The Role of Endoscopic Ultrasound

in Hepatobiliary Disease
David A. Schwartz, MD and Maurits J. Wiersema, MD

Address biliary disease, including choledocholithiasis, microlithia-

Division of Gastroenterology and Hepatology, Mayo Clinic, sis, biliary strictures, obstructive jaundice of unknown
200 First Street SW, Rochester, MN 55905, USA. etiology, cholangiocarcinoma, gallbladder polyps, and
E-mail: Wiersema.maurits@mayo.edu
liver lesions.
Current Gastroenterology Reports 2002, 4:72–78
Current Science Inc. ISSN 1522-8037
Copyright © 2002 by Current Science Inc.
Choledocholithiasis
Choledocholithiasis occurs in up to 20% of patients with
The evaluation of hepatobiliary disease continues to change
gallbladder stones [3]. The diagnosis of common duct
as new technologies are introduced. One of the more
stones can be difficult because clinical and biochemical
recent advances in hepatobiliary imaging has been the appli-
evidence can be subtle or even absent. Standard non-
cation of endoscopic ultrasound (EUS). Endoscopic retro-
invasive imaging modalities, including US and CT, can
grade cholangiopancreatography, computed tomography,
accurately diagnose gallstones, but are insensitive for
conventional ultrasound, and magnetic resonance imaging
choledocholithiasis [4]. The gold standard for diagnosing
were previously the primary means of evaluating hepato-
common duct stones has been ERCP. This modality is very
biliary disease. The role of EUS as both a substitute and a
sensitive (>90%) in detecting choledocholithiasis. If a
complement for these historical imaging modalities contin-
stone is present, ERCP allows for removal of the stones at
ues to evolve. This review examines the current literature
the same time. However, ERCP is invasive and can cause
on EUS in the evaluation of hepatobiliary diseases, including
pancreatitis. Ideally, a sensitive and relatively noninvasive
choledocholithiasis, microlithiasis, biliary strictures,
study would document the presence of a common duct
obstructive jaundice, cholangiocarcinoma, gallbladder
stone first before proceeding with ERCP. EUS is ideal for
polyps, and liver lesions.
this role. With EUS, biliary gallstones typically appear as
semicircular hyperechoic lesions with posterior acoustic
shadowing (Figs. 1 and 2). Sludge within the gallbladder
Introduction or common bile duct (CBD) appears as layered low-
The diagnostic applications of endoscopic ultrasound amplitude echoes within the lumen (Fig. 3).
(EUS) have grown tremendously over the last decade and a The accuracy of EUS for detecting CBD stones has been
half. EUS now plays an important role in the evaluation of demonstrated in a number of comparative studies. Several
a multitude of divergent conditions ranging from lung and studies have shown EUS to be superior to CT or US for the
rectal cancer to perianal Crohn’s disease. One of the initial diagnosis of choledocholithiasis [4–6]. In the largest
uses of EUS was in the diagnosis of hepatobiliary diseases of these three studies, Amouyal et al. [4] prospectively
[1,2]. Because the liver and bile ducts are closely related to compared US, CT, and EUS in 62 consecutive patients who
the distal stomach and the proximal duodenum, EUS is presented with signs of choledocholithiasis. Cholangio-
ideally suited for examination of the biliary tract. This graphy was used as the gold standard. EUS demonstrated a
anatomy allows one to examine the pancreas and biliary 97% sensitivity rate for detecting CBD stones, compared
tract without obstruction from abdominal fat or gas within with only 75% for CT and 25% for US. The difference in
the small intestine. sensitivity was even more pronounced when the subgroup
Historically, endoscopic retrograde cholangiopancrea- of patients with nondilated ducts or with stones less than
tography (ERCP) and standard imaging modalities such as 1 cm in diameter were separated out. EUS was 100%
computed tomography (CT), conventional ultrasound sensitive in these clinical scenarios, compared with 0% for
(US), and magnetic resonance imaging (MRI) have been ultrasound and less than 80% for CT. In addition, the
the primary means of evaluating patients with hepato- negative predictive value of endosonography was 97%,
biliary disease. The role of EUS both as a substitute and as compared with 78% for CT and 56% for US.
a complement to these historical imaging modalities Similarly, several studies have compared EUS with
continues to evolve. This review examines the current ERCP and found the two modalities to be equally accurate
literature on the use of EUS in the evaluation of hepato- [7–9]. In one of the first large series looking at the accuracy
 The Role of Endoscopic Ultrasound in Hepatobiliary Disease • Schwartz and Wiersema 73

