Surgical Science, 2012, 3, 489-493
http://dx.doi.org/10.4236/ss.2012.310097 Published Online October 2012 (http://www.SciRP.org/journal/ss)
Comparison of Endoscopic Retrograde
Cholangiopancreatography (ERCP) and Magnetic
Resonance Cholangiopancreatography
(MRCP) in Bile Duct Imaging
Mehmet Ali Eryılmaz1, Ömer Karahan1, İsmet Tolu2, Ahmet Okuş1, Serden Ay2*, Barış Sevinç1,
Ahmet Hakan Halıcı1
1Konya Training and Research Hospital General Surgery Clinic, Konya, Turkey
2Konya Training and Research Hospital Radiology Clinic, Konya, Turkey
Received August 10, 2012; revised September 11, 2012; accepted September 22, 2012
ABSTRACT
Purpose: The aim of this study was to compare magnetic resonance cholangiopancreatography (MRCP) with endo-
scopic retrograde cholangiopancreatography (ERCP) in diagnosing bile duct pathologies. Materials and Methods: We
documented the data of 171 patients with both ERCP and MRCP between January 2009 and December 2010 at the
Konya Education and Research Hospital. Results: Of the 171 patients, 100 (58.5%) were female and 71 (41.5%) were
male. The median age was 63 (55 to 89). ERCP was used to diagnose bile duct stones in 102 (59%) patients, bile duct
tumour in 14 (8%) patients, hydatic cysts opening up to the bile duct in 4 (2%) patients and bile duct stenosis in 3 (1.8%)
patients. For the detection of bile duct stones, MRCP had a sensitivity of 92%, a specificity of 74% and a diagnostic
accuracy of 83%. For bile duct tumours, MRCP had a sensitivity of 85%, a specificity of 98% and a diagnostic accuracy
rate of 92%. Conclusion: In our centre, the results of MRCP and ERCP were similar for the last two years. However,
MRCP was superior with respect to diagnosis as it was cheaper and non-invasive. Thus, ERCP should be preferred for
therapeutic processes.
Keywords: ERCP; MRCP; Bile Duct Disease; Diagnosis; Treatment
1. Introduction
In the diagnosis of bile duct pathologies, clinical labora-
tory findings and imaging studies play important roles.
Frequently used imaging techniques are ultrasonography
(USG), magnetic resonance cholangiopancreatography
(MRCP) and endoscopic retrograde cholangiopancrea-
tography (ERCP). Abdominal USG is useful in detecting
bile duct morphology; however, its sensitivity in detect-
ing the ethiology of bile duct obstruction is low. In the
diagnosis of bile duct stones, USG must be combined
with other imaging methods [1]. The advantages of
MRCP are it is not invasive or radionuclear and provides
3D images [2]. ERCP is another way of detecting those
pathologies. Although it is an invasive method, some
therapeutic investigations like sphincterotomy, stone
extraction and stent placement can be performed while
performing a diagnostic study [3]. According to some
authors, 10% of MRCP and 5% of ERCP performed in
our country are optimal [4]. Therefore, we aimed to de-
tect the differences and the advantages of the two tech-
niques by comparing their diagnostic accuracy, cost, time
and morbidity in our hospital.
2. Materials and Methods
2.1. Patients
At the Konya Training and Research Hospital between
January, 2009 and December 2010, 655 patients under-
went MRCP, and 283 patients, ERCP, for bile duct pa-
thologies. 171 patients who had both MRCP and ERCP
were included in the study. MRCP was used on patients
who had clinical jaundice symptoms with or without pain
in the upper right quadrant of the abdomen, and labora-
tory results showing higher values than GGT = 40 IU/L,
ALP = 120 IU/L, direct bilirubin = 0.2 mg/dl, and
USG-detected choledoch duct wider than 7 mm, who had
pre-diagnoses of bile duct disorders. Those who received
diagnoses such as bile duct stones or stenosis, bile leak-
age, parasitosis, pancreatic tumors, papilla water tumors,
*Corresponding author.
C
opyright © 2012 SciRes. SS

M. A. ERYILMAZ ET AL. 490
underwent ERCP in order to perform treatment or biopsy.
