vandervoort J et al. Pallia-
tion of malignant gastric outlet obstruction using an endo-
scopically placed Wallstent. Gastrointest. Endosc. 1998; 47:
267.
44 Speer AG, Cotton PB, Russell RCG et al. Randomized trial
of endoscopic vs. percutaneous stent insertion in malignant
obstructive jaundice. Lancet 1987; ii: 57.
45 Stahl TJ, Allen MO’C, Ansel HJ
et al. Partial biliary
obstruction caused by chronic pancreatitis: an appraisal of
indications for surgical biliary drainage. Ann. Surg. 1988;
207: 26.
46 Stewart CJR, Mills PR, Carter R et al. Brush cytology in the
assessment of pancreatico-biliary strictures: a review of 406

cases. J. Clin. Pathol. 2001; 54: 449.
47 Taheri S, Meeran K. Islet cell tumours: diagnosis and
medical management. Hosp. Med. 2000; 61: 824.
48 Tersmette AC, Petersen GM, Offerhaus GJA et al. Increased
risk of incident pancreatic cancer among first-degree rela-
tives of patients with familial pancreatic cancer. Clin. Cancer
Res. 2001; 7: 738.
49 Traverso LW, Longmire WPJr. Preservation of the pylorus in
pancreaticoduodenectomy. Surg. Gynecol. Obstet. 1978; 146:
959.
50 Van Laethem J-L, Vertongen P, Deviere J et al. Detection
of c-K-ras gene codon 12 mutations from pancreatic duct
brushings in the diagnosis of pancreatic tumours. Gut
1995; 36: 781.
51 Warshaw AL, Zhuo-yun G, Wittenberg J et al. Preoperative
staging and assessment of resectability of pancreatic cancer.
Arch. Surg. 1990; 125: 230.
52 Yamaguchi K, Enjoji M. Carcinoma of the ampulla of Vater.
A clinico-pathologic study and pathologic staging of 109
cases of carcinoma and five cases of adenoma. Cancer 1987;
59: 506.
53 Yamaguchi K, Enjoji M. Adenoma of the ampulla of Vater:
putative precancerous lesion. Gut 1991; 32: 1558.
54 Zoepf T, Zoepf D-S, Arnold JC et al. The relationship
between juxtapapillary duodenal diverticula and disorders
of the biliopancreatic system: analysis of 350 patients. Gas-
trointest. Endosc. 2001; 54: 56.
646 Chapter 36
Benign lesions of the gallbladder
At ultrasound, polypoid lesions of the gallbladder are

occasionally seen and there is usually concern as to their
nature and how to manage them. The vast majority are
benign. They may be true tumours or pseudo-tumours.
True tumours comprise adenoma, lipoma and leiomy-
oma. Pseudo-tumours include cholesterol polyps,
inflammatory polyps and adenomyomatosis.
These lesions are seen most often as an echogeneic
focus that projects into the gallbladder lumen, does not
cast an acoustic shadow, and does not move when the
patient is moved (unlike a stone). The diagnostic accu-
racy of ultrasound for the commonest lesions is 50–90%
depending on the pathology [34].
Cholesterol polyps are usually multiple, with a higher
echogenicity than liver, a pedicle and a mulberry-like
surface [34]. They may contain a hyperechoic spot.
Pathologically they consist of hypertrophied villi laden
with cholesterol.
Adenoma is seen as a polypoid lesion which on ultra-
sound has an echogenicity similar to the liver, a smooth
surface and usually no pedicle [34].
80%–90% of gallbladder polypoid lesions do not
change in size on follow-up scans [41, 57]. However, they
cause concern because of the low chance of malignancy
in adenomas. Cholecystectomy will be done for the
symptomatic patient. This is also appropriate for the
lesion greater than 10mm in diameter, where the risk of
malignancy is greater [34]. Other features of malignant
tumour are a sessile lesion, isoechoic with the liver,
growing rapidly on serial ultrasounds.
Patients with a smaller lesion without these features

should undergo a second scan. Some lesions disappear
but the majority remain, and these patients may be
offered a cholecystectomy for peace of mind. Alterna-
tively, a repeat scan is done at 6-monthly intervals to
detect any change in size [39]. In practice ultrasound
lesions less than 10mm in diameter with benign appear-
ances in an asymptomatic patient tend to be treated con-
servatively but follow-up scanning is important. If the
result of trans-abdominal ultrasound is inconclusive,
endoscopic ultrasound if available is useful with a diag-
nostic accuracy for neoplastic lesions of 80% [10].
Carcinoma of the gallbladder [1]
This is an uncommon neoplasm. Gallstones coexist in
about 75% of cases and chronic cholecystitis is a frequent
association. There is a clear association with large, multi-
ple gallbladder stones [62], but a causal relationship is
unproven.
The calcified (porcelain) gallbladder is particularly
likely to become cancerous [51]. An anomalous pancre-
atico-biliary ductal union, greater than 15mm from the
papilla of Vater, is associated with congenital cystic
dilatation of the common bile duct and with gallbladder
carcinoma [40]. Regurgitation of pancreatic juice may
be tumorigenic. The common gallbladder cholesterol
polyps are not precancerous.
Chronic typhoid infection of the gallbladder increases
the risk of gallbladder carcinoma by 167-fold [6], empha-
sizing the need for antibiotic treatment to eradicate
the chronic typhoid and paratyphoid carrier state, or for
elective cholecystectomy.

Papillary adenocarcinoma starts as a wart-like excres-
cence. It grows slowly into, rather than through, the
wall until a fungating mass fills the gallbladder. Mucoid
change is associated with more rapid growth, early
metastasis and gelatinous peritoneal carcinomatosis.
Squamous cell carcinoma and scirrhous forms are recog-
nized. The anaplastic type is particularly malignant. The
most common tumour is a differentiated adenocarci-
noma [1, 16] which may be papillary.
The tumour usually arises in the fundus or neck,
but rapid spread may make the original site difficult to
locate. The rich lymphatic and venous drainage of the
gallbladder leads to early spread to related lymph nodes,
causing cholestatic jaundice and widespread dissemina-
tion. The liver bed is invaded and there may also be local
spread to the duodenum, stomach and colon resulting in
fistulae or external compression.
Clinical. The patient is usually an elderly, white
female, complaining of pain in the right upper
quadrant, nausea, vomiting, weight loss and jaundice.
Sometimes an unsuspected carcinoma is found in a
cholecystectomy specimen at histology. These small
lesions may not even be recognized at the time of
operation [13].
647
Chapter 37
Tumours of the Gallbladder and
Bile Ducts
Examination may reveal a hard and sometimes tender
mass in the gallbladder area.

Serum, urine and faeces show the changes of cholestatic
jaundice if the bile duct is compressed.
Ultrasound scanning shows a mass in the gallbladder
lumen or totally replacing the gallbladder. With early
lesions the differentiation between gallbladder carci-
noma and a thickened wall due to acute or chronic chole-
cystitis is difficult.
CT may also show a mass in the area of the gallblad-
der. Ultrasound and CT detect carcinoma of the gallblad-
der in 60–70% of cases [45].
By the time an abnormality is shown by ultrasound or
CT, extension is likely and the chance of total removal
low. Endoscopic ultrasound images correlate with histo-
logical depth of invasion and are useful in staging [23].
ERCP shows external compression of the bile duct in
the jaundiced patient. Angiography shows displacement
of hepatic and portal blood vessels by the mass.
In only 50% of patients is a correct pre-operative diag-
nosis made [12].
Prognosis
This is generally hopeless because the majority are inop-
erable at the time of diagnosis. Distant metastases are
already present in 50% of cases [16]. The only long-term
survivors are those in whom the tumour was found inci-
dentally at the time of cholecystectomy for gallstones
(carcinoma in situ).
Median survival from diagnosis is 3 months, with only
14% alive at 1 year [12]. Patients with papillary and well-
differentiated adenocarcinomas have longer survival
than those with tubular and undifferentiated types [28].