Figure 1. A 20-mm gallstone is shown within the gallbladder. Figure 2. A 7-mm stone is shown within the common bile duct.
This typically appears as a semicircular hyperechoic lesion with post- Note the easily identified postacoustic shadowing.
acoustic shadowing.

of EUS for CBD stones, Palazzo et al. [7] compared EUS

with the surgical findings in 185 patients. In this series,
endosonography had a sensitivity rate of 95% and a
specificity rate of 98%. EUS identified all of the stones in a
separate group of 219 patients who underwent ERCP
instead of surgery.
In a very well-designed study, Prat et al. [8] prospec-
tively performed EUS followed within 2 hours by ERCP in
119 patients who presented with suspected choledo-
cholithiasis. Endoscopic sphincterotomy with exploration
of the CBD was used as the gold standard. The sensitivity
rate of EUS for choledocholithiasis was 93%, and specific-
ity was 97%. The sensitivity and specificity rates for ERCP
in this study were 89% and 100%, respectively. As a result,
the authors concluded that EUS should be performed
before a laparoscopic cholecystectomy if there are predic-
tive factors favoring a CBD stone or in young patients in
order to prevent the long-term sequelae of sphincterotomy.
Several studies looking at EUS prior to laparoscopic
cholecystectomy support this recommendation [10–12]. Figure 3. Sludge within the gallbladder is shown. Sludge typically
appears as layered low-amplitude echoes within the lumen, usually
In a large study from South Korea, 132 patients who without shadowing.
presented with symptomatic cholelithiasis were evaluated
with US, US plus EUS, and US plus ERCP prior to laparo-
scopic cholecystectomy [11]. EUS had a sensitivity rate for South Carolina [13•]. In this study, the authors found that
detecting a CBD stone similar to that of ERCP (89% vs the least costly strategy for imaging in patients with a
93%) and superior to that of US alone (65%). In addition, suspected CBD stone depended upon the patient’s risk for
because a complication rate of 5.3% was associated with stones. Procedural costs and expertise were also important
ERCP, compared with 0% for EUS in the study, the authors factors. EUS was found to be the most cost-effective
recommend reserving ERCP exclusively for therapeutic procedure for patients who were at intermediate risk for CBD
indications prior to laparoscopic cholecystectomy. stones (10% to 50%). Patients at high risk for choledo-
The role of EUS prior to laparoscopic cholecystectomy cholithiasis (recent cholangitis and jaundice, acute pancrea-
was further clarified in a decision analysis from researchers in titis, or elevated alkaline phosphatase and a dilated CBD)
74 Liver

