THE AMERICAN JOURNAL OF GASTROENTEROLOGY
1, 2000
© 2000 by Am. Coll. of Gastroenterology
Published by Elsevier Science Inc.
Role of ERCP in Asymptomatic Orthotopic Liver
Transplant Patients With Abnormal Liver Enzymes
Devin E. Eckhoff, M.D., Todd H. Baron, M.D., William G. Blackard, M.D., Desiree E. Morgan, M.D.,
Ralph Crowe, M.D., Marty Sellers, M.D., Brendan McGuire, M.D., Juan L. Contreras, M.D., and
J. Steve Bynon, M.D.
University of Alabama at Birmingham, Birmingham, Alabama
OBJECTIVE: The safety and efficacy of endoscopic retro-
grade cholangiopancreatography (ERCP) in the evaluation
and management of biliary tract complications after ortho-
topic liver transplantation (OLT) have been previously dem-
onstrated. However, the role of ERCP in evaluating asymp-
tomatic OLT patients with abnormal liver enzymes with a
previously normal biliary tree remains poorly defined. We
sought to assess the utility of ERCP in this subset of pa-
tients.
METHODS: A retrospective analysis of asymptomatic OLT
patients with abnormal liver enzymes evaluated by ERCP
was undertaken. In addition to ERCP, all these patients
had a diagnostic abdominal Doppler ultrasound, and a
percutaneous liver biopsy. All patients had choledocho-
choledochostomy at the time of transplant and normal
T-tube cholangiograms 3 months postoperatively. A radi-
ologist, blinded to clinical findings, interpreted the ultra-
sound as normal, biliary dilation, or vascular abnormal-
ities. The same radiologist interpreted ERCP findings. A
pathologist, blinded to clinical findings, graded liver bi-
opsies as normal, diagnostic, or abnormal but nondiag-
nostic.
RESULTS: Twenty-two patients underwent 23 ERCPs.
Twenty-two of the 23 ERCPs were normal (96%), and one
abnormal ERCP finding did not explain the liver enzyme
abnormality. Liver biopsy was diagnostic in 13 of 22 (57%)
and in each case the ERCP was normal. The remaining 10
liver biopsies were abnormal but nondiagnostic. Ultrasound
was abnormal in five of 22 cases, but in the three cases
suggesting biliary dilation, the ERCP was interpreted as
normal.
CONCLUSION: Routine use of ERCP in evaluation of asymp-
tomatic OLT patients with liver function test abnormalities
and normal cholangiograms at 3 months was not diagnos-
tically useful. In this subset of patients, liver biopsy was
usually abnormal and frequently diagnostic and should be
the initial invasive diagnostic procedure. (Am J Gastroen-
terol 2000;95:141–144. © 2000 by Am. Coll. of Gastroen-
terology)
Vol. 95, No.
ISSN 0002-9270/00/$20.00
PII S0002-9270(99)00734-0
INTRODUCTION
Orthotopic liver transplantation (OLT) was introduced in
1963 and biliary reconstruction was rapidly recognized as
the technical Achilles heel of this surgical procedure (1).
Despite improvements in the preservation solutions, surgi-
cal techniques, and understanding of vascular supply, biliary
tract complications still account for 2–35% of post-OLT
morbidity (2– 6). The increasing availability and expertise in
endoscopic retrograde cholangiopancreatography (ERCP)
has allowed for effective noninvasive management of these
biliary complications, particularly bile leaks, choledocholi-
thiasis, and anastomotic strictures (7, 8). However, the role
of ERCP in evaluating allograft dysfunction in asymptom-
atic OLT patients with a previously normal biliary tree
remains poorly defined. In this report, we attempt to assess
the diagnostic efficacy of ERCP for evaluation of abnormal
liver function tests (LFTs) in asymptomatic OLT patients
whose cholangiograms were previously shown to be normal
3 months after transplantation.
MATERIALS AND METHODS
Between March 1992 and January 1998, 400 patients un-
derwent OLT at the University of Alabama at Birmingham;
51% had duct- to-duct anastomosis. For the purpose of this
study, we excluded patients with indwelling T-tubes, clinical
cholangitis, or previously recognized bile duct pathology.