The results of radical resection including partial hepa-
tectomy and radical lymphadenectomy are conflicting
[9, 16] with some series showing no survival benefit
and others claiming increased survival.
Treatment
Cholecystectomy has been recommended for all patients
with gallstones in an effort to prevent the development
of carcinoma in the gallbladder. This seems drastic for a
common condition, and would lead to a large number of
unnecessary cholecystectomies.
The pre-operative diagnosis of carcinoma of the gall-
bladder should not preclude laparotomy although the
results of surgical treatment are disappointing. Radical
resection including partial hepatectomy has been
attempted but with unsatisfactory results and no con-
vincing evidence of improved survival [16]. The same
applies to radiotherapy and chemotherapy [1].
Endoscopically or percutaneously placed biliary pros-
theses relieve bile duct obstruction.
Other tumours
Rarely other tumours develop in the gallbladder
including leiomyosarcoma, rhabdomyosarcoma, oat cell
carcinoma and carcinoid tumours.
Benign tumours of the extra-hepatic
bile duct
These extremely rare tumours usually remain unde-
tected until there is evidence of biliary obstruction and
cholangitis. They are rarely diagnosed pre-operatively.
Recognition is important as resection is curative.
Papilloma is a polypoid tumour which projects into the

lumen of the common bile duct. It is a small, soft, vascu-
lar tumour, which may be sessile or pedunculated. These
tumours may be single or multiple; they may be cystic.
Occasionally they undergo malignant change. Cholan-
giography may show a smooth mass projecting into the
bile duct. Mucus secretion from the tumour can cause
obstructive cholangitis.
Adenomyoma can be found anywhere in the biliary
tract. It is firm and well circumscribed and varies in size
up to 15cm in diameter. It is cured by resection [11].
Fibroma is small and firm and causes early bile duct
obstruction.
Granular cell tumour is of mesenchymal origin. It
affects young women, usually black, causing cholestasis
[5]. It must be distinguished from cholangiocarcinoma or
localized sclerosing cholangitis. Tumours are uniformly
resectable and curable.
Carcinoma of the bile duct
(cholangiocarcinoma)
Carcinoma may arise at any point in the biliary tree from
the small intra-hepatic bile ducts to the common bile
duct (fig. 37.1).
The incidence of intra-hepatic cholangiocarcinoma is
increasing. Studies from England and Wales [59] and the
USA[48] show a 10-fold increase between the early 1970s
and the mid-1990s. The explanation is unclear. Although
improved diagnostic techniques for cholangiocarcinoma
and primary sclerosing cholangitis may have played a
part, they do not alone explain the marked increase in
incidence and mortality. Mortality from extra-hepatic

cholangiocarcinoma fell over the same period.
The treatment depends on the site. Resection is the
rule for extra-hepatic tumours. For hilar cholangiocarci-
noma surgical resection should always be considered
but requires particular expertise because of the inaccessi-
bility of hilar tumours, the proximity of the hepatic
artery and portal vein and the need for hepatic resection
in some patients. Even if not curative, surgical treatment
may prolong survival with a good quality of life.
648 Chapter 37
Suspicion of cholangiocarcinoma, for example after
ultrasound scan, should lead to referral to a specialist
unit. This is to co-ordinate the work-up to evaluate
resectability of the tumour. Modern CT techniques and
MRI with MR cholangiography allow a high degree of
non-invasive evaluation. The necessity and timing of
invasive investigations depends on clinical circum-
stances. ERCP and non-operative drainage of bile per-
cutaneously or after ERCP frequently introduce sepsis
which may compromise later treatment [29, 42]. These
aspects emphasize the importance of a multidisciplinary
approach.
In those who are inoperable, biliary drainage by inter-
ventional radiologist or endoscopist relieves pruritus
and usually jaundice to allow satisfactory palliation.
Associations
Bile duct cancer is associated with ulcerative colitis
with or without sclerosing cholangitis (Chapter 15). The
majority of patients with primary sclerosing cholangitis
who develop cholangiocarcinoma have ulcerative

colitis. Patients with primary sclerosing cholangitis and
ulcerative colitis who also have colorectal neoplasia
(dysplasia/carcinoma) are at greater risk of cholangio-
carcinoma than those without colonic neoplasia [4].
In a group of 70 patients with primary sclerosing
cholangitis followed prospectively for a mean of 30
months, 15 patients died of liver failure. Five of 12
patients (40%) having an autopsy had cholangiocarci-
noma
—
7% of the total group [54].
Biliary malignancy is not necessarily a late complica-
tion of primary sclerosing cholangitis. 30% of patients in
one series had a diagnosis of cholangiocarcinoma made
within 1 year of the first evidence of underlying liver
disease based on abnormal liver function tests [37].
Clinical features associated with malignancy were
epigastric pain, weight loss and raised CA 19-9 and
carcinoembryonic antigen (CEA) [37].
All members of the congenital fibropolycystic family
may be complicated by adenocarcinoma (Chapter 33).
These include congenital hepatic fibrosis, cystic dilata-
tion (Caroli’s syndrome), choledochal cyst, polycystic
liver and von Meyenburg complexes. Cholangiocarci-
noma may be associated with biliary cirrhosis due to
biliary atresia.
The liver fluke infestations of the Orient may be com-
plicated by intra-hepatic (cholangiocellular) cholangio-
carcinoma. In the Far East (China, Hong Kong, Korea,
Japan), where Clonorchis sinensis is prevalent, cholangio-

carcinoma accounts for 20% of primary liver tumours.
These arise in the heavily parasitized bile ducts near the
hilum.
Opisthorchis viverrini infestation is important in Thai-
land, Laos and western Malaysia [35]. These parasites
induce DNA changes and mutations through the pro-
duction of carcinogens and free radicals, and the stimu-
lation of cellular proliferation of intra-hepatic bile duct
epithelium [46].
The risk of extra-hepatic bile duct carcinoma is signifi-
cantly lower 10 years or more after cholecystectomy, sug-
gesting a link with gallstones [17].
Pathology
The confluence of the cystic duct with the main hepatic
duct or the right and left main hepatic ducts at the porta
hepatis are common sites of origin (fig. 37.1). Tumours of
the hepatic ducts extend into the liver. They cause com-
plete obstruction of the extra-hepatic bile ducts with
intra-hepatic biliary dilatation and enlargement of the
liver. The gallbladder is collapsed and flaccid. If the
tumour is restricted to one hepatic duct, biliary obstruc-
tion is incomplete and jaundice absent. The lobe of
the liver drained by this duct atrophies and the other
hypertrophies.
In the common bile duct the tumour presents as a
firm nodule or plaque which causes an annular stricture
which may ulcerate. It spreads along the bile duct and
through its wall.
Local and distant metastases, even at autopsy, are
found in only about half of the patients. They involve

peritoneum, abdominal lymph nodes, diaphragm, liver
Tumours of the Gallbladder and Bile Ducts 649
U
pp
er third 58
%
Middle third 17
%
Lower third 18
%
Diffuse 7
%
Fig. 37.1. Site of cholangiocarcinoma. The majority occur in
the upper third of the bile duct [60].
or gallbladder. Blood vessel invasion is rare and extra-
abdominal spread is unusual.
Histologically the tumour is usually a mucin-secreting
adenocarcinoma with cuboidal or columnar epithelium
(fig. 37.2). Spread along neural sheaths may be noted.
Tumours around the hilum are sclerosing with an abun-
dant fibrous stroma. More distal ones are nodular or
papillary.
Molecular changes
Point mutations in codon 12 of the K-ras oncogene
are found in cholangiocarcinoma [65]. p53 protein is
expressed particularly in high-grade mid and distal duct
cholangiocarcinomas [15]. Aneuploidy (divergence from
the normal chromosome content) is found in hilar
cholangiocarcinoma [55] and is associated with neural
invasion and shorter survival.

Cholangiocarcinoma cells contain somatostatin receptor
RNAand cell lines have specific receptors. Cell growth is
inhibited by somatostatin analogues.
Clinical features
This tumour tends to occur in the older age group,
patients being about 60 years old. Slightly more males
than females are affected.
Jaundice is the usual presenting feature, followed by
pruritus
—
a point of distinction from primary biliary cir-
rhosis where itching usually comes first. Jaundice may
be delayed if only one main hepatic duct is involved. The
trend of the serum bilirubin level is always upward, but
periods of clearing of jaundice are found in up to 50%
[31].
Pain, usually epigastric and mild, is present in about
one-third of patients. Diarrhoea may be related to steat-
orrhoea. Weakness and weight loss are marked.
The condition may be associated with chronic ulcera-
tive colitis, often following long-standing cholestasis
due to sclerosing cholangitis.
Examination. Jaundice is deep. Cholangitis is unusual
unless the bile ducts have been interfered with surgi-
cally, endoscopically or percutaneously.
The liver is large and smooth, extending 5–12cm
below the costal margin. The spleen is not felt. Ascites is
unusual.
Investigations
Serum biochemical findings are those of cholestatic jaun-

dice. The serum bilirubin, alkaline phosphatase and g-
glutamyl transpeptidase levels may be very high.
Fluctuations may reflect incomplete obstruction or
primary involvement of one hepatic duct.
The serum mitochondrial antibody test is negative
and a-fetoprotein is not increased.
The faeces are pale and fatty and occult blood is often
present. Glycosuria is absent.
Anaemia may be greater than that seen with ampullary
carcinoma; the explanation is unknown
—
it is not due to
blood loss. The leucocyte count is high normal with
increased polymorphs.
Liver biopsy shows features of large bile duct obstruc-
tion. In primary sclerosing cholangitis biliary dysplasia
raises the possibility of coexistent cholangiocarcinoma
[20].
Cytology taken at the time of ERCP or percutaneous
drainage is worthwhile, but requires cytological exper-
tise for interpretation. Brush cytology is better than
analysis of aspirated bile, with a sensitivity of 60% [56].
Other approaches include fine-needle aspiration cytol-
ogy from the suspected tumour, done under ultrasound
or fluoroscopy. Aspiration cytology guided by endo-
scopic ultrasound is also valuable but this expertise is
rare [22].
In primary sclerosing cholangitis, brush cytology of
dominant strictures at ERCP has a sensitivity of 60% for
cholangiocarcinoma [52]. p53 and K-ras mutation ana-