should have an ERCP, whereas those at low risk for choledo- patients. Two studies prospectively compared EUS with
cholithiasis (no history of stones, normal liver tests, and bile crystal analysis for the diagnosis of microlithiasis
normal-size CBD) should be treated with expectant manage- [19,20]. In a study from France, 45 consecutive patients
ment. One caveat to these recommendations concerns with biliary-type pain and two normal transabdominal US
patients who are scheduled for surgery several days after the examinations were evaluated with EUS and duodenal bile
onset of symptoms. Frossard et al. [14] performed ERCP at crystal analysis [20]. The gold standard was confirmation
various times following EUS in patients with CBD stones. at cholecystectomy (33 patients) or clinical follow-up over
They found the rate of spontaneous stone migration to be a median of 17 months (12 patients). EUS was 96%
21% within 3 days of symptoms and an additional 20% sensitive and 86% specific for detecting gallstone micro-
between days 3 and 27. Therefore, in high-risk patients in lithiasis or sludge, compared with 67% and 91%, respec-
whom more than 3 days have passed from the onset of tively, for bile analysis.
symptoms, EUS should be considered instead of an ERCP. In The recent availability of small high-frequency intra-
this setting, the likelihood of the patient spontaneously ductal ultrasound (IDUS) catheter probes has added a
passing the CBD stone may have altered the current risk for valuable tool for the evaluation of CBD stones. This
choledocholithiasis to the intermediate level. modality is especially useful for those patients suspected of
The utility of endosonography has also been docu- having microlithiasis or residual stones after attempted
mented in the evaluation of patients with acute pancrea- stone extraction [21,22]. ERCP has difficulty diagnosing
titis in whom an urgent ERCP is not indicated (ie, absence small stones and differentiating stones from air bubbles
of obstructive jaundice or biliary sepsis) [15–18]. In a within the duct. IDUS should have the capability to over-
recent large study from Hong Kong, 100 consecutive come this difficulty. In a study from Japan, IDUS (using a
patients who presented with acute pancreatitis were 20-MHz probe) was compared with ERCP in 31 patients
evaluated by EUS followed immediately by ERCP [15]. [22]. The investigators found IDUS to be superior to
EUS was shown to be as sensitive as ERCP in detecting cholangiography (97% vs 81%). The IDUS catheter was
choledocholithiasis (both 97%) with a specificity rate of inserted over a wire. Sphincterotomy was not needed in
98% for EUS versus 95% for ERCP. In addition, ERCP was any of the patients.
unsuccessful in five patients, and four patients developed
bleeding following sphincterotomy. There were no
complications associated with EUS, and the CBD was Biliary Strictures
visualized in all patients. As a result, the authors concluded Biliary strictures are notoriously difficult to evaluate. The
that EUS should be used to select patients for therapeutic sensitivity of brush cytology and biopsy for identifying a
ERCP in order to avoid the complications associated with malignancy is poor. Endosonography, and specifically
diagnostic ERCP in patients presenting with suspected IDUS, have been used to aid in the evaluation of biliary
gallstone pancreatitis. strictures with moderate success (Fig. 4). In a study by
The same group also looked at the role of EUS in the Gress et al. [23], 36 patients with a biliary stricture were
evaluation of idiopathic pancreatitis [16]. Patients were assessed with IDUS. The investigators were unable to
diagnosed with idiopathic pancreatitis if they presented identify a reliable indicator of malignancy in the cohort.
with acute pancreatitis and no evidence of stones was found In fact, 45% of the malignant strictures were symmetri-
by conventional US or CT. In a group of 89 patients who cal, and 33% of the benign strictures were asymmetrical.
presented with acute pancreatitis, 18 patients were labeled A 12.5-MHz catheter was used in this study. This lower-
as having idiopathic pancreatitis. In these 18 patients, EUS frequency probe probably prohibited reliable imaging of
detected small stones in 14 patients. It should be noted that the CBD wall. Tamada et al. [24] used a 20-MHz probe to
ERCP was performed in 13 of those 18 patients and that no examine 42 patients with biliary strictures. The accuracy
stones were identified. No evidence of cholelithiasis was of IDUS in this study using the higher-frequency probe
found in the remaining four patients after a median follow- was only 76%. Despite the low accuracy of IDUS, the
up of 22 months. authors were able to identify several subgroups in which
IDUS might be helpful. In patients in whom IDUS shows
a normal bile duct wall, no further studies are needed.
Microlithiasis No malignancies were found in the seven patients with a
Patients who present with recurrent right upper-quadrant normal bile duct wall. A bile duct wall layering that was
pain and a negative transabdominal ultrasound can be interrupted by a tumor was diagnostic of a malignancy.
clinically challenging. Once the more common causes of A malignancy was confirmed in 25 of 26 patients with
the symptoms, such as irritable bowel syndrome or peptic this finding. The one false-positive finding occurred in
ulcer disease, have been ruled out, the clinician needs to a patient who had had previous biliary surgery. Because
consider the alternate diagnoses, including microlithiasis. of the limitations of IDUS in this setting, the remainder
Because EUS is highly accurate in detecting microlithiasis of the findings at IDUS examination need further
and biliary sludge, it has a role in the evaluation of these diagnostic testing.
 The Role of Endoscopic Ultrasound in Hepatobiliary Disease • Schwartz and Wiersema 75