The remaining 22 patients with abnormal liver enzymes
(elevated alkaline phosphatase, bilirubin, and/or transami-
nases), and without symptoms of pain referable to the biliary
tree or fever are the subjects of this study. All patients had
undergone duct-to-duct anastomosis using a T-tube, which
was removed between 3 and 6 months posttransplant (after
a cholangiogram revealed no biliary tract abnormalities).
Some patients underwent an ERCP and an ultrasound de-
spite a specific diagnosis on biopsy if it was believed the
diagnosis could not adequately explain the abnormal liver
enzymes.
To determine if the pattern of enzyme abnormalities was
helpful in defining the cause of the biochemical changes, the
patients’ liver enzymes were categorized into three groups:
142 Eckhoff et al.
Table 1. Demographic and Biochemical Data of Asymptomatic Orthotopic Liver Transplant (OLT) Patients With Abnormal Liver
Enzymes
Cholestatic
(n ⫽ 5)
Age, yr 37.6 ⫾ 16.1
Male:Female ratio 1:4
Time since OLT (months) 26.8 ⫾ 17.9
Total bilirubin (mg/dl) 0.5 ⫾ 0.12
NL ⫽ (0.0–1.0)
Alkaline phosphatase (U/L) 198.0 ⫾ 22.3
NL ⫽ (39–117)
AST (U/L) 32.0 ⫾ 4.5
NL ⫽ (0.37)
ALT (U/L) 39.0 ⫾ 6.9
NL ⫽ (7–56)
AST ⫽ aspartate aminotransferase; ALT ⫽ alanine aminotransferase; NL ⫽ normal.
cholestatic, hepatocellular, and mixed. The cholestatic en-
zyme abnormalities were defined as an alkaline phosphatase
⬎1.5 times the upper limit of normal ⫾ elevated bilirubin.
The hepatocellular group was defined as abnormal transami-
nases ⬎3 times the upper limit of normal, with alkaline
phosphatase levels ⬍1.5 times the upper limit of normal.
The mixed group was defined as a mixture of both types of
enzyme abnormalities. All patients underwent duplex Dopp-
ler abdominal ultrasound, percutaneous liver biopsy, and
ERCP. The ultrasound and ERCP were reviewed by an
experienced radiologist in a blinded fashion. The ultrasound
was interpreted as normal if there was no biliary dilation
(common bile duct ⬍10 mm and no intrahepatic dilation) or
vascular abnormality. Liver biopsies were reviewed in a
blinded fashion by a pathologist and interpreted as normal;
diagnostic, if abnormal with a specific diagnosis (i.e., hep-
atitis, rejection, bile duct obstruction, and/or vascular in-
sult); or nondiagnostic, if abnormal but without a specific
diagnosis.
RESULTS
Twenty-two patients (seven men, 15 women) underwent 23
diagnostic ERCPs. The average age of these patients was
40.9 yr. Indications for transplantation included cryptogenic
cirrhosis (five), hepatitis C virus (HCV, five), primary bil-
iary cirrhosis (four), Laennec’s cirrhosis (two), autoimmune
cirrhosis (one), hepatitis A virus (HAV, one), hepatitis B
virus (HBV, one), Wilson’s (one), glycogen storage disease
(one), and aldomet toxicity (one). The mean time from
transplantation was 20.4 months (Table 1). No patient in this
study had a complication related to their diagnostic proce-
dure, either ERCP or liver biopsy, and the complication rate
for these procedures at our department is similar to previ-
ously published reports (9, 10).
