ORIGINAL PAPERS

13-sozan.qxd
7/10/09 12:42 PM
Page 297
ORIGINAL PAPERS
Adv Clin Exp Med 2009, 18, 3, 297–302
ISSN 1230−025X
© Copyright by Wroclaw Medical University
TOMASZ SOZAŃSKI1, VIOLETTA SOKOLSKA2, KRZYSZTOF MOROŃ3
Magnetic Resonance Cholangiopancreatography
After Failed or Incomplete Endoscopic Retrograde
Cholangiopancreatography
Zastosowanie cholangiopankreatografii rezonansu magnetycznego
w przypadku nieudanej lub niekompletnej wstecznej endoskopowej
cholangiopankreatografii
1
Department of Pharmacology, Wroclaw Medical University, Poland
Department of General Radiology, Interventional Radiology and Neuroradiology, Chair of Radiology,
Wroclaw Medical University, Poland
3
Department of Radiology, Wroclaw Medical University, Poland
2
Abstract
Background. MRCP is a new noninvasive technique for examining the biliary and pancreatic ducts without the
need for contrast medium and without any known side effects. MRCP could replace diagnostic ERCP and other
invasive bile and pancreatic duct examination methods.
Objectives. Evaluation of the usefulness of MRCP to delineate bile and pancreatic ducts and depict pathological
fluid collection in the biliopancreatic region in patients after failed or incomplete ERCP.
Material and Methods. MRCP was performed in 36 patients who had previously undergone an attempted ERCP.
However, with ERCP complete duct visualization was obtained in only 17 patients, but it failed in 6 patients and
opacification was incomplete in 13. MRCP was done using a 1.5−T Picker Edge Eclipse system (EXPRESS, TR: ∞,
TEeff: 268 ms, ETL: 200, slice thickness: 50–70 mm, matrix: 336 × 512, FOV: 38 cm).
Results. Complete visualization of the ducts was obtained with MRCP in all 36 patients, whereas ERCP was suc−
cessful in only 21 cases. In 6 patients, MRCP showed pseudocysts missed by ERCP. In 15 patients, no further inva−
sive procedures were needed after MRCP.
Conclusions. MRCP is a valuable diagnostic tool for evaluating patients after failed or unsuccessful ERCP. MRCP
is a noninvasive technique that may avoid the use of other, invasive procedures (Adv Clin Exp Med 2009, 18, 3,
297–302).
Key words: MRCP, ERCP, bile ducts, pancreatic ducts.
Streszczenie
Wprowadzenie. MRCP jest nowoczesną, nieinwazyjną metodą diagnostyki dróg żółciowych i przewodu
trzustkowego, która nie wymaga podania środków kontrastowych i nie daje powikłań. MRCP może zastąpić diag−
nostyczne ERCP i inne inwazyjne metody diagnostyki dróg żółciowych i przewodu trzustkowego.
Cel pracy. Ocena uwidocznienia dróg żółciowych i przewodów trzustkowych oraz patologicznych zbiorników
płynowych w ich okolicy w badaniu MRCP u pacjentów po nieudanym lub niekompletnym ERCP.
Materiał i metody. Badanie MRCP wykonano u 36 pacjentów, u których wcześniej podjęto próbę wykonania
ERCP, przy czym jedynie u 17 pacjentów (47,2%) w ERCP uzyskano pełny obraz badanych przewodów, u pozos−
tałych 19 pacjentów bądź nie udało się wykonać ERCP (6 pacjentów), bądź jego obraz był niekompletny
(13 chorych). Badania wykonano z użyciem aparatu Picker Edge Eclipse (sekwencje EXPRESS, TR – ∞, TEeff –
268 ms, ETL – 200, grubość warstw – 50–70 mm, matryca – 336 × 512, pole widzenia – 38 cm).
Wyniki. U wszystkich 36 pacjentów (100%) w MRCP uzyskano kompletny obraz badanych przewodów, podczas
gdy w ERCP udało się to jedynie w 21 przypadkach. U 6 chorych w MRCP uwidoczniono pseudotorbiele, które
nie były widoczne w ERCP. U 15 pacjentów na podstawie wyniku MRCP podjęto decyzję o niewykonywaniu dal−
szych inwazyjnych zabiegów.