solely on appearance; fine-needle aspiration (FNA) was not

performed in these studies. In our experience, these patients
tend to have a large amount of reactive adenopathy,
probably from subclinical cholangitis. This is very difficult
to differentiate from malignant involvement by appearance
alone. We routinely perform FNA of suspicious nodes in
these patients. FNA should help to improve the N-staging
accuracy of EUS for cholangiocarcinoma. For example, in a
recent study looking at EUS-guided FNA in 10 patients with
hilar cholangiocarcinomas in whom brush cytology was
negative or unavailable, EUS-guided FNA was able to
provide diagnostic material in nine patients [32]. In
addition, three patients were confirmed to have metastatic
lymph nodes (two with hilar nodes and one with celiac
nodes). A 22-gauge needle was used to perform the FNA
in this study.
In an effort to improve the T-staging accuracy of EUS,
several centers (including ours) have begun using IDUS to
assess these tumors. Sugiyama et al. [33] and Tamada et al.
[34–37] have conducted the majority of the studies defin-
Figure 4. This 20-MHz intraductal ultrasound examination shows an ing IDUS use in this setting. IDUS has also shown accuracy
asymmetrical echo-poor lesion within the common bile duct that is in the assessment of portal vein or hepatic artery invasion
consistent with a malignant stricture.
[33–35], pancreatic parenchymal involvement [36], and
the amount of tumor ingrowth into biliary stents [37,38].
Recently, a study from Germany compared IDUS with We typically use IDUS to help characterize an indetermi-
EUS in 56 patients with biliary strictures of unknown cause nate stricture or to better determine vascular involvement.
prior to surgery [25•]. A 20-MHz catheter probe was used
for this study. The investigators applied the recommenda-
tions from the previously mentioned study from Tamada Obstructive Jaundice of Unknown Cause
et al. [24] to interpret the IDUS findings, and as a result Generally, patients who present with cholestasis are evalu-
they were able to achieve improved accuracy. In this study, ated with US or CT. However, these investigations are fre-
the accuracy of IDUS was 89%, compared with 76% for quently inconclusive, and additional evaluation is required
conventional EUS. to establish the diagnosis. In the past, the next step would
have been to proceed with ERCP. Over the last decade,
however, several studies have demonstrated that EUS is also
Cholangiocarcinoma effective in this setting [39,40]. In a study by Dancygier and
Patients with cholangiocarcinoma tend to present with Nattermann [41], 41 patients who presented with obstruc-
painless jaundice, and standard imaging methods such as tive jaundice were evaluated by US, EUS, and ERCP within
CT and US usually show evidence only of biliary dilation 24 hours of their admission to the hospital. ERCP was
without an identifiable cause. Because endosonography is considered the gold standard. EUS was found to be superior
very sensitive for small lesions, it is ideally suited for the to US (98% vs 60%) and equivalent to ERCP. EUS provided
evaluation of biliary malignancies. Two studies have the additional benefit of directly imaging the tumor if a
found EUS to be superior to CT and US and equal to tumor was present. EUS-guided FNA now gives a further
ERCP in the diagnosis of small biliary neoplasms [26,27]. advantage by providing a tissue diagnosis during the same
In addition, EUS has the advantage of providing examination. However, EUS cannot perform a therapeutic
staging information. procedure such as stenting. Therefore, EUS and ERCP are
The initial studies looking at the accuracy of EUS in stag- complementary procedures, and the exact role for each in
ing cholangiocarcinoma were reported by Tio et al. this setting is still being defined.
[1,28,29]. In a study of 76 patients with cholangiocarci- Two studies have helped to clarify further the role of EUS
noma, these authors found the T-staging accuracy of EUS to in this setting. Erickson et al. [42•] looked at the cost savings
be 83% for CBD cancers and 85% for common hepatic duct accomplished by performing EUS as the first endoscopic
malignancies [29]. However, the accuracy of EUS for assess- procedure in 147 patients who presented with obstructive
ing lymph nodes was poor (53% to 55%). Other authors jaundice of unknown cause. In this study cohort, the most
have found similar results for T-staging and N-staging common causes of jaundice were malignancy (65%) and
accuracy with EUS [27,30,31]. Lymph node determinations choledocholithiasis or cholecystitis (18%). As a result of the
in these studies (as in the studies by Tio et al.) were based high prevalence of malignancy in this study, 54% of the
76 Liver