Seven patients had liver enzyme abnormalities catego-
rized as hepatocellular with serum bilirubin 0.8 ⫾ 2.3
(mean ⫾ SEM), alkaline phosphatase 154.0 ⫾ 37.6 U/L,
aspartate aminotransferase (AST) 244.0 ⫾ 168.3 (u/L), and
alanine aminotransferase (ALT) 299.9 ⫾ 239.6 U/L. These
AJG – Vol. 95, No. 1, 2000
Hepatocellular Mixed Total
(n ⫽ 7) (n ⫽ 10) (n ⫽ 22)
44.4 ⫾ 10.4 40.0 ⫾ 8.4 40.9 ⫾ 5.2
1:6 5:5 7:15
19.0 ⫾ 16.9 20.8 ⫾ 18.3 20.4 ⫾ 17.8
0.8 ⫾ 2.3 8.74 ⫾ 4.2 4.4 ⫾ 1.9
154.0 ⫾ 37.6 416.3 ⫾ 169.5 283.3 ⫾ 84.5
244.0 ⫾ 168.3 387.3 ⫾ 138.3 260.7 ⫾ 94.9
299.9 ⫾ 239.6 533.4 ⫾ 229.6 351.3 ⫾ 118.5
patients underwent a total of seven ERCPs. Six patients had
an ultrasound that was considered normal. This was con-
firmed by ERCP in all six cases (100%). In two of the six
cases with normal ultrasound and normal ERCP, liver bi-
opsy was diagnostic. Both cases revealed acute rejection.
The remaining four patients had abnormal but nondiagnostic
liver biopsies and demonstrated portal inflammatory re-
sponse (PIR) only. One patient in this hepatocellular group
had an abnormal ultrasound, suggesting a dilated extrahe-
patic tree; however, the ERCP was normal and the liver
biopsy demonstrated acute rejection. Table 1 summarizes
demographic and biochemical data for all groups.
The group with mixed liver enzymes comprised 10 pa-
tients who underwent 11 ERCPs. These patients had serum
bilirubin 8.7 ⫾ 4.2 mg/dl (mean ⫾ SEM), alkaline phos-
phatase 416.3 ⫾ 169.5 U/L, AST 387.3 ⫾ 138.3 U/L, and
ALT 533.4 ⫾ 229.6 U/L. Ultrasound was normal in eight of
10 cases with ERCP, confirming this in all cases (100%). In
six of the eight cases, the liver biopsy was diagnostic, with
diagnosis including acute rejection (three), recurrent viral
hepatitis (two), and acute and chronic rejection (one). The
other two patients in this group who demonstrated normal
ultrasound and ERCP findings had a nondiagnostic liver
biopsy, demonstrating portal inflammatory response. The
two remaining patients in this mixed liver enzyme group had
abnormal ultrasounds. One of these patients had an ultra-
sound suggesting extrahepatic dilation. The ERCP was nor-
mal in this patient, but the liver biopsy was diagnostic,
demonstrating acute rejection. The other abnormal ultra-
sound had low resistive indexes, suggestive of hepatic artery
stenosis. It is our current policy not to revise hepatic arteries
based on an ultrasound finding, and no further evaluation of
the hepatic artery was done in this case. In this patient the
ERCP demonstrated a biliary venous communication and
was thought to be unrelated to LFT abnormalities; therefore,
an arteriogram to rule out hemobilia was not done. Liver
biopsy in this patient revealed cholestasis.
Five patients had liver enzymes classified as cholestatic,
with serum bilirubin 0.56 ⫾ 0.12 mg/dl (mean ⫾ SEM) and
alkaline phosphatase 198.0 ⫾ 22.3 U/L, AST 32.0 ⫾ 4.5
AJG – January, 2000
(U/L), and ALT 39 ⫾ 6.9 (U/L). Three of the five patients
had a normal ultrasound and confirmed by ERCP. Of these
three, one liver biopsy demonstrated acute rejection and two
were abnormal but nondiagnostic, both exhibiting portal
inflammatory response. One patient had an ultrasound dem-
onstrating moderate dilation of the biliary system; however,
ERCP was normal and liver biopsy was diagnostic, showing
acute rejection. The ultrasound of the fifth patient suggested
hepatic artery stenosis and the ERCP in this patient was
normal. Liver biopsy exhibited cholestatic features. Thus,
all ERCPs in these patients were normal despite a choles-
tatic pattern of LFT abnormalities.