lysis does not increase sensitivity. However, K-ras
mutations may be found ahead of the diagnosis of
cholangiocarcinoma in patients with primary sclerosing
cholangitis [33].
The serum concentration of the tumour marker CA
19-9 is often increased in patients with biliary tract
650 Chapter 37
Fig. 37.2. Bile duct carcinoma: with irregular fibrous stroma.
(H & E, ¥ 40.)
malignancy. Levels are also increased by cholangitis
and cholestasis. The sensitivity for detecting cholangio-
carcinoma in primary sclerosing cholangitis is 50–60%
[3, 47]. Combining CA 19-9 with CEA measurement
does not increase sensitivity. Although CA 19-9 only has
a moderate sensitivity, it has a role in surveillance of
selected patients with primary sclerosing cholangitis if
only because early detection of bile duct malignancy by
any means is difficult.
Scanning
Ultrasound shows dilated intra-hepatic ducts with a
normal extra-hepatic biliary tree. Atumour mass may be
shown in up to 80% of cases. Ultrasound (real-time
together with Doppler) accurately detects neoplastic
involvement of the portal vein, both occlusion and wall
infiltration, but is less good in showing hepatic arterial
involvement [43]. Intra-duct ultrasound is still experi-
mental but can provide important information on
tumour extension in and around the bile duct [58].
Enhanced CT shows bile duct dilatation, and detects
tumours in 40–70% of patients. Helical CT detects

cholangiocarcinoma, as small as 15mm diameter, in 90%
of patients, and provides information on parenchymal,
intra-hepatic bile duct and portal involvement [19]. It
underestimates extra-hepatic bile duct involvement,
hepatic arterial invasion and lymph node spread.
Overall, magnetic resonance cholangiography has an
accuracy for bile duct stones and strictures of over 90%.
In cholangiocarcinoma (fig. 37.3) it correctly delineates
duct obstruction and the extent of hilar tumour exten-
sion in 80% of patients. It underestimates tumour exten-
sion in 20% [66]. MRI with MRCP is an important
technique for planning the treatment of malignant hilar
strictures but does not replace invasive cholangiography
which also allows brushings to be taken for cytology and
bile drainage if indicated.
CT and MRI therefore may show bile duct obstruction
and a mass but differentiation between inflammatory
and malignant biliary strictures generally depends upon
invasive techniques to obtain cytology and biopsy.
There is a diagnostic problem particularly in primary
sclerosing cholangitis. PET scanning using [
18
F] fluoro-2-
deoxy-
D-glucose has been reported to detect cholangio-
carcinomas in patients with and without primary
sclerosing cholangitis with a sensitivity of 90% [30, 32].
If substantiated this would represent an advance in the
management particularly of patients with primary scle-
rosing cholangitis, although false-positive scans have

been seen in normal individuals [32].
Cholangiography
With the newer non-invasive imaging modalities
now available, the role of direct cholangiography has
changed. Some specialist units rely on Doppler
ultrasound and MRCP and avoid instrumentation of
the biliary tree before surgery [7]. Hepatobiliary units
have their individual approach and tailor this to the
patient.
Endoscopic or percutaneous cholangiography, or
both, still have a role (fig. 37.3). However, as emphasized
above, they are not done immediately after ultrasound
has shown a hilar bile duct obstruction. Jaundice is
often felt to be dangerous, needing immediate treat-
ment, but this is only so for the septic patient or when
there is renal failure. It is prudent to investigate the
patient non-invasively to judge the nature and extent of
the hilar lesion, and then consider direct cholangiogra-
phy, cytology and drainage when the management plan
is clear. MRCP may be used to select the duct system
for endoscopic stent drainage [27] or percutaneous
puncture.
In hilar cholangiocarcinoma, ERCP shows the normal
common bile duct and gallbladder with obstruction at
the hilum (fig. 37.3c). Contrast usually passes through
the stricture(s) into dilated bile ducts above. The stric-
ture is passed with a guide-wire, cytology is done, and a
stent placed.
Percutaneous cholangiography shows the dilated
intra-hepatic ducts down to the stricture (fig. 37.3d). A

drain is inserted. When right and left hepatic ducts are
individually obstructed, puncture of both systems may
be necessary to outline the obstruction accurately. Some
specialist surgeons prefer the percutaneous rather than
endoscopic approach to the system because it provides
detailed information on tumour extension within the
liver and intra-hepatic bile ducts. This is of more value
than the appearances of the extra-hepatic biliary tree
below the stricture.
Angiography
Digital subtraction angiography (DSA) shows the
hepatic artery and portal vein and their intra-hepatic
branches. Its use depends upon the information derived
from ultrasound, CT and MRI.
Diagnosis
In the patient with deepening cholestatic jaundice the
clinical diagnosis is likely to be carcinoma of the peri-
ampullary region. Other possibilities are drug jaundice,
primary sclerosing cholangitis (Chapter 15) and primary
biliary cirrhosis (Chapter 14). Cholangiocarcinoma is an
Tumours of the Gallbladder and Bile Ducts 651
652 Chapter 37
Fig. 37.3. A 75-year-old woman presenting with cholestatic jaundice. Ultrasound showed dilated intra-hepatic ducts, a hilar mass
and a normal common bile duct. (a) MRCP shows dilated intra-hepatic ducts with at least three segments obstructed in the right lobe
and the left hepatic system obstructed at the hilum. If non-surgical drainage is to be done, these appearances favour drainage of the
left- rather than the right-sided system (D, duodenum). (b) MRI scan shows a mass in the liver (arrow) above the hilum. (c) Non-
operative drainage was chosen since the patient was considered inoperable. ERCP shows a normal common duct with a hilar
structure. Astent could not be placed. (d) Following on the MRCP appearances, the left-sided duct system was chosen for
percutaneous cholangiography and a stent inserted.
(a) (b)

(c) (d)
unusual cause but it should be detected if an orderly
work-up is used. History and examination are usually
unhelpful.
The first step in the cholestatic patient is ultrasound
scanning. Intra-hepatic bile ducts will be dilated in
cholangiocarcinoma. The common duct is normal,
equivocal or may be dilated down to an extra-hepatic
tumour. If there is a suspicion of hilar cholangiocarci-
noma and other clinical features do not indicate inoper-
ability, the choice is MRCP or, if this is not available,
referral to a specialist hepato-biliary unit.
If ultrasound does not show dilated bile ducts in the
cholestatic patient, other causes (Chapter 13) need to
be considered including drug jaundice (history) and
primary biliary cirrhosis (anti-mitochondrial antibody).
Liver histology will help. If primary sclerosing cholangi-
tis is possible, cholangiography is diagnostic.
With scanning and cholangiography it should be pos-
sible to diagnose the bile duct stricture due to cholangio-
carcinoma. At the hilum, the differential diagnosis is a
benign stricture [25] or metastatic gland, in the mid-duct
carcinoma of the gallbladder, and in the peri-ampullary
region carcinoma of the pancreas. Differentiation will
depend upon history and other imaging techniques.
Prognosis
Prognosis depends on the site of the tumour. Those dis-
tally placed are more likely to be resectable than those
at the hilum. The histologically differentiated do better
than the undifferentiated. Polypoid cancers have the

best prognosis.
If unresected, the 1-year survival for cholangio carci-
noma is 50%, with 20% surviving at 2 years and 10% at
3 years [18]. This reflects that some tumours are slow
growing and metastasize late. Jaundice can be relieved
surgically or by endoscopic or percutaneous stenting.
The tumour kills by its site making it inoperable, rather
than by its malignancy. Average survival after resection
is longer, making proper assessment in patients fit for
surgery essential.
Staging [7]
If the clinical state of the patient does not rule out
surgery the resectability and extent of tumour is
assessed. Metastases, usually late, should be sought.
Low and mid common bile duct lesions are usually
resectable although vascular imaging is needed to
exclude invasion.
Hilar cholangiocarcinoma is more problematic (table
37.1). If cholangiography shows involvement of the sec-
ondary hepatic ducts in both hepatic lobes (fig. 37.4, type
IV) or imaging shows encasement of the main portal
vein or hepatic artery, the lesion is irresectable. A pallia-
tive procedure is needed.
If the tumour is limited to the hepatic duct bifurca-
tion, affecting one lobe of the liver only, or only obstructs
the portal vein or hepatic artery on the same side,
the lesion may be resectable. Pre-operative imaging is
aimed at establishing whether after surgical removal
a viable unit of liver remains [7]. This must contain
a biliary radicle large enough to anastomose to bowel,

and a normal portal vein and hepatic arterial branch.
At surgery, further assessment is done with intra-
operative ultrasound and a search for lymph node
involvement.
In a department with a high resection rate, pre-
operative cholangiography predicted clinical manage-
ment in 62% of patients, and angiography determined
management in 80% [38].
Treatment
Surgery
Tumours of the lower bile duct may be resected with a 1-
year survival of about 70%. More proximal tumours may
be resected by local or major liver surgery including
excision of the whole bifurcation of the common
hepatic duct, lobectomy if necessary and bilateral
hepatico-jejunostomy.
Tumours of the Gallbladder and Bile Ducts 653
Table 37.1. Criteria of irresectability for hilar
cholangiocarcinoma
Bilateral bile duct involvement or multifocal disease on
cholangiography
Main trunk of portal vein encased/occluded
Bilateral involvement of hepatic arterial or portal vein
branches or both
Unilateral hepatic artery involvement and extensive
contralateral bile duct involvement
I
II
III
a