patients underwent EUS-guided FNA. Only 53% of the

patients required an ERCP after EUS. The majority of ERCP
procedures were for extraction of a CBD stone or placement
of a stent. This EUS-first strategy led to a cost savings of
between $1007 and $1313 per patient. The second study to
examine the role of EUS in this clinical setting was from
Burtin et al. [43]. These authors looked at the most efficient
diagnostic strategy for patients who presented with obstruc-
tive jaundice of unknown cause. This study specifically
assessed how the health status of a patient and the need for
endoscopic therapy affected efficiency. In this study of 68
patients who presented with indeterminate cholestasis, the
patients who were better operative candidates were less
likely to need ERCP after EUS than were patients who were
poor surgical candidates. In other words, patients who were
at higher risk for complications from surgery were more
likely to be sent for ERCP in order to obviate surgery.
Conversely, patients who were at low risk for complications Figure 5. EUS FNA with a 22-gauge needle shows a hepatic metasta-
sis discovered during the staging examination of a patient with esoph-
from surgery (ie, young and healthy patients) tended to go
ageal cancer. The lesion was not seen on CT.
directly to surgery after EUS. Therefore, the authors
concluded that EUS is the most efficient initial endoscopic
procedure for patients who are potential surgical candidates. including echo pattern (as noted previously), layer
structure, margin of polyps, stalk, and number of polyps
[48]. They applied this scoring system retrospectively to
Gallbladder Polyps and Cancer 53 patients who had both EUS and a cholecystectomy for
Gallbladder polyps are a common incidental finding on gallbladder polyps over a 2.5-year period. Using receiver-
transabdominal ultrasound. They have been reported to operating characteristic curves (ROC) analysis, they found
occur in 5% of the population [44]. Gallbladder cancer, by that, in the group of patients with polyps between 5 and 15
comparison, is relatively uncommon but carries a very mm in size, a score greater than 5 (maximum score of 20)
poor prognosis even if it is identified early. Previously, with was associated with an 84.6% sensitivity and specificity
transabdominal ultrasound, the only way to distinguish rate for identifying neoplastic polyps. Clinically, this is the
between the more common cholesterol polyp and a most difficult group to differentiate because of the inter-
neoplastic polyp was on the basis of polyp size. Polyps mediate risk of a neoplasm in this size of polyps.
larger than 1 cm in diameter have a higher likelihood of EUS is also useful if gallbladder cancer has already devel-
being neoplastic. Because of the limitations of the previous oped. Mitake et al. [49] performed EUS on 39 patients with
screening methods, patients with polyps larger than 1 cm gallbladder cancer prior to surgery. They found the T-staging
were sent for prophylactic cholecystectomy despite the fact accuracy to be 76.9%. The N-staging accuracy was 89.7%.
that in some studies more than 30% of those polyps were A recent study from Japan found comparable accuracy in
not neoplastic [45]. In addition, a large percentage (about their series [50].
40%) of polyps less than 1 cm were neoplastic [45].
Clearly, a better screening modality was needed.
The group from Tokyo began using EUS to better Liver Lesions
characterize gallbladder polyps in the late 1980s. They One of the newer indications for EUS has been the evalua-
were able to recognize several endosonographic features tion of liver lesions. Prior to 3 years ago, it was not part of
that differentiated neoplastic from nonneoplastic polyps. our routine staging of malignancies to examine the liver for
They found that polyps that did not contain echogenic possible metastases. In 1999, the group from University of
spots, multiple microcysts, or a comet tail artifact were California at Irvine reported on a prospective study in
likely to be neoplastic. In addition, sessile lesions were which they systematically looked for liver lesions in 574
more likely to be malignant [45,46]. In a recent study, they patients who had presented for the evaluation of a possible
compared EUS with US using this characterization in gastrointestinal or pulmonary malignancy [51••]. In this
194 patients who presented with gallbladder polyps of less cohort, 14 patients (2.4%) were found to have liver lesions
than 2 cm [47]. The results were confirmed surgically in and subsequently underwent EUS-guided FNA. Strikingly,
58 patients. EUS was 97% accurate, compared with only CT had shown a liver lesion in only three of the 14 patients
76% accuracy with US. (21%). The lesions ranged from 0.8 to 5.2 cm in diameter.
A group from Korea expanded on these findings and The majority of lesions in this study were metastases from
developed a scoring system based on five EUS variables pancreatic cancer and appeared as hypoechoic lesions on
 The Role of Endoscopic Ultrasound in Hepatobiliary Disease • Schwartz and Wiersema 77

Figure 6. A, A hepatic metastasis in a patient

with pancreatic cancer is shown. Note the
hypoechoic halo around the hyperechoic
lesion. This is the most common appearance
of a metastatic lesion within the liver. B, A
hepatic metastasis in a patient with colon
cancer is shown. Lesions that are hypervascu-
lar, such as colon cancer, renal cell cancer, or
neuroendocrine tumors, usually appear to be
echo-rich on ultrasound. Note that no post-
acoustic shadowing is present. This helps to
distinguish the lesion from a hemangioma.