Age, gender, and details of MRCP and ERCP reports of
the patients included in the study were obtained retro-
spectively from the hospital automation system.
2.2. MRCP Technique
MRCP was performed before ERCP. In order to have an
empty stomach and a maximally full gall bladder, the
patients didn’t have any food for 6 hours before the pro-
cedure. The MRCP device was 1.5 Tesla and had 4 phase
channels (Avanto, Simens Medical Systems, Germany).
The patients were not sedated for the MRCP procedure.
Total procedure duration was approximately 4-5 minutes.
MRCP images were evaluated by an experienced radi-
ologist (Figures 1 and 2). No contrast material was used
to obtain the MRCP images. The Social Security Institu-
tion paid TL 65 for the MRCP procedure in 2010.
Figure 1. T2 axial MRCP image showing a bile duct stone.
Figure 2. 3D MRCP image showing a bile duct stone.
2.3. ERCP Technique
ERCP was performed in order to confirm bile duct pa-
thologies and to provide treatment, after informed con-
sent was received from the patients. The equipment used
during the ERCP procedures were C-Arm (Siemens, In-
dia), bipolar cautery (Olympus, Germany) and duodeno
video endoscopy (Olympus, Japan). Every patient was
admitted for treatment after fasting for eight hours. Par-
enteral Buscopan was administered to all patients. Endo-
scopic sphincterotomy (ES) and stone extraction were
performed in the cases of stone, debris or parasitosis. In
cases of tumoral lesions biopsy was performed for diag-
nostic confirmation, and stent placement was carried out
where needed. The ERCP procedure took approximately
30 minutes. The Social Security Institution paid TL 284
for ERCP in 2010.
2.4. Statistics
Sensitivity, specifity, positive and negative predictive
values and total accuracy rates were used. P value was
calculated by using SPSS (SPSS Inc., v16.0, 2009, Chi-
cago, IL, USA). A P value of <0.05 was accepted as sig-
nificant.
3. Results
One hundred (58.5%) of the hundred and seventy one
patients were female and 71 (41.5%) were male. The
median age was 63 years (55 to 89). In nine patients
(5.26%) the ERCP procedure could not be carried out. In
six of these patients, the choledoch could not be cannu-
lated, and 3 had pyloric stenosis. There was no failure
during the MRCP procedures. The results of MRCP and
ERCP in diagnosing bile duct disorders are given in Ta-
bles 1 and 2.
Of the patients who had both MRCP and ERCP,
MRCP detected bile duct stones in 103, and ERCP de-
tected 102. Endoscopic sphincterotomy and stone extrac-
tion with ERCP were performed on these patients. Of the
32 (19%) patients diagnosed by MRCP, ERCP found 26
(15.20%) to have dilated bile ducts. However, neither
method was able to distinguish pathology to explain the
dilatation. ERCP detected tumors in 14 out of the 15
(8.77%) patients diagnosed by MRCP with tumors, and
biopsies were taken by ES. ERCP could not be per-
formed on one patient due to inability to access the duo-
denum because of pyloric stenosis.
The diagnosis of 4 of the 5 (2.92%) patients diagnosed
by MRCP was confirmed by ERCP. ES was performed,
and cystic vesicles and membranes removed. Four pa-
tients (2.33%) were diagnosed with bile duct stenosis,
two were treated with ES, and one with ES + balloon
dilatation. Stent placement was performed on one.

M. A. ERYILMAZ ET AL. 491
Table 1. MRCP and ERCP results of patients with bile duct disease.
MRCP n (%) ERCP n (%) P
Unsuccessful investigation 0 9 (5) 0.002
Normal bile ducts 6 (4) 5 (3) 0.759
Bile duct stone 103 (60) 102 (59) 0.912
Bile duct dilatation (distal stenosis) 32 (19) 26 (15) 0.387
Bile duct tumor 15 (9) 14 (8) 0.846
Bile duct cystic disease 5 (3) 4 (2) 0.736
Bile duct stenosis 4 (2) 3 (1.7) 0.703
Bile duct inflammatory disease 2 (1.5) 4 (2) 0.410
Others 4 (2) 4 (2) 1
TOTAL 171 (100) 171 (100)
Table 2. Sensitivity test for MRCP.