Wnioski. MRCP jest skuteczną metodą diagnostyczną u pacjentów po nieudanym lub niekompletnym ERCP.
13-sozan.qxd
7/10/09 12:42 PM
Page 298
298
T. SOZAŃSKI, V. SOKOLSKA, K. MOROŃ
MRCP jako metoda nieinwazyjna może zapobiegać wykonywaniu innych, inwazyjnych zabiegów (Adv Clin Exp
Med 2009, 18, 3, 297–302).
Słowa kluczowe: MRCP, ERCP, drogi żółciowe, przewody trzustkowe.
There are numerous traditional imaging meth−
ods used to examine bile and pancreatic ducts,
including oral cholecystography, intravenous
cholangiography, endoscopic retrograde cholan−
giopancreatography (ERCP), percutaneous tran−
shepatic cholangiography (PTC), and inta− and
post−surgical cholangiography. Each, however, has
some limitations. Oral and intravenous cholecys−
tography frequently yields images of insufficient
quality, and ERCP and PTC are invasive proce−
dures which may lead to complications. Intra− and
post−surgical cholangiography are available only
for patients undergoing surgery [1–3].
ERCP is the most commonly used traditional
method for the diagnostics and therapy of the bile
and pancreatic ducts. It was first performed by
McCune et al. in 1968 [4]. It consists of introduc−
ing a fiberscope into the duodenum, finding the
orifice of the common bile and pancreatic ducts,
catheterization, and introducing contrast medium.
Classic X−ray images are then taken using an
appropriate time sequence. ERCP provides
detailed visualization of the bile and pancreatic
ducts, providing images of very high spatial reso−
lution. As a therapeutic method, it allows Vater’s
papilla incision and the removal of concrements
from the common bile duct [5]. In bile duct steno−
sis it allows their dilation using inflatable balloons
or the introduction of stents. Moreover, it allows
direct the visualization and taking of a biopsy of
the Vater’s papilla area [6]. The method is, how−
ever, burdened with numerous possible complica−
tions. The procedure requires sedation, endoscopy,
introduction of a catheter into the bile ducts, and
the direct application of contrast medium.
Pancreatitis or cholangitis as well as allergic reac−
tions to the applied iodine−based contrast media
are encountered in approx. 5–8% of ERCP cases.
In therapeutic procedures, additional risks include
bile duct perforation, bleeding, and infection.
Diagnostic ERCP is associated with a 0.96% mor−
tality rate, which increases to 2.3% if the proce−
dure is combined with sphincterotomy. Moreover,
technical problems during the procedure mean that
complete visualization of the bile and pancreatic
ducts is achieved in only 80–90% of cases [7, 8].
As a therapeutic method. ERCP is and will remain
the method of choice in the coming years.
However, because of its invasiveness and risk of
associated complications, as a diagnostic method
it should be replaced by other, noninvasive meth−
ods in many patients [8–10].
The development of diagnostic techniques
based on magnetic resonance allowed the intro−
duction of a new imaging method, magnetic reso−
nance cholangiopancreatography (MRCP). The
method allows obtaining images of quality compa−
rable to those obtained by invasive methods.
Compared with the previously used methods it is
noninvasive. It does not require the administration
of contrast media or patient sedation [1–3]. MRCP
is based on fast spin−echo sequences (FSE) with a
very long effective echo time (pseudoecho) of
140–240 milliseconds. This results in obtaining
strongly T2 time−dependent images in which only
structures with long transverse relaxation time are
visualized. That ensures strong contrast between
strong signals from stationary or slowly moving
fluids in the bile and pancreatic ducts and weak
signals from soft tissues, concrements, and blood
flowing in vessels [1–3, 11, 12].
The first MRCP applications described by
Wallner et al. in 1991 used sequences of gradient
echoes (steady−state free precession, SSFP) [7].