Overall, in asymptomatic OLT patients with abnormal
LFTs, 17 of 22 screening ultrasound examinations were
normal and confirmed by ERCP in all 22 cases. Three
ultrasounds suggested biliary dilation, but the ERCP was
considered normal in all three, as biliary dilation was not
demonstrated (common bile duct ⬍ 10 mm). The other two
abnormal ultrasounds suggested vascular abnormalities (he-
patic artery stenosis) with ERCP normal in one and abnor-
mal in the other. In this latter patient, the ERCP abnormal-
ities did not explain the elevated laboratory values or require
intervention. In fact, only one of 23 ERCPs was abnormal
(demonstrated biliary venous fistula). However, in this case
the ERCP did not suggest biliary dilation, explain the ele-
vated laboratory values, or, more importantly, affect patient
management. In contrast, liver biopsy was diagnostic in 13
of 22 patients (57%) and the remaining biopsies were all
abnormal though not diagnostic of a specific treatable cause.
Eight of the nine nondiagnostic liver biopsies demonstrated a
portal inflammatory response. Four patients in this group were
transplanted for viral hepatitis and this probably represents
early recurrent disease. Specific diagnosis in the other five
patients was not possible but could potentially represent recur-
rent disease (cryptogenic and primary biliary cirrhosis).
DISCUSSION
Biliary tract complications after OLT are frequently multi-
factorial and most often related to vascular, technical, or
immunological factors (11). Patients who develop biliary
tract complications have three common presentations: fever
with fluctuating liver function tests, cholangitis, and gradual
deterioration of liver function tests without symptoms (12,
13). These presentations, however, are not pathognomonic
of bile duct pathology and must be differentiated from other
causes of abnormal liver function tests, including rejection
(acute and chronic), viral infection, primary graft dysfunc-
tion, and hepatic artery thrombosis.
The use of ERCP for evaluation of abnormal LFTs in
asymptomatic OLT patients whose cholangiograms were
previously shown to be normal is of limited value (statistical
analysis was not done because of small sample size). ERCP
was abnormal in only one case, and, in this patient, the
abnormality neither correlated with ultrasound findings nor
provided an explanation for the biochemical abnormalities.
ERCP in Asymptomatic OLT Patients 143
In the other 18 patients, the ERCP was interpreted as nor-
mal. However, manometry was not routinely performed and
therefore sphincter or Oddi dysfunction cannot be excluded.
Based on these findings, one could postulate that use of
ERCP in patients with previously demonstrated normal biliary
ducts should be reserved for situations in which liver biopsy is
consistent with large duct obstruction or abnormal and other-
wise unexplained LFTs have persisted. These findings proba-
bly can be extended to patients who have had a duct-to-duct
anastomosis with or without a T-tube, who, ⱖ3 months post-
transplant, have had normal biliary tracts demonstrated by
either ERCP or percutaneous transhepatic cholangiogram.
In this study, ultrasound was not helpful in determining
biliary pathology. The ultrasound was normal in 17 of 22
patients and, in the remaining five patients, abnormal biliary
duct dilation on ultrasound was not supported by findings on
subsequent ERCP. Other reports have confirmed that ultra-
sound is both an insensitive and nonspecific screening test of
biliary disease in OLT patients and by one report is only
46% sensitive for biliary tract complications (13). Because
most OLT patients with a screening ultrasound suggesting
biliary dilation have cholangiograms demonstrating stable
or only slightly increased extrahepatic bile ducts, some
authors have suggested that mild biliary dilation not asso-
ciated with a specific cause of obstruction or hepatic dys-
function does not portend significant biliary obstruction and
is clinically benign (11). Possible explanations for this non-
obstructive biliary dilation include variations in the under-
lying duct laxity, subclinical low-grade obstruction, and the
extrahepatic duct serving as a reservoir of bile in the absence
of the gallbladder (14).
In contrast to the utility of ultrasound and ERCP, liver
biopsy in this population was diagnostic in 13 of 22 patients
(57%). Significantly, in each case, the liver biopsy sug-
gested an appropriate clinical course of action. In the other
10 cases, the liver biopsy was abnormal but nondiagnostic
(i.e., failed to satisfy specific pathological criteria), but was
suggestive of a specific pathology. However, when the find-
ings on liver biopsy were taken in the clinical settings, the
biopsy was believed to adequately explain the liver enzyme
abnormalities and no further diagnostic or therapeutic inter-
ventions were required. The data suggests that liver biopsy
is frequently diagnostic in this situation (57% in this study)
and should be the preferred initial approach.