Positive predictive
value
Negative predictive
value
Sensitivity
(%)
Specifity
(%)
Total accuracy
(%)
Bile duct stone 0.79 0.89 92 74 83
Bile duct tumor 0.80 0.98 85 98 92
Bile duct cystic disease 0.80 1 100 99 99
Bile duct stenosis 0.75 1 100 99 99
Complications developed in 5 (2.92%) of the cases
that underwent stone extraction with ERCP. Two patients
suffered duodenum perforation. Sphincteroplasty and
primary repairs of the perforation were carried out
through open surgery in one case. The other case recov-
ered through conservative treatment. One patient devel-
oped hemorrhage, another developed pancreatitis. Both
recovered through conservative treatment. The stone
basket compacted in one patient. Both the basket wire
and the stone were extracted with open surgery. Neither
technique gave rise to mortality.
It was observed that ERCP and MRCP led to similarly
accurate diagnoses, and there was no significant differ-
ence between the two techniques (P > 0.05) (Tables 1 and
2).
4. Discussion
MRCP is an important diagnostic tool for biliary pa-
thologies [4]. ERCP is an invasive procedure, and is used
for treatment rather than diagnosis [3,5]. The diagnosis
determined by MRCP can be confirmed by ERCP and/or
surgical methods. In our study, 171 of the 655 patients
that underwent MRCP underwent ERCP with the possi-
ble necessity of endoscopic treatment in mind. MRCP
defined stones in the bile ducts of 103 patients. ERCP
confirmed the presence of stones in 102 of those patients.
ES and stone extraction were performed. In one patient,
the choledoch could not be cannulated due to diverticula
in the duodenum. ES was performed on 14 patients that
ERCP showed to have bile duct stenosis due to tumors,
and biopsies were taken. In one case where a tumor in
the papilla was observed along with stones, ES + stone
extraction and biopsy were performed with ERCP. In one
case, stent placement was performed to ensure bile flow
(Table 1).
There are various studies investigating the sensitivities,
specifities and precision of MRCP and ERCP in deter-
mining the pathologies of the bile ducts. In a prospective
study by M. G. Scaffidi, et al., MRCP and ERCP results
in determining choledoch stones were compared, and
MRCP’s sensitivity was found as 88%, specifity as 72%,
and diagnostic accuracy as 83% [6]. In the study carried
out by Lopez Hanninen, et al., on 66 patients, the sensi-
tivity of MRCP in showing tumoral lesions of the bile
ducts was reported as 83%, its specifity as 96% and di-
agnostic accuracy as 90% [7]. In our study, in concor-
dance with literature, the sensitivity of MRCP in identi-
fying bile duct stones was determined as 92%, its speci-
fity as 74%, diagnostic accuracy as 83%, and in identi-
fying tumoral lesions, its sensitivity as 85%, specifity as
98% and diagnostic accuracy as 92% (Table 2).
Stenosis may develop in bile ducts due to prior surgery or

M. A. ERYILMAZ ET AL. 492
to primary sclerosing cholangitis. According to literature,
MRCP’s sensitivity in showing the localization and length
of the stenosis is 100% its specifity is 91%, and diagnos-
tic accuracy is 95% [8]. Our study results are similar. In
this field, we found MRCP’s sensitivity to be 100%,
specifity, 99% and diagnostic accuracy, 99%.
Bile duct obstruction due to hydatic cysts opening into
the bile duct can be seen in 5% - 25% of patients. ERCP
and ES are one of the treatment methods [9]. In our study,
hydatic cyst lesions opening into the bile duct were iden-
tified in 4 (2.33%) patients. ES was performed by ERCP,
and the vesicles and membranes of the cyst were re-
moved. In our study, the sensitivity of MRCP was deter-
mined as 100%, specifity as 99%, and diagnostic accu-
racy as 99%. There was no significant difference be-
tween MRCP and ERCP (P > 0.05).