III
b
IV
Fig. 37.4. Classification of hilar cholangiocarcinoma according to the involvement of bile ducts. Resectability of type I to III depends
on angiographic findings. Type IV (bilateral involvement of secondary hepatic ducts) indicates incurable disease. In inoperable
patients median survival after stent insertion depends upon the extent of tumour [53].
Some advocate caudate lobectomy, based on the
observation that two to three bile ducts from this lobe
drain directly into the main bile ducts adjacent to the
confluence of the hepatic ducts and thus are likely to be
involved by tumour.
The proportion of cholangiocarcinomas being
resected has increased from 5–20% of patients in the
1970s to 30% or more in specialist centres in the 1990s.
This relates to earlier diagnosis and referral to a tertiary
centre, more accurate and complete pre-operative
assessment, and a more aggressive surgical approach.
The problem is to achieve a resection with tumour-
negative margins. Median survival after aggressive
resection of hilar cholangiocarcinoma is 18–40 months
with good palliation for most of this time [7, 38]. Local
resection of Bismuth type I and II tumours (fig. 37.4)
carries a peri-operative mortality of 5% or less. Liver
resection is needed for type III lesions, and carries a
greater morbidity and mortality.
Liver transplantation is not appropriate for cholangio-
carcinoma because of early recurrence in the majority
[64].
Surgical palliative procedures include anastomosis of
jejunum to the segment III duct in the left lobe which is

usually accessible despite the hilar tumour (fig. 37.5).
Jaundice is relieved for at least 3 months in 75% of
patients [26]. If segment III bypass is not possible
(atrophy, metastases), a right-sided intra-hepatic anasto-
mosis to the segment V duct can be done.
Non-surgical palliation
In those patients unfit for surgery or with irresectable
tumours, jaundice and itching may be relieved by
placing an endoprosthesis across the stricture either by
the endoscopic or percutaneous route.
By the endoscopic route, stents can be inserted suc-
cessfully in about 90% of patients if a combined endo-
scopic/percutaneous procedure is included after a failed
endoscopic attempt. The major early complication is
cholangitis (7%). Thirty-day mortality is between 10 and
28% depending upon the extent of the tumour at the
hilum and the mean survival is 20 weeks [53]. Stenting of
only one lobe is necessary [14].
Percutaneous trans-hepatic endoprosthesis insertion
is also successful but carries with it a higher risk of com-
plications such as bleeding and bile leakage (Chapter
32). Metal mesh endoprostheses, which expand to 10mm
diameter in the stricture after insertion of a 5 or 7 French
catheter, are more expensive than plastic types, but have
longer patency for peri-ampullary strictures [36]. They
may be used for hilar strictures. Studies suggest a similar
advantage over plastic endoprostheses [63] but their
insertion requires an experienced operator.
There are no trials comparing surgical versus non-
surgical palliation. There are benefits and disadvantages

of both approaches [44]. Generally, non-operative tech-
niques are appropriate for high-risk patients expected to
have a shorter survival. Because of recurrent stent block-
age requiring replacement [24], surgical bypass should
be considered as an alternative palliative approach.
Internal radiotherapy using an iridium-192 wire or
radium needles may be combined with biliary drainage
[21]. The value of this technique is unproven. Cytotoxic
drugs are ineffective. External radiotherapy has ap-
peared to show some benefit in retrospective studies but
in a randomized trial showed no benefit [50]. Intra-duct
photodynamic therapy combined with stenting has
given encouraging results in Bismuth type III and IV
cholangiocarcinoma [2]. There is a local tumour response
of 30–75% and hilar bile ducts occlusion can be reversed.
The treatment is costly and controlled studies are needed
to establish survival benefit. Symptomatic treatment is
that of chronic cholestasis (Chapter 13).
Cholangiocellular carcinoma
This intra-hepatic bile duct derived tumour is classified
as a primary hepatic carcinoma. It becomes symptomatic
as it enlarges producing abdominal pain rather than
jaundice [8]. It grows rapidly with early metastasis and a
particularly poor prognosis. There is an association with
Thorotrast (thorium dioxide), an intravenous contrast
medium used many years ago. Scanning shows an intra-
hepatic mass. Distinction from hepato-cellular carci-
654 Chapter 37
Fig. 37.5. Check cholangiogram after surgical bypass for hilar
cholangiocarcinoma. The anastomosis is between the jejunum

and the third segment duct of the left lobe (arrow).
noma may be difficult. Hepatic venous and portal vein
involvement is rare. Surgery is the only chance for
effective treatment. Resection is possible in 30–60% of
cases [8]. One-year survival after resection is 50–60%.
Transplantation for irresectable tumour gives a median
survival of 5 months [49].
Metastases at the hilum
Cholestatic jaundice developing following the diagnosis
of carcinoma elsewhere (in particular the colon) may
be due to diffuse metastases within the liver or duct
obstruction by nodes at the hilum. Differentiation
between the two is by ultrasound. If dilated bile ducts
are shown and the patient is symptomatic with itching,
biliary obstruction can be relieved by insertion of an
endoprosthesis by the endoscopic or percutaneous
approach [61]. Palliation is achieved depending upon
the extent of tumour but the 30-day mortality is greater
and the survival significantly shorter compared with
endoprosthesis insertion for primary bile duct malig-
nancy [53].
References
1 Bartlett DL. Gallbladder cancer. Semin. Surg. Oncol. 2000; 19:
145.
2 Berr F, Wiedmann M, Tannapfel A et al. Photodynamic
therapy for advanced bile duct cancer: evidence for
improved palliation and extended survival. Hepatology
2000; 31: 291.
3 Björnsson E, Kilander A, Olsson R. CA 19-9 and CEA are
unreliable markers for cholangiocarcinoma in patients with

primary sclerosing cholangitis. Liver 1999; 19: 501.
4 Broomé U, Löfberg R, Veress B et al. Primary sclerosing
cholangitis and ulcerative colitis: evidence for increased
neoplastic potential. Hepatology 1995; 22: 1404.
5 Butterly LF, Schapiro RH, LaMuraglia GM et al. Biliary gran-
ular cell tumour: a little-known curable bile duct neoplasm
of young people. Surgery 1988; 103: 328.
6 Caygill CPJ, Hill MJ, Braddick M et al. Cancer mortality in
chronic typhoid and paratyphoid carriers. Lancet 1994; 373:
83.
7 Chamberlain RS, Blumgart LH. Hilar cholangiocarcinoma:
a review and commentary. Ann. Surg. Oncol. 2000; 7: 55.
8 Chen M-F. Peripheral cholangiocarcinoma (cholangiocellu-
lar carcinoma): clinical features, diagnosis and treatment. J.
Gastroenterol. Hepatol. 1999; 14: 1144.
9 Chijiiwa K, Tanaka M. Carcinoma of the gallbladder: an
appraisal of surgical resection. Surgery 1994; 115: 751.
10 Choi W-B, Lee S-K, Kim M-H et al. Anew strategy to predict
the neoplastic polyps of the gallbladder based on a scoring
system using EUS. Gastrointest. Endosc. 2000; 52: 372.
11 Cook DJ, Salena BJ, Vincic LM. Adenomyoma of the
common bile duct. Am. J. Gastroenterol. 1988; 83: 432.
12 Cubertafond P, Gainant A, Cucchiaro G. Surgical treatment
of 724 carcinomas of the gallbladder. Results of the French
Surgical Association Survey. Ann. Surg. 1994; 219: 275.
13 De Aretxabala X, Roa I, Brugos L et al. Gallbladder cancer in
Chile: a report on 54 potentially resectable tumours. Cancer
1992; 69: 60.
14 De Palma GD, Galloro G, Siciliano S et al. Unilateral vs. bilat-
eral endoscopic hepatic duct drainage in patients with

malignant hilar biliary obstruction: results of a prospective,
randomized, and controlled study. Gastrointest. Endosc.
2001; 53: 547.
15 Diamantis I, Karamitopoulou E, Perentes E
et al. p53 protein
immunoreactivity in extrahepatic bile duct and gallbladder
cancer: correlation with tumour grade and survival. Hepa-
tology 1995; 22: 774.
16 Donohue JH, Nagorney DM, Grant CS et al. Carcinoma of
the gallbladder: does radical resection improve outcome?
Arch. Surg. 1990; 125: 237.
17 Ekbom A, Hsieh C, Yuen J et al. Risk of extra hepatic bile
duct cancer after cholecystectomy. Lancet 1993; 372: 1262.
18 Farley DR, Weaver AL, Nagorney DM. ‘Natural history’ of
unresected cholangiocarcinoma: patient outcome after non-
curative intervention. Mayo Clin. Proc. 1995; 70: 425.
19 Feydy A, Vilgrain V, Denys Aet al. Helical CT assessment in
hilar cholangiocarcinoma: correlation with surgical and
pathologic findings. Am. J. Roentgenol. 1999; 172: 73.
20 Fleming KA, Boberg KM, Glaumann H et al. Biliary dyspla-
sia as a marker of cholangiocarcinoma in primary sclerosing
cholangitis. J. Hepatol. 2001; 34: 360.
21 Fletcher MS, Brinkley D, Dawson JL et al. Treatment of hilar
carcinoma by bile drainage combined with internal radio-
therapy using 192-iridium wire. Br. J. Surg. 1983; 70: 733.
22 Fritscher-Ravens A, Broering DC, Sriram PVJ et al. EUS-
guided fine-needle aspiration cytodiagnosis of hilar cholan-
giocarcinoma: a case series. Gastrointest. Endosc. 2000; 52:
534.
23 Fujita N, Noda Y, Kobayashi G et al. Diagnosis of the depth