EUS. There were no early or late complications associated 2. Strohm W, Kurtz W, Classen M: Detection of biliary stones by
with EUS FNA. means of endosonography. Scand J Gastroenterol 1984,94
(suppl):60–64.
As a result of this study, we have changed our practice 3. Mitchell S, Clark R: A comparison of computed tomography
and now include examination of the liver during the and sonography in choledocholithiasis. Am J Roentgenol 1984,
staging evaluation (Fig. 5). As in the study by Nguyen et al. 142:729–733.
4. Amouyal P, Amouyal G, Levy P, et al.: Diagnosis of
[51••], we do not routinely give prophylactic antibiotics choledocholithiasis by endoscopic ultrasonography.
when we perform an FNA of a liver lesion unless there is an Gastroenterology 1994, 106:1062–1067.
element of cholestasis present or if the lesion is cystic. 5. Edmundowicz SA, Aliperti G, Middleton WD: Preliminary
As we obtain more experience looking at the liver during experience using endoscopic ultrasonography in the diag-
nosis of choledocholithiasis. Endoscopy 1992, 24:774–778.
EUS, several types of lesions both benign and malignant 6. Polkowski M, Palucki J, Regula J, et al.: Helical computed
have become apparent. In fact, in application it can be very tomographic cholangiography versus endosonography for
difficult to differentiate between benign and malignant suspected bile duct stones: a prospective blinded study in
non-jaundiced patients. Gut 1999, 45:744–749.
lesions based solely on endosonographic appearance. It is 7. Palazzo L, Girollet PP, Salmeron M, et al.: Value of endoscopic
important to remember that there is no pathognomonic ultrasonography in the diagnosis of common bile duct
feature that differentiates a benign from a malignant stones: comparison with surgical exploration and ERCP.
Gastrointest Endosc 1995, 42:225–231.
lesion. Based on the radiology literature, lesions that are
8. Prat F, Amouyal G, Amouyal P, et al.: Prospective controlled
hypoechoic or have a small hypoechoic rim are more likely study of endoscopic ultrasonography and endoscopic retro-
to be malignant (most cancers) (Fig. 6A) [52•]. Hyper- grade cholangiography in patients with suspected common-
echoic lesions that do not have postacoustic shadowing bile duct lithiasis. Lancet 1996, 347:75–79.
9. Norton SA, Alderson D: Prospective comparison of endo-
(ie, colon cancer) are more likely to be malignant as well scopic ultrasonography and endoscopic retrograde cholan-
(Fig. 6B). Conversely, hyperechoic lesions without a rim giopancreatography in the detection of bile duct stones.
or with postacoustic shadowing are most likely benign Br J Surg 1997, 84:1366–1369.
10. Aubertin JM, Levoir D, Bouillot JL, et al.: Endoscopic ultra-
(ie, hemangioma) sonography immediately prior to laparoscopic cholecystec-
tomy: a prospective evaluation. Endoscopy 1996, 28:667–673.
11. Shim CS, Joo JH, Park CW, et al.: Effectiveness of endoscopic
Conclusions ultrasonography in the diagnosis of choledocholithiasis
prior to laparoscopic cholecystectomy. Endoscopy 1995,
EUS is a rapidly emerging technology that has become an 27:428–432.
invaluable tool in the evaluation of various hepatobiliary 12. Erickson RA, Carlson B: The role of endoscopic retrograde
diseases, including choledocholithiasis, obstructive cholangiopancreatography in patients with laparoscopic
cholecystectomies. Gastroenterology 1995, 109:252–263.
jaundice of unknown cause, biliary strictures, cholangio- 13.• Sahai AV, Mauldin PD, Marsi V, et al.: Bile duct stones and
carcinoma, gallbladder polyps, and liver lesions. As the laparoscopic cholecystectomy: a decision analysis to assess
technology continues to improve and the expertise to the roles of intraoperative cholangiography, EUS, and ERCP.
Gastrointest Endosc 1999, 49:334–343.
perform EUS and EUS-guided FNA becomes more avail- The most appropriate management strategy for patients with choledo-
able, these modalities should continue to be components cholithiasis is dependent on their risk for the disease. High-risk
of routine clinical care. patients should undergo ERCP, whereas intermediate-risk patients
are better assessed with EUS first.
14. Frossard JL, Hadengue A, Amouyal G, et al: Choledocholithia-
sis: a prospective study of spontaneous common bile duct
References and Recommended Reading stone migration. Gastrointest Endosc 2000, 51:175–179.
Papers of particular interest, published recently, have been 15.• Liu CL, Lo CM, Chan JK, et al.: Detection of choledocholithia-
highlighted as: sis by EUS in acute pancreatitis: A prospective evaluation in
100 consecutive patients. Gastrointest Endosc 2001,
• Of importance
54:325–330.
•• Of major importance Sensitivity and specificity rates for EUS compared with ERCP in
identifying choledocholithiasis were 97% and 98% for EUS, versus
1. Tio TL, Tytgat GN: Endoscopic ultrasonography of bile duct
97% and 95% for ERCP. The authors conclude that EUS should be
malignancy and the preoperative assessment of local resect-
considered prior to ERCP to reduce the risk associated with diagnostic
ability. Scand J Gastroenterol 1986, 123(suppl):151–157.
ERCP in the setting of suspected gallstone pancreatitis.
78 Liver