Stones that are smaller than the section thickness of
the MR device may give false negative results, and in
patients where there is a gap of more than 72 hours be-
tween the MRCP and the ERCP, although MRCP has
indicated stones, ERCP may not find stones due to their
spontaneous passing [10]. In such cases, ES is a thera-
peutical procedure [11]. In our study, 32 patients by
MRCP and 26 patients by ERCP were detected to have
dilated bile ducts that could not be explained by any ob-
structive pathology such as stones or tumors, and ES was
performed.
MRCP and diagnostic ERCP were compared for cost
effectiveness through a literature scan over 28 articles by
Kaltenthaler et al., and it was decided that MRCP was
more economical [12]. In our study, it was determined
that the cost of one MRCP event cost TL 65, while one
ERCP event cost approximately TL 284.
Even in experienced hands, ERCP has a cannulation
failure rate of up to 3% - 9% [13]. In our study, it has
been observed that ERCP didn’t give satisfactory results
in 5.26% of the cases. ERCP, considered a reference
method presently, is accepted to have a morbidity rate of
0.8% - 10% [14]. Morbidity was 2.92% in our study. In
one case where a bile duct stone was detected, the mor-
bidity was due to basket compaction. Two cases had
perforations, one had a hemorrhage and one had pa-
creatitis. T. Obana et al., have observed no mortalities
connected to the ERCP procedure [15]. However, low
that it may be, ERCP has a mortality rate of 0.7% - 0.9%
[16]. There were no mortalities in our study.
Our study has shown the ERCP procedures which are
routinely carried out by gastroenterologists can be carried
out successfully by experienced General Surgeons. 10%
of the MRCPs and 5% of the ERCPs carried out in our
country in the beginning of the 2000s were optimally
successful [11]. In the 2010 plus years, MRCP can be
carried out on all cases where clinicians deem it neces-
sary without the need for sedation, and ERCP is carried
out with success rates of up to 95% (Table 1).
Our study has some shortcomings. While MRCP was
carried out on all patients with bile duct pathologies, se-
lected patients underwent ERCP. ERCP and MRCP were
not carried out on the same day. The diagnosis reached
with MRCP was confirmed with ERCP. Surgery was
performed in cases where ERCP was unsuccessful. In our
country, the cost of MRCP and ERCP are based upon the
unit price paid by the social security institution, and may
differ from other countries.
5. Conclusion
In the identification of bile duct pathologies, MRCP has
accuracy values near to that of ERCP. It is approximately
four times more economical, and takes far less time.
Therefore, MRCP should be used for diagnosis, and
ERCP should be performed for treatment, if necessary.
REFERENCES
[1] E. Elüdokuz, E. Tankurt, H. Akbaylar, E. İğci, O. Dicle, İ.
Şimşek, et al., “The Sensitivity of Ultrasonography ın
Cases Choledocholithiasis Diagnosed with ERCP,” Tur-
kiye Klinikleri Journal of Gastroenterohepatology, Vol. 7,
1996, pp. 108-110.
[2] P. D. Prabhakar, A. M. Prabhakar, H. B. Prabahakar and
D. Sahani, “Magnetic Resonance Cholangiopancreato-
graphy of Benign Disorders of the Biliary System,” Mag-
netic Resonance Imaging Clinics of North America, Vol.
18, No. 3, 2010, pp. 497-514.
doi:10.1016/j.mric.2010.08.007
[3] G. Ersöz, “Endoscopic Retrograde Cholangiopancreato-
graphy in Bile Duct Disorders,” Türkiye Klinikleri Jour-
nal of Medical Sciences, Vol. 3, 2007, pp. 12-16.
[4] H. Şentürk, B. Canbakan and E. Özden, “ERCP: Indica-
tions and Risks,” Series of Clinical Approach in Gastro-
enterology, Vol. 30, 2004, pp. 101-109.
[5] F. Maccioni, M. Martinelli, N. Al Ansari, A. Kagarmanova,
et al., “Magnetic Resonance Cholangiography: Past, Pre-
sent and Future: A Review,” European Review for Medi-
cal and Pharmacological Sciences, Vol. 14, No. 8, 2010,
pp. 721-725.