of invasion of gallbladder carcinoma by EUS. Gastrointest.
Endosc. 1999; 50: 659.
24 Gerhards MF, den Hartog D, Rauws EAet al. Palliative treat-
ment in patients with unresectable hilar cholangiocarci-
noma: results of endoscopic drainage in patients with type
III and IV hilar cholangiocarcinoma. Eur. J. Surg. 2001; 167:
274.
25 Gerhards MF, Vos P, van Gulik TM et al. Incidence of benign
lesions in patients resected for suspicious hilar obstruction.
Br. J. Surg. 2001; 88: 48.
26 Guthrie CM, Banting SW, Garden OJ et al. Segment III
cholangiojejunostomy for palliation of malignant hilar
obstruction. Br. J. Surg. 1994; 81: 1639.
27 Hintze RE, Abou-Rebyeh H, Adler A et al. Magnetic reso-
nance cholangiopancreatography-guided unilateral endo-
scopic stent placement for Klatskin tumours. Gastrointest.
Endosc.
2001; 53: 40.
28 Hisatomi K, Haratake J, Horie Aet al. Relation of histopatho-
logical features to prognosis of gallbladder cancer. Am. J.
Gastroenterol. 1990; 85: 567.
29 Hochwald SN, Burke EC, Jarnagin WR et al. Association of
preoperative biliary stenting with increased postoperative
infectious complications in proximal cholangiocarcinoma.
Arch. Surg. 1999; 134: 261.
30 Keiding S, Hansen SB, Rasmussen HH et al. Detection of
cholangiocarcinoma in primary sclerosing cholangitis by
positron emission tomography. Hepatology 1998; 28: 700.
31 Klatskin G. Adenocarcinoma of the hepatic duct at its bifur-
cation within the porta hepatis. An unusual tumour with

Tumours of the Gallbladder and Bile Ducts 655
distinctive clinical and pathological features. Am. J. Med.
1965; 38: 24.
32 Kluge R, Schmidt F, Caca K et al. Positron emission tomogra-
phy with [
18
F] fluoro-2-deoxy-D-glucose for diagnosis and
staging of bile duct cancer. Hepatology 2001; 33: 1029.
33 Kubicka S, Kohnel F, Flemming P et al. K-ras mutations in
the bile of patients with primary sclerosing cholangitis. Gut
2001; 48: 403.
34 Kubota K, Bandai Y, Noie T et al. How should polypoid
lesions of the gallbladder be treated in the era of laparo-
scopic cholecystectomy? Surgery 1995; 117: 481.
35 Kurathong S, Lerdverasirikul P, Wongpaitoon V et al.
Opisthorchis viverrini infection and cholangiocarcinoma. A
prospective, case-controlled study. Gastroenterology 1985; 89:
151.
36 Lammer J, Hausegger KA, Flockiger F et al. Common bile
duct obstruction due to malignancy: treatment with plastic
vs. metal stents. Radiology 1996; 201: 167.
37 Leidenius M, Höckerstedt K, Broomé U et al. Hepatobiliary
carcinoma in primary sclerosing cholangitis: a case control
study. J. Hepatol. 2001; 34: 792
38 Lillemoe KD, Cameron JL. Surgery for hilar cholangiocarci-
noma: the Johns Hopkins approach. J. Hepatobil. Pancreat.
Surg. 2000; 7: 115.
39 Mainprize KS, Gould SWT, Gilbert JM. Surgical manage-
ment of polypoid lesions of the gallbladder. Br. J. Surg. 2000;
87: 414.

40 Misra SP, Dwivedi M. Pancreaticobiliary ductal union. Gut
1990; 31: 1144.
41 Moriguchi H, Tazawa J, Hayashi Y et al. Natural history of
polypoid lesions in the gall bladder. Gut 1996; 39: 860.
42 Nakeeb A, Pitt HA. The role of preoperative biliary decom-
pression in obstructive jaundice. Hepatogastroenterology
1995; 42: 332.
43 Neumaier CE, Bertolotto M, Perrone R et al. Staging of hilar
cholangiocarcinoma with ultrasound. J. Clin. Ultrasound
1995; 23: 173.
44 Nordback IH, Pitt HA, Coleman J et al. Unresectable hilar
cholangiocarcinoma: percutaneous vs. operative palliation.
Surgery 1994; 115: 597.
45 Oikarinen H, Paivansalo M, Lahde S et al. Radiological find-
ings in cases of gallbladder carcinoma. Eur. J. Radiol. 1993;
17: 179.
46 Parkin DM, Ohshima H, Srivatanakul P et al. Cholangiocar-
cinoma: epidemiology, mechanisms of carcinogenesis and
prevention. Cancer Epidemiol. Biomarkers Prev. 1993; 2: 537.
47 Patel AH, Harnois DM, Klee GG et al. The utility of CA 19–9
in the diagnoses of cholangiocarcinoma in patients without
primary sclerosing cholangitis. Am. J. Gastroenterol. 2000; 95:
204.
48 Patel T. Increasing incidence and mortality of primary intra-
hepatic cholangiocarcinoma in the United States. Hepatology
2001; 33: 1353.
49 Pichlmayr R, Lamesch P, Weimann A et al. Surgical treat-
ment of cholangiocellular carcinoma. World J. Surg. 1995; 19:
83.
50 Pitt HA, Nakeeb A, Abrams RA et al. Perihilar

cholangiocarcinoma. Postoperative radiotherapy does not
improve survival. Ann. Surg. 1995; 221: 788.
51 Polk HC. Carcinoma and calcified gallbladder. Gastroen-
terology 1966; 50: 582.
52 Ponsioen CY, Vrouenraets SME, van Milligen de Wit AWM
et al. Value of brush cytology for dominant strictures in
primary sclerosing cholangitis. Endoscopy 1999; 31: 305.
53 Polydorou AA, Cairns SR, Dowsett JF et al. Palliation of
proximal malignant biliary obstruction by endoscopic
endoprosthesis insertion. Gut 1991; 32: 685.
54 Rosen CB, Nagorney DM, Wiesner RH et al. Cholangiocar-
cinoma complicating primary sclerosing cholangitis. Ann.
Surg. 1991; 213: 21.
55 Sato Y, van Gulik TM, Bosma A et al. Prognostic significance
of tumour DNA content in carcinoma of the hepatic duct
confluence. Surgery 1994; 115: 488.
56 Stewart CJR, Mills PR, Carter R et al. Brush cytology in the
assessment of pancreatico-biliary strictures: a review of 406
cases. J. Clin. Pathol. 2001; 54: 449.
57 Sugiyama M, Atomi Y, Yamato T. Endoscopic ultrasono-
graphy for differential diagnosis of polypoid gall bladder
lesions: analysis in surgical and follow up series. Gut 2000;
46: 250.
58 Tamada K, Ido K, Ueno N et al. Preoperative staging of
extrahepatic bile duct cancer with intraductal ultrasonogra-
phy. Am. J. Gastroenterol.
1995; 90: 239.
59 Taylor-Robinson SD, Toledano MB, Arora S et al. Increase in
mortality rates from intrahepatic cholangiocarcinoma in
England and Wales 1968–98. Gut 2001; 48: 816.

60 Tompkins RK, Saunders KD, Roslyn JJ et al. Changing pat-
terns in diagnosis and management of bile duct cancer. Ann.
Surg. 1990; 211: 614.
61 Valiozis I, Zekry A, Williams SJ et al. Palliation of hilar
biliary obstruction from colorectal metastases by endo-
scopic stent insertion. Gastrointest. Endosc. 2000; 51: 412.
62 Vitetta L, Sali A, Little P et al. Gallstones and gall bladder
carcinoma. Aust. NZ J. Surg. 2000; 70: 667.
63 Wagner H-J, Knyrim K, Vakil N et al. Plastic endoprostheses
vs. metal stents in the palliative treatment of malignant
hilar biliary obstruction: a prospective randomized trial.
Endoscopy 1993; 25: 213.
64 Wall WJ. Liver transplantation for hepatic and biliary malig-
nancy. Semin. Liver Dis. 2000; 20: 425.
65 Watanabe M, Asaka M, Tanaka J et al. Point mutation of K-
ras gene codon 12 in biliary tract tumours. Gastroenterology
1994; 107: 1147.
66 Zidi SH, Prat F, Le Guen O et al. Performance characteristics
of magnetic resonance cholangiography in the staging of
malignant hilar strictures. Gut 2000; 46: 103.
656 Chapter 37
In 1955, Welch performed the first transplantation of the
liver in dogs [118]. In 1963, Starzl and his group carried
out the first successful hepatic transplant in man [101].
The number of transplants is escalating and, in 1997,
4099 patients were transplanted in the USA. Elective
liver transplantation in low-risk patients has a 90%
1-year survival. Improved results can be related to more
careful patient selection, to better surgical techniques
and post-operative care, and to greater willingness to