16. Liu CL, Lo CM, Chan JK, et al.: EUS for detection of occult 34. Tamada K, Ido K, Ueno N, et al.: Assessment of portal vein
cholelithiasis in patients with idiopathic pancreatitis. invasion by bile duct cancer using intraductal ultrasono-
Gastrointest Endosc 2000, 51:28–32. graphy. Endoscopy 1995, 27:573–578.
17. Frossard JL, Sosa-Valencia L, Amouyal G: Usefulness of endo- 35. Tamada K, Ido K, Ueno N, et al.: Assessment of hepatic artery
scopic ultrasonography in patients with ‘idiopathic’ acute invasion by bile duct cancer using intraductal ultrasono-
pancreatitis. Am J Med 2000, 109:196–200. graphy. Endoscopy 1995, 27:579–583.
18. Chak A, Hawes RH, Cooper GS, et al.: Prospective assessment 36. Tamada K, Ueno N, Ichiyama M, et al.: Assessment of
of the utility of EUS in the evaluation of gallstone pancrea- pancreatic parenchymal invasion by bile duct cancer using
titis. Gastrointest Endosc 1999, 49:599–604. intraductal ultrasonography. Endoscopy 1996, 28:492–496.
19. Dill JE, Hill S, Callis J, et al.: Combined endoscopic ultra- 37. Tamada K, Wada S, Ohashi A, et al.: Intraductal US in assess-
sound and stimulated biliary drainage in cholecystitis and ing the effects of radiation therapy and prediction of patency
microlithiasis: diagnoses and outcomes. Endoscopy 1995, of metallic stents in extrahepatic bile duct carcinoma.
27:424–427. Gastrointest Endosc 2000, 51:405–411.
20. Dahan P, Andant C, Levy P, et al.: Prospective evaluation of 38. Hyodo T, Yamanaka T: Intraductal ultrasonography in six
endoscopic ultrasonography and microscopic examination patients with endoscopic biliary stenting. J Gastroenterol 1999,
of duodenal bile in the diagnosis of cholecystolithiasis in 34:105–110.
45 patients with normal conventional ultrasonography. 39. Materne R, Van Beers BE, Gigot JF, et al.: Extrahepatic biliary
Gut 1996, 38:277–281. obstruction: magnetic resonance imaging compared with
21. Ohashi A, Ueno N, Tamada K, et al.: Assessment of residual endoscopic ultrasonography. Endoscopy 2000, 32:3–9.
bile duct stones with use of intraductal US during endo- 40. Amouyal P, Palazzo L, Amouyal G, et al.: Endosonography:
scopic balloon sphincteroplasty: comparison with balloon promising method for diagnosis of extrahepatic cholestasis.
cholangiography. Gastrointest Endosc 1999, 49:328–333. Lancet 1989, 2:1195–1198.
22. Ueno N, Nishizono T, Tamada K, et al.: Diagnosing extra- 41. Dancygier H, Nattermann C: The role of endoscopic ultra-
hepatic bile duct stones using intraductal ultrasonography: sonography in biliary tract disease: obstructive jaundice.
a case series. Endoscopy 1997, 29:356–360. Endoscopy 1994, 26:800–802.
23. Gress F, Chen YK, Sherman S, et al.: Experience with a catheter- 42.• Erickson RA, Garza AA: EUS with EUS-guided fine-needle
based ultrasound probe in the bile duct and pancreas. aspiration as the first endoscopic test for the evaluation of
Endoscopy 1995, 27:178–184. obstructive jaundice. Gastrointest Endosc 2001, 53:475–484.
24. Tamada K, Ueno N, Tomiyama T, et al.: Characterization of This well-written cost analysis demonstrates that the most cost-
biliary strictures using intraductal ultrasonography: compar- efficient diagnostic strategy for patients presenting with unexplained
ison with percutaneous cholangioscopic biopsy. Gastrointest obstructive jaundice is to perform EUS as the first endoscopic test.
Endosc 1998, 47:341–349. 43. Burtin P, Palazzo L, Canard JM, et al.: Diagnostic strategies