[6] E. C. Kaltenthaler, S. J. Walters, J. Chilcott, A. Blake-
borough, Y. B. Vergel and S. Thomas, “MRCP Compared
to Diagnostic ERCP for Diagnosis When Biliary Obstruc-
tion Is Suspected: A Systematic Review,” BMC Medical
Imaging, Vol. 6, 2006, pp. 9-24.
doi:10.1186/1471-2342-6-9
[7] M. G. Scaffidi, C. Luigiano, P. Consolo, R. Pellicano, G.
Giacobbe, M. Gaeta, et al., Magnetic Resonance Cholan-
giopancreatography versus Endoscopic Retrograde Cho-
langiopancreatography in the Diagnosis of Common Bile
Duct Stones: A Prospective Comparative Study,” Miner-
va Medica, Vol. 100, No. 5, 2009, pp. 341-348.
[8] E. L. Hänninen, H. Amthauer, N. Hosten, J. Ricke, M.
Böhmig, J. Langrehr, et al., “Prospective Evaluation of
Pancreatic Tumours: Accuracy of MR Imaging with MR

M. A. ERYILMAZ ET AL. 493
Cholangiopancreatography and MR Angiography,” Radi-
ology, Vol. 224, No. 1, 2002, pp. 34-41.
doi:10.1148/radiol.2241010798
[9] P. Angulo, D. H. Pearce, C. D. Johnson, J. J. Henry, N. F.
LaRusso, B. T. Petersen, et al., “Magnetic Resonance
Cholangiopancreatography in Patients with Biliary Dis-
ease: Its Role in Primary Sclerosing Cholangitis,” Journal
of Hepatology, Vol. 33, No. 4, 2000, pp. 520-527.
doi:10.1016/S0168-8278(00)80002-1
[10] G. Galati, A. V. Sterpetti, M. Caputo, M. Adduci, G. Lu-
candri, S. Brozzetti, et al., “Endoscopic Retrograde Cho-
langiography for Intrabiliary Rupture of Hydatic Cyst,”
The American Journal of Surgery, Vol. 191, No. 2, 2006,
pp. 206-210. doi:10.1016/j.amjsurg.2005.09.014
[11] M. M. Calvo, L. Bujanda, A. Calderon, J. L. Cabrıada, A.
Bernal, V. Orive, et al., “Role of Magnetic Resonance
Cholangiopancreatography in Patients with Suspected
Choledocholithiasis,” Mayo Clinic Proceedings, Vol. 77,
No. 5, 2002, pp. 422-428.
doi:10.1016/S0025-6196(11)62210-6
[12] E. Kaltenthaler, Y. B. Vergel, J. Chilcott, S. Thomas, T.
Blakeborough, S. J. Waters, et al., “A Systematic Review
and Economic Evaluation of Magnetic Resonance Cholan-
giopancreatography Compared with Diagnostic Endoscopic
Retrograde Cholangiopancreatography,” Health Technol-
ogy Assessment, Vol. 8, No. 10, 2004, pp. 1-89.
[13] R. Schöfl, “Diagnostic Endoscopic Retrograde Cholan-
giopancreatography,” Endoscopy, Vol. 33, No. 2, 2001,
pp. 147-157. doi:10.1055/s-2001-11667
[14] M. L. Freeman, D. B. Nelson, S. Sherman, G. B. Haber,
M. E. Herman, P. J. Dorsher, et al., “Complications of
Endoscopic Biliary Sphincterotomy,” The New England
Journal of Medicine, Vol. 335, No. 13, 1996, pp. 909-919.
doi:10.1056/NEJM199609263351301
[15] T. Obana, N. Fujita, Y. Noda, G. Kobayashi, K. Ito, J.
Horaguchi, et al., “Efficacy and Safety of Therapeutic
ERCP for the Elderly with Choledocholithiasis: Compa-
rison with Younger Patients,” Internal Medicine, Vol. 49,
No. 18, 2010, pp. 1935-1941.
doi:10.2169/internalmedicine.49.3660
[16] K. Hekimoğlu, Y. Üstündağ, A. Duşak, Z. Erdem, B.
Karademir, S. Aydemir, et al., “MRCP vs ERCP in the
Evaluation of Biliary Pathologies: Review of Current Lit-
erature,” Journal of Digestive Diseases, Vol. 9, No. 3,
2008, pp. 162-169.
doi:10.1111/j.1751-2980.2008.00339.x