re-transplant after rejection. Better immunosuppression
has contributed.
Selection of patients
The patient selected for transplant should suffer from
irreversible, progressive disease for which there is no
acceptable, alternative therapy. The patient and the
family must understand the magnitude of the under-
taking and be prepared to face the difficult early post-
operative period and life-long immunosuppression.
Improved results have led to a greater acceptance of
the procedure. Demand has exceeded supply of donor
organs (fig. 38.1). The time spent awaiting transplant
and deaths occurring before it can be performed have
increased. The waiting time for low-risk patients is
approximately 6–12 months. Although in general this
may be longer for those of blood group B and AB, group
O recipients may have the longest waiting time because
group O is the universal donor type. Depending on the
system of organ distribution, such livers can be given to
recipients having any ABO group. Donor livers suitable
for children are particularly rare and this has led to the
split-liver technique (see fig. 38.5).
The equitable distribution of the precious donor livers
is difficult. Results (and costs) are much better if
the patient is low risk (ambulatory) compared with high
risk (intensive care). Decisions are usually made by a
multi-disciplinary panel including the patient and
patient’s family. In the USA, the United Network for
Organ Sharing (UNOS) guidelines are followed (tables
38.1, 38.2) [63]. A modified Child–Pugh score is used as

the basis by which to evaluate the severity of liver
disease. The priorities expressed by the general public
are not the same as those of clinicians. There has been a
perceived lack of fairness in organ allocation [76]. The
UNOS website (www.unos.org) allows the public and
clinicians access to transplant activities and outcomes.
Livers should be allocated for medical need and not on
the basis of financial or other considerations. Unaccept-
able criteria include the patient’s contribution to society,
inability to comply with treatment (e.g. antisocial behav-
iour), the patient’s contribution to the disorder (e.g. drug
or alcohol abuse) and the past use of medical resources
[27]. Recipients are broadly defined as having an intoler-
able quality of life because of liver disease or having an
anticipated length of life of less than 1 year because of
liver failure. There are few guidelines predicting sur-
vival. Patients more than 65 years old have a substan-
tially worse 5-year survival, but age itself is not a
contraindication.
Candidates: outcome (table 38.3)
In Europe, the pattern of primary indication for liver
transplantation is changing. The main indication is cir-
rhosis, including primary biliary cirrhosis. More patients
with acute and sub-acute hepatic failure are being
657
Chapter 38
Hepatic Transplantation
0
6000
5000

4000
3000
2000
1000
7000
8000
1988 1990 1992
Year
Waiting
Donors
Transplant
Deaths
Number
1994 1996 1998
Fig. 38.1. Over the last 8 years availability of donors has not
kept up with the demand for transplants. Waiting time and
deaths have increased.
658 Chapter 38
Table 38.1. Child–Pugh scoring system to assess severity of liver disease
Points
12 3
Encephalopathy (grade) None 1–2 3–4
Ascites Absent Slight or controlled At least moderate despite
by diuretics diuretic treatment
Bilirubin (mg/dl) <2 2–3 >3
Albumin (g/dl) >3.5 2.8–3.5 <2.8
Prothrombin time (seconds <4 4–6 >6
prolonged)
Or INR <1.7 1.7–2.3 >2.3
For primary biliary cirrhosis, <4 4–10 >10

primary sclerosing cholangitis
or other cholestatic liver
diseases: bilirubin (mg/dl)
Table 38.2. Liver transplantation listing criteria for UNOS status 1, 2A, 2B and 3
Status 1
Fulminant hepatic failure. Onset within 8 weeks of initial symptoms and one of the following:
stage 2 encephalopathy
bilirubin >15 mg/dl
INR >2.5
hypoglycaemia (glucose level <50 mg/dl)
Primary non-function of graft transplanted within 7 days
Hepatic artery thrombosis occurring within 7 days of transplantation
Acute decompensated Wilson’s disease
Status 2A
Patient with chronic liver failure and a Child–Pugh score ≥10, in the critical care unit, with a life expectancy without a liver transplant of less
than 7 days, with at least one of the following criteria:
unresponsive active variceal haemorrhage with failure or contraindication of surgical or transjugular intra-hepatic shunt
hepato-renal syndrome
refractory ascites/hepato-renal syndrome (hydrothorax)
stage 3–4 encephalopathy unresponsive to therapy
Contraindications to status 2A listing:
extra-hepatic sepsis unresponsive to antimicrobial therapy
requirement for high dose or two or more pressor agents to maintain an adequate blood pressure
severe, irreversible multi-organ failure
Status 2B
Patients with chronic liver disease and a Child–Pugh score ≥10, or ≥7 and one or more of the following clinical considerations:
unresponsive variceal haemorrhage
hepato-renal syndrome
spontaneous bacterial peritonitis
refractory ascites/hepato-renal syndrome (hydrothorax)

Liver transplant candidates with hepato-cellular carcinoma can be registered as status 2B if they meet the following criteria:
thorough assessment has excluded metastatic disease
recipient has one nodule £5 cm or three or fewer nodules all £3cm
patient is not a resection candidate
Status 3
Patients with chronic liver disease and a Child–Pugh score ≥7
included and fewer with cancer because of the poor sur-
vival due to recurrence of tumour (table 38.4).
Cirrhosis
All patients with end-stage cirrhosis should be consid-
ered for liver transplantation. Selection of the right time
is difficult. The patient must not be moribund, so that the
transplant will fail, or be capable of leading a relatively
normal life for a long period so that transplant is un-
necessary. Indications include a prothrombin time more
than 5s prolonged, a serum albumin concentration of less
than 30g/l, and intractable ascites. Bleeding oesophageal
varices, after failure of medical treatment, is a good indi-
cation. The cost of transplant is little different from that of
long-term medical and surgical management of compli-
cations such as bleeding, coma and ascites.
The patients are poor operative risks because of
impaired blood coagulation and portal hypertension, so
that blood loss is great. The technical difficulties are
greater when cirrhosis is present, particularly when the
liver is small and difficult to remove. Survival is much
the same for all forms of cirrhosis.
Autoimmune chronic hepatitis
Post-transplant 5-year survival is 91% and graft survival
83% [85]. Despite triple immunosuppression, 33%

develop recurrent chronic hepatitis of autoimmune
type [30]. This is usually asymptomatic but may be
severe, leading to graft failure [86]. Control is usually re-
established by adjusting corticosteroid dose.
Chronic viral hepatitis
Hepatic transplantation performed for acute fulmi-
nant viral hepatitis (A, B, D and E) is not followed by
graft re-infection as the viral levels are very low. In the
chronic situation, however, graft re-infection is very
common.
Hepatitis B
Post-transplant recurrence is usual and is related to
viral replication in extra-hepatic sites, particularly
monocytes. HBV-DNA is measured frequently. A severe
fibrosing cholestatic hepatitis may develop with ballooning
of hepatocytes and ground-glass change. This may be
related to high cytoplasmic expression of viral antigens
in the presence of immunosuppression [7, 31]. HBV may
sometimes be cytopathic.
Hepatitis B immune globulin (HBIG), given before and
after transplant, improves survival. However, it does not
eliminate the virus and breakthrough replication can
occur. It must be given indefinitely and is costly. Levels
of protective antibody may be achieved by a new
HBIG preparation (CMRI Hb) obtained from screened
vaccinated donors [4].
HBV vaccination, following discontinuation of HBIG,
may be associated with the development of protective
serum titres of anti-HBs [93].
Lamivudine given before and after transplant may

control re-infection. It has allowed transplant in HBV
DNA and HBeAg-positive patients. The high develop-
ment of lamivudine failure [71] with lamivudine-resis-
tant mutants is associated with re-infection [55, 65, 73].
The combination of lamivudine and HBIG as prophy-
laxis against reinfection is effective, so long as lamuvi-
dine resistance has not occurred pre-transplant [92].
Hepato-cellular cancer can develop in the trans-
planted liver.
Hepatic Transplantation 659
Table 38.4. Percentage survival of 9966 patients according to
diagnosis of cirrhosis, acute liver failure and cancer. (Data from
European Liver Transplant Registry, 1993)
Survival (%)
Diagnosis 1 year 2 year 3 year
Cirrhosis 80 73 71
Acute liver failure 60 56 54
Cancer 64 42 36
Table 38.3. Possible candidates for hepatic transplantation
Cirrhosis
Cryptogenic
Autoimmune
Virus B (HBV DNA negative)
Virus D
Virus C
Alcoholic (Chapter 22)
Cholestatic liver disease
Primary biliary cirrhosis
Biliary atresia
Primary sclerosing cholangitis