25.• Menzel J, Poremba C, Dietl KH, Domschke W: Preoperative for extrahepatic cholestasis of indefinite origin: endoscopic
diagnosis of bile duct strictures: comparison of intraductal ultrasonography or retrograde cholangiography? Results of
ultrasonography with conventional endosonography. a prospective study. Endoscopy 1997, 29:349–355.
Scand J Gastroenterol 2000, 35:7–82. 44. Jorgensen T, Jensen KH: Polyps in the gallbladder: a preva-
IDUS was 81% accurate in classification of biliary strictures as benign lence study. Scand J Gastroenterol 1990, 25:281–286.
or malignant. 45. Sugiyama M, Atomi Y, Kuroda A, et al.: Large cholesterol
26. Yasuda K, Nakajima M, Kawai K: Disease of the biliary tract polyps of the gallbladder: diagnosis by means of US and
and the papilla of Vater. In Endoscopic Ultrasonography in Gastro- endoscopic US. Radiology 1995, 196:493–497.
enterology. Edited by Kawai K. Tokyo: Igaku-Shoin; 1988:87–95. 46. Sugiyama M, Xie XY, Atomi Y, Saito M: Differential diagnosis
27. Mukai H, Yasuda K, Nakajima M: Tumors of the papilla of small polypoid lesions of the gallbladder: the value of
and distal common bile duct: diagnosis and staging by endo- endoscopic ultrasonography. Ann Surg 1999, 229:498–504.
scopic ultrasonography. Gastrointest Endosc Clin North Am 47. Sugiyama M, Atomi Y, Yamato T: Endoscopic ultrasonography
1995, 5:763–772. for differential diagnosis of polypoid gall bladder lesions:
28. Tio TL, Coene PP, den Hartog Jager FC, Tytgat GN: Preoperative analysis in surgical and follow up series. Gut 2000,
TNM classification of esophageal carcinoma by endosono- 46:250–254.
graphy. Hepatogastroenterology 1990, 37:376–381. 48. Choi WB, Lee SK, Kim MH, et al.: A new strategy to predict the
29. Tio TL, Cheng J, Wijers O: Endosonograhpic TMN staging of neoplastic polyps of the gallbladder based on a scoring
extrahepatic bile duct cancer: comparison with pathological system using EUS. Gastrointest Endosc 2000, 52:372–379.
staging. Gastroenterology 1991, 100:1351–1361. 49. Mitake M, Nakazawa S, Naitoh Y, et al.: Endoscopic ultra-
30. Qilian Z, Weidong N, Lanbo Z, Jinyu L: Endoscopic ultra- sonography in diagnosis of the extent of gallbladder
sonography assessment in preoperative staging for carci- carcinoma. Gastrointest Endosc 1990, 36:562–566.
noma of ampulla of Vater and extrahepatic bile duct. 50. Fujita N, Noda Y, Kobayashi G, et al.: Diagnosis of the depth of
Chin Med J 1996, 109:622–625. invasion of gallbladder carcinoma by EUS. Gastrointest Endosc
31. Fujita N, Noda Y, Kobayashi G, et al.: Staging of bile duct 1999, 50:659–663.
carcinoma by EUS and IDUS. Endoscopy 1998, 30:A132–A134. 51.•• Nguyen P, Feng JC, Chang KJ: Endoscopic ultrasound (EUS)
32. Fritscher-Ravens A, Broering DC, Sriram PV, et al.: EUS-guided and EUS-guided fine-needle aspiration (FNA) of liver lesions.
fine-needle aspiration cytodiagnosis of hilar cholangiocarci- Gastrointest Endosc 1999, 50:357–361.
noma: a case series. Gastrointest Endosc 2000, 52:534–540. The initial study that showed that EUS and EUS-guided FNA were a
33. Sugiyama M, Hagi H, Atomi Y, Saito M: Diagnosis of portal sensitive and safe means of evaluating liver lesions.
venous invasion by pancreatobiliary carcinoma: value of 52.• Harvey CJ, Albrecht T: Ultrasound of focal liver lesions.
endoscopic ultrasonography. Abdom Imaging 1997, Eur Radiol 2001, 11:1578–1593.
22:434–438. A concise review of the ultrasound appearance of various liver lesions.