These results suggest that bile duct complications, in the
absence of hepatic artery thrombosis, tend to present within
the first 3 months after transplantation. Therefore, one can
realistically propose that a normal cholangiogram, or ERCP
in the cases where T-tubes were not used, obtained 3 months
after transplantation may serve to define a patient population
that will not require cholangiographic reevaluation unless
symptoms, ultrasound, or liver biopsy more definitely dem-
onstrate a bile duct problem. Based on our observation, we
would suggest that all asymptomatic OLT patients with
abnormal liver enzymes undergo a liver biopsy as the initial
evaluation (Fig. 1). If the liver biopsy is diagnostic, treat-
144 Eckhoff et al.
Figure 1. Suggested algorithmic approach to asymptomatic Orthotopic Liver Transplant (OLT) patients with abnormal liver enzymes with
previously normal cholangiograms.
ment should be instituted as appropriate. If the biopsy is
nondiagnostic, but abnormal enough to explain the findings,
we would propose observing the patient over time and
considering further diagnostic studies such as endoscopic
ultrasound, magnetic resonance cholangiopancreatography,
ultrasound with Doppler, or ERCP only if abnormalities
persist. Finally, if the biopsy suggests bile duct obstruction,
ERCP should be performed.
Reprint requests and correspondence: Devin E. Eckhoff, M.D.,
University of Alabama at Birmingham, 619 South 19th Street,
LHRB 748, Birmingham, AL 35294-0007.
Received Mar. 16, 1999; accepted Aug. 25, 1999.
REFERENCES
1. Calne, RY. A new technique for biliary drainage in orthotopic
liver transplantation utilizing the gallbladder as a pedicle graft
conduit between the donor and recipient common bile ducts.
Ann Surg 1976;184:605.
2. Lerut J, Gordon RJ, Iwatsuki S, et al. Biliary tract complica-
tions in human orthotopic liver transplantation. Transplanta-
tion 1987;43:47–51.
3. Klein AS, Savader S, Burdick JF, et al. Reduction of morbidity
and mortality from biliary complications after liver transplan-
tation. Hepatology 1991;14:818 –23.
4. Lopez RR, Benner KG, Ivancev K, et al. Management of
biliary tract complications after liver transplantation. Am J
Surg 1992;163:519 –24.
AJG – Vol. 95, No. 1, 2000
5. Porayko MK, Kondo M, Steers JL. Liver transplantation: Late
complications of the biliary tract and their management. Semin
Liv Dis 1995;15:139 –55.
6. Koneru B, Tzakis A, Bowman J, et al. Postoperative surgical
complications. Gastroenterol Clin North Am 1988;17:71– 86.
7. Rolles K, Dawson K, Novell R, et al. Biliary anastomosis after
liver transplantation does not benefit from T-tube splintage.
Transplantation 1994;57:402– 4.
8. Popescu I, Sheiner P, Mor E, et al. Biliary complications in
400 cases of liver transplantation. Mt Sinai J Med 1994;61:
57– 60.
9. Erickson RA, Carlson B. The role of endoscopic retrograde
cholangiopancreatography in patients with laparoscopic cho-
lecystectomies. Gastroenterology 1995;109:252– 63.
10. McGill DB, Rakela J, Zinsmeister AR, et al. A 21-year expe-
rience with major hemorrhage after percutaneous liver biopsy.
Gastroenterology 1990;99:1396 – 400.
11. Donovan J. Nonsurgical management of biliary tract disease
after liver transplantation. Gastroenterol Clin N Am 1993;2:
317–36.
12. Van Thiel DH, Fagiuoli S, Wright HI, et al. Biliary compli-
cations of liver transplantation. Gastrointest Endosc 1993;39:
455– 60.
13. Zajko AB, Zemel G, Skolnick ML, et al. Percutaneous trans-
hepatic cholangiography rather than ultrasound as a screening
test for postoperative biliary complications in liver transplant
patients. Transplant Proc 1988;20:678 – 81.
14. Campbell WL, Foster RG, Miller WJ, et al. Changes in ex-
trahepatic bile duct caliber in liver transplant recipients with-
out evidence of biliary obstruction. Am J Roentgenol 1992;
158:997–1000.