Secondary sclerosing cholangitis
Graft-versus-host disease
Chronic hepatic rejection
Cholestatic sarcoidosis (Chapter 28)
Chronic drug reactions (rare)
Primary metabolic disease (see table 38.5)
Acute liver failure (Chapter 8)
Malignant disease (Chapter 31)
Hepato-cellular carcinoma
Epithelioid haemangio-endothelioma
Hepatoblastoma
Miscellaneous
Budd–Chiari syndrome (Chapter 11)
Short-bowel syndrome
Hepatitis delta
Transplantation is almost always followed by infection
of the graft. HDV RNAand HDAg can be detected in the
new liver and HDV RNA in the serum [124]. Hepatitis
only develops if there is concomitant or superinfection
with HBV.
Hepatitis B infection is inhibited by delta infection
and hepatitis B recurrence may be reduced by delta
infection. In general, survival is good after transplant for
delta-infected patients.
Hepatitis C virus
Hepatitis C is the commonest indication for liver
transplantation in most centres. All patients who are
positive for HCV by PCR pre-transplant will remain
positive, and 97% will develop recurrent hepatitis C
post-transplant (fig. 38.2). Genetic sequencing of vari-

able regions of the viral genome pre- and post-transplant
will determine whether the infection is recurrent or
acquired. Infection of the graft can come from infected
mononuclear cells which contain negative strand viral
RNA
—
the replicative intermediate of the viral genome.
Early results suggested that the overall 5-year survival of
patients with HCV was not worse than that with other
liver diseases [40]. However, graft loss is increased and a
11–12% mortality is seen in the first year [24]. Whether
long-term survival will decline due to recurrent disease
awaits further study [13].
The post-transplant course is very variable; 50% of
HCV-positive recipients, despite viraemia, will not have
graft hepatitis within 2 years of the operation [13].
Lobular hepatitis, frequently asymptomatic, may
develop 1–4 months post-transplant. It is unrelated to
serum ALT. The hepatitis usually progresses to chronic
hepatitis, which is often severe, with progressive fibrosis
[8] and eventually graft loss [41]. Progressive fibrosing
cholestatic hepatitis may be seen [95]. Re-grafting is
usually necessary.
Patients with pre-treatment HCV viral titres exceed-
ing 1¥10
6
have a 5-year survival of 57% as opposed to
84% for those with lower titres [24].
Prediction of a severe course is difficult. Studies of the
relationship to genotype and quasi-species are conflict-

ing [8, 42]. Earlier recurrence of HCV relates to the
number of rejection episodes [98], with consequent
immunosuppression and hence, increasing HCV
viraemia. The risk of severe fibrosis/cirrhosis may be
greater with more aggressive immunosuppression [20].
There is no difference between cyclosporin- and
tacrolimus-treated patients.
Interferon is ineffective in those with particularly high
serum viral titres and so does not affect survival or graft
loss [97]. It may increase the risk of graft rejection.
A pilot study has shown that interferon–ribavirin
therapy can result in 50% clearance of HCV at the end of
therapy. Of initial responders, 50% will relapse [12, 14].
Further controlled trials and longer follow-up are
needed.
Neonatal hepatitis
This disease of unknown aetiology is associated with
jaundice, giant cell hepatitis and rarely liver failure
necessitating liver transplant which is curative [23].
Alcoholic liver disease
In the West, these patients are likely to provide the
largest number of candidates for transplant. The
selection and the results obtained are discussed in
Chapter 22.
Cholestatic liver disease
End-stage biliary disease, usually involving the small
intra-hepatic bile ducts, is an excellent indication
for hepatic transplantation (fig. 38.3). Hepato-cellular
function is usually preserved until late and the timing of
660 Chapter 38

Increased
immunosuppression
Donor
HCV positive
Reinfection -
from recipient
mononuclear cells
Blood products
Severe
Usually mild
Post-transplant
HCV
Fig. 38.2. Mechanisms for recurrent HCV hepatitis after liver
transplantation.
Biliary atresia
Primary biliary cirrhosis
Graft-vs host disease
Chronic rejection
Cholestatic sarcoidosis
Chronic drug cholestasis
Sclerosing cholangitis
Transplant
Loss
of
bile
ducts
Fig. 38.3. Diseases with disappearing bile ducts treated by
liver transplantation.
the transplant is easy. In every case the liver shows an
advanced biliary cirrhosis, often combined with loss of

bile ducts (‘disappearing bile duct syndrome’).
Primary biliary cirrhosis (Chapter 14)
One-year patient survival is 80% [96]. Recurrence is
evident on hepatic histology, but there are no reports of
subsequent graft failure.
Extra-hepatic biliary atresia (Chapter 26)
This indication comprises 35–67% of paediatric liver
transplants. Calculated 1-year survival is 75%. Results
are excellent and long-term survivors have good phy-
sical and mental development, although re-transplant
and post-transplant surgery is often necessary.
A previous Kasai procedure increases the operative
difficulty and the morbidity.
Alagille’s syndrome
Transplant is required only in very severe sufferers [22].
Associated cardio-pulmonary disease may be fatal and
careful pre-operative assessment is necessary.
Primary sclerosing cholangitis (Chapter 15)
Sepsis and previous biliary surgery provide technical
problems. Nevertheless, the results for transplantation
are good, 1-year survival being 70% and 5-year survival
57%. Cholangiocarcinoma is a complication that greatly
reduces long-term survival. Colon cancer is the most
frequent cause of death [72].
Langerhans’ cell histiocytosis accounts for 15–39% of
sclerosing cholangitis. It has been successfully treated by
liver transplant [123].
Other end-stage cholestatic diseases
Hepatic transplantation has been performed for
graft-versus-host cirrhosis in a bone marrow recipient.

Other rare indications include cholestatic sarcoidosis
(Chapter 28) and chronic drug reactions.
Primary metabolic disease
Liver homografts retain their original metabolic speci-
ficity. Consequently, liver transplantation is used for
patients with inborn errors that result from defects in
hepatic metabolism. Patients suffering from these condi-
tions are good candidates. Selection depends on the
prognosis and the likelihood of the later complication of
primary liver tumours.
Liver transplantation for metabolic disorders is
divided into those performed for end-stage liver disease or
pre-malignant change and those performed for major
extra-hepatic features (table 38.5). Overall survival is
85.9% over 5.5 years’ follow-up.
End-stage liver disease
a
1
-Antitrypsin deficiency
This is the most common metabolic disease leading to
liver transplantation. Macronodular cirrhosis will
develop in about 15% before the age of 20 years. Hepato-
cellular carcinoma is a complication. The plasma a
1
-
antitrypsin deficiency is corrected and the lung disease
stabilizes after the transplant. Advanced pulmonary
disease is a contraindication unless both lungs and liver
are transplanted.
Genetic haemochromatosis (Chapter 23)

This is an uncommon indication for transplantation.
Survival is lower than for other indications, because of
infection and cardiac problems. Clear-cut recurrence of
hepatic iron has not been reported but follow-ups are
short [16].
Wilson’s disease (Chapter 24)
Liver transplants have to be considered in patients pre-
senting with fulminant hepatitis, in young cirrhotic
patients with severe hepatic decompensation who have
failed to improve after 3 months’ adequate d-penicil-
lamine treatment, and in effectively treated patients who
have developed severe hepatic decompensation follow-
ing discontinuance of penicillamine.
The overall survival is 72% increasing to 90% where
the indication is fulminant Wilson’s disease [94].
Hepatic Transplantation 661
Table 38.5. Liver transplantation for metabolic disorders
End-stage disease or pre-malignant change
a
1
-Antitrypsin deficiency
Wilson’s disease
Tyrosinaemia
Galactosaemia
Glycogen storage diseases
Protoporphyria
Neonatal haemochromatosis
b-thalassaemia
Cystic fibrosis
Byler’s disease

Major extra-hepatic features
Primary oxaluria type 1
Homozygous hypercholesterolaemia
Crigler–Najjar syndrome
Primary coagulation disorders (factor VIII, IX, protein C)
Urea cycle defects
Mitochondrial respiratory chain defects
Primary familial amyloidosis
Survivors transplanted for severe neurological com-
plications showed significant improvement [37].
Glycogen storage diseases
Liver transplantation has been successfully performed
for types I and IV, with survival and continued growth
into adult life.
Galactosaemia
A few patients diagnosed late develop advanced cirrho-
sis in childhood and early adult life and are candidates
for transplantation [81].
Protoporphyria
This can lead to end-stage cirrhosis and so be an
indication for liver transplantation [52]. Post-
operatively, the high level of protoporphyrin in erythro-
cytes and faeces persists and the disease is not cured.
Tyrosinaemia
Hepatic transplantation is curative and should be con-
sidered early before the development of hepato-cellular
carcinoma [68].
b-Thalassaemia
Combined heart and liver transplantation has been
reported for end-stage, iron-induced organ failure in an

adult with homozygous b-thalassaemia [78].
Cystic fibrosis
Hepatic transplantation is indicated for predominant
liver involvement. Combined liver–lung transplant is
often necessary. The 3-year survival of young patients
with end-stage respiratory failure complicated by cirrho-
sis is 70% [28].
Byler’s disease
Byler’s disease (Progressive Familial, Intra-hepatic
Cholestasis type 1) results in death from cirrhosis or
heart failure. The low serum apolipoprotein A1 concen-
tration is corrected by transplant performed for cirrhosis
[19].
Correction of extra-hepatic features
Oxaluria
Primary oxaluria type I, due to deficiency of hepatic
peroxisomal alanine-glyoxylate aminotransferase, is
corrected by simultaneous hepatic and renal transplan-
tation [117]. Cardiac dysfunction reverses. The hepatic
transplantation should possibly be done before renal
damage has developed.
Homozygous hypercholesterolaemia
Liver transplant produces an 80% decrease in serum
lipids. Cardiac transplant or coronary bypass are also
usually necessary [88].
Crigler–Najjar syndrome
Liver transplant is indicated to prevent neurological
sequelae when the serum bilirubin level is very high and
cannot be controlled by phototherapy.
Primary coagulation disorders

The usual indication is HCV cirrhosis. Transplant cures
the haemophilia but the effects of HIV infection
and recurrent viral hepatitis remain post-transplant
complications [46].
Urea cycle enzyme deficiencies
Transplantation has been performed for ornithine trans-
carbamylase deficiency as urea cycle enzymes are pre-
dominantly located in the liver [108]. The decision
concerning the need for transplantation is difficult as
some urea cycle disorders allow a normal lifestyle.
Mitochondrial respiratory chain defects
These may cause liver disease in neonates associated
with hypoglycaemia and postprandial hyperlactic-
acidaemia. They have been treated by liver transplant.
Primary familial amyloidosis
Transplant, often by the domino technique, is performed
for intractable polyneuropathy. Neurological impro-
vement is variable.
Acute liver failure (Chapter 8)
Indications include fulminant viral hepatitis, Wilson’s
disease, acute fatty liver of pregnancy, drug overdose
(for instance, paracetamol), and drug-related hepatitis
[11].
Malignant disease (Chapter 31)
Hepatic transplantation has been disappointing in
patients with liver tumours despite pre-operative
attempts at identifying extra-hepatic spread. Patients
662 Chapter 38
with cancer have a low operative mortality, but the worst
long-term survival. Carcinomatosis is the usual cause of

death. Tumour recurs in 60%, perhaps because of the
immunosuppressants necessary to prevent rejection.
The peri-operative survival is 76%, but the 1-year sur-
vival only 50% and the 2-year survival 31%. For all
tumours transplanted, the overall actual 5-year survival
is 20.4%.
Hepato-cellular carcinoma (Chapter 31)
The tumour must be 5cm or less. If multifocal, only three
tumours less than 3cm each should be considered.
Staging laparoscopy is important at the time of trans-
plant. Vascular invasion, even microscopic, increases the
recurrence rate and mortality.
The 2-year survival is 50 versus 83% for non-
malignant conditions.
Transplantation may be preferable to resection for
small tumours discovered incidentally in a patient with
compensated cirrhosis.
Fibro-lamellar carcinoma
The tumour is localized to the liver and cirrhosis is
absent. This may be the best tumour candidate for trans-
plantation.
Epithelioid haemangio-endothelioma
This presents as multiple focal lesions in both lobes of an
otherwise normal liver. The course is unpredictable and
recurrence is likely in 50%. Metastatic spread does
not contraindicate surgery and this does not correlate
with survival. It can be successfully treated by liver
transplantation.
Hepatoblastoma
Transplantation results in a 50% survival at 24–70

months. Microscopic vascular invasion and anaplastic
epithelium with extra-hepatic spread are bad signs.
Neuro-endocrine tumours
When resection is not possible, worthwhile palliation
can result from hepatic transplantation [62].
Abdominal cluster operations for right upper
quadrant malignancy
Most of the organs derived from the embryonic foregut
are removed including liver, duodenum, pancreas,
stomach and intestine. With powerful immunosuppres-
sion, donor lympho-reticular cells circulate without
causing clinical graft-versus-host disease and become
those of the recipient without causing rejection [104].
The procedure is clearly very radical and patients
usually die from recurrent tumour.
Cholangiocarcinoma
This is an unsatisfactory indication as tumour recurrence
is usual and 3-year survival is poor, being zero in some
series. In some countries, patients with cholangiocarci-
noma are not accepted as transplant candidates.
Miscellaneous
Budd–Chiari syndrome (Chapter 11)
Hepatic transplantation is used in those who are too ill to
perform decompressive surgery and where previous
portal-systemic shunts have failed [89]. The 5-year sur-
vival is 67–69%. Recurrence of thrombosis is likely, espe-
cially in those who have an underlying coagulopathy.
Non-alcoholic fatty liver disease (NAFLD)
The end-stage is macronodular fatty cirrhosis. This is
treated by transplantation, but 50% develop liver biopsy

evidence of NAFLD post-transplant [79].
Absolute and relative contraindications
(table 38.6)
Absolute
These include uncorrectable cardio-pulmonary disease,
ongoing infection, metastatic malignancy and severe
brain damage.
Hepatic Transplantation 663
Table 38.6. Absolute and relative contraindications to liver
transplantation
Absolute
Psychological, physical and social inability to tolerate the procedure
Active sepsis
Metastatic malignancy
Cholangiocarcinoma
AIDS
Advanced cardiopulmonary disease
Relative (higher risk)
Age more than 60 or less than 2 years
Prior-porta-caval shunt
Prior complex hepato-biliary surgery
Portal vein thrombosis
Re-transplant
Multi-organ transplants
Obesity
HIV
Serum creatinine more than 2 mg/dl (180mmol/l)
Cytomegalovirus mismatch
Advanced liver disease
Transplant should not be done if the patient cannot

comprehend the magnitude of the undertaking and the
exceptional physical and psychological commitment
required [64].
Relative (higher risk)
Patients are at higher risk if they have advanced
liver disease (UNOS status 2A) and are being treated
in an intensive care unit and particularly if they are
ventilation-dependent.
Children do particularly well but technical difficulties
increase below the age of 2 years.
Risk increases with a body weight of more than 100kg.
Multi-organ transplant adds to the risk.
A pre-transplant serum creatinine level exceeding
2.0mg/dl (180mmol/l) is the most accurate predictor of
post-transplant death [29].
CMV mismatch (recipient negative, donor positive)
adds to the risk.
Portal vein thrombosis makes the transplant more
difficult and survival is reduced. However, the operation
is usually possible [105]. An anastomosis is made
between the donor portal vein and the recipient con-
fluence of superior mesenteric vein and splenic vein, or a
venous graft from the donor is used.
Previous surgical porta-caval shunts make the opera-
tion more difficult and a distal spleno-renal shunt creates
least problems. TIPS for variceal bleeding is the most
satisfactory preliminary to transplantation [2]. Careful
positioning of the stent is important, avoiding an exces-
sive length down the portal vein.
Previous complex surgery in the upper abdomen also

makes the transplant technically very difficult.
Re-transplant
In Europe, primary transplant is associated with
a 71% survival at 1 year. This is reduced to 47% for
re-transplantation.
General preparation of the patient
The usual clinical, biochemical and serological investiga-
tion of any patient with liver disease is detailed.
Blood group, HLA and DR antigens are recorded.
Antibodies to cytomegalovirus and hepatitis C are mea-
sured and markers of hepatitis B infection noted.
In patients with malignant disease, metastases must
be sought by all possible techniques.
Cardio-pulmonary assessment should be thorough
including the exclusion of hepato-pulmonary syndrome.
Imaging. Splanchnic vasculature and particularly the
hepatic artery and portal vein must be visualized as a
guide to surgery. Doppler ultrasound is routine. The
hepatic arterial tree is also shown in contrast-enhanced
helical CT [77].
MRI may be used as an alternative, or together
with CT to exclude vascular abnormalities and silent
malignancy.
The bile ducts are visualized by MRI cholangiography
[44] or, if necessary, by ERCP.
The pre-transplant medical ‘work-up’ takes about 10
days. It includes psychiatric counselling and confirma-
tion of the diagnosis. The patient may wait many months
for a suitable donor liver and, during this period, inten-
sive psychosocial support is necessary.

Donor selection and operation
Donation may be informed with consent from the family,
the clinician ensuring that the family have been con-
sulted, or presumed consent including the patient having
specifically indicated their wish to donate. Those coun-
tries practising presumed consent tend to have higher
transplant rates than those using informed consent.
However, Spain, with the highest donation transplanta-
tion rate in Europe has the custom of informed consent,
but a very well-resourced programme of trained co-
ordinators. Better education, support and advice is
needed for all clinical staff who have contact with
potential donors [76].
Donor shortage has encouraged the use of livers for-
merly regarded as unsatisfactory. These include livers
from donors with abnormal liver tests, elderly donors,
those with prolonged ICU stay receiving inotropes, or
with steatosis which was formerly an exclusion criterion.
Use of these marginal livers does not seem to have
increased graft loss.
Donors are considered between 2 months and 60–65
years of age, victims of brain injury that has resulted in
brain death. Cardio-vascular and respiratory functions
are sustained by mechanical ventilation. The recovery of
livers and other vital organs from heart-beating cadavers
minimizes the ischaemia that occurs at normal body tem-
peratures and is a major contribution to graft success.
Transplant across A, B and O blood groups may be fol-
lowed by severe rejection. It should be avoided unless
necessitated by an emergency situation [48].

HLA matching is more difficult and indeed there is
some evidence that selected HLA class II mismatches
may be advantageous, particularly in preventing the
vanishing bile duct syndrome [74].
Hepatitis B and C viral markers, CMV antibodies and
HIV testing should be done.
The donor operation is as follows. The hepatic struc-
tures are dissected and the liver is pre-cooled through
the portal vein with Ringer’s lactate and 1000ml of Uni-
versity of Wisconsin (UW) solution perfused through the
aorta and portal vein. A cannula in the distal inferior
664 Chapter 38