They allowed the detection of dilated bile and pan−
creatic ducts. However, because of their low spatial
resolution, they were insufficient for the visualiza−
tion of non−dilated ducts. Moreover, they were
highly susceptible to motion artifacts (intestinal
peristalsis, abdominal integument movements,
aorta pulsation) and magnetic field inhomogeneity
(loss of signal caused by the presence of metal clips
or duodenal air). Moreover, it was necessary for the
patient to hold a breath for as long as one minute
during the examination. Because of these consider−
ation, they were replaced by FSE sequences, which
are a variation of traditional spin−echo sequences
[13, 14]. FSE sequences were first used for MRCP
imaging in 1994 by Takehara et al. [15]. Compared
with gradient sequences, they allow visualization of
both dilated and regular bile and pancreatic ducts
[16]. The next advance was the application of sin−
gle−shot FSE by Laubenberger in 1995. The use of
single−shot shortens the sequence significantly, to 4
seconds in the case of a thick−layer technique, and
significantly improves spatial resolution [17]. The
next step was the introduction of a semi−Fourier
acquisition technique in place of the previously
used Fourier transformation [18]. This allowed lim−
iting the time necessary to obtain a single scan to
2 seconds in the case of the thick−layer technique
and to 18 seconds for thin layers. The semi−Fourier
image transformation technique is currently recog−
nized as the best available method [19, 20].
13-sozan.qxd
7/10/09 12:42 PM
Page 299
299
MRCP After Failed or Incomplete ERCP
In the multisection technique it is possible to
obtain thick source images and reconstructing
them, most commonly using maximum intensity
projection (MIP). This allows the reconstruction
of three−dimensional (3D) images. With this tech−
nique, high−definition images are obtained, but
small pathological lesions may be invisible in the
resulting reconstructions, which is a result of sig−
nal averaging of adjacent layers. Therefore, in the
multisection technique there is always the necessi−
ty to analyze the initial scans. The thick−layer pro−
jection technique is an alternative. It shortens the
scanning time significantly [1, 20, 21]. Respi−
ratory movement−related artifact reduction is
obtained by the repetition of single acquisition,
currently not commonly used. Respiratory gating
is most commonly used, consisting of gathering
information on the same respiratory cycle phase
with the examination performed while the patient
holds a breath. Some authors suggest administra−
tion of hyoscine butylbromide (Buscopan) or
glucagon directly before the examination to reduce
intestinal peristalsis.
Generally, patients do not require any special
preparation for MRCP. The only recommendation
is restraining from food and drink for approxi−
mately six hours before the examination. This
allows emptying the stomach and duodenum,
improving the quality of the resulting images and,
in case of a need for classic MRI scans, it allows
safe administration of contrast medium. No
MRCP−associated side effects have been noted.
Besides the general contraindications for MRI,
there are no specific contraindications for MRCP.
The examination is absolutely forbidden in
patients with ferromagnetic vascular clips and
pacemakers. Moreover, it may prove difficult to
complete in patients with ferromagnetic implants
causing artifacts in the porta hepatis area and in
individuals suffering from claustrophobia.
Claustrophobic patients may be examined under
general anesthesia or using modern open MRI sys−
tems [1, 2, 18].
Material and Methods
One hundred sixteen MRCP examinations
were performed at the Chair and Department of
Radiology, Wroclaw Medical University. Thirty−
six patients were selected from that group in
whom MRCP was preceded by an attempt to per−
form an ERCP examination. That group of patients
was analyzed. The study group of 36 patients
included 21 males (58.3%) and 15 females (41.7%).
The patients’ ages ranged between 19 and 71 and
the average was 48 years.
The examinations were conducted using
Picker Edge Eclipse 1.5−T equipment. The patients
fasted for six hours before the examination. The
procedure started with three pilot scans in FSE
sequence. The first scan, the so−called localizing
one, was performed in the sagittal plane and the
following ones in the axial and frontal planes. The
proper epigastrial region scans in FSE sequence
were then performed in the axial projection, cov−
ering the region from the hepatic vault to the infe−
rior limit of the pancreas. Two projections in T1
and T2 sequences were taken. The MRCP images
were obtained using a single−shot method, with the
semi−Fourier transformation of the resulting image
using the parameters presented in Table 1.
Table 1. MRCP scan parameters
Tabela 1. Parametry techniczne skanów MRCP
Sequence
Express
Repetition time – TR (ms)
∞
Effective time echo – TEeff (ms)
268
Echo train lenght – ETL
200
Slice thickness (mm)
50–70
Matrix
336–512
Field of view – FOW (cm)
38
Results
In the group of 36 patients in whom ERCP was
attempted, the examination was completely unsuc−
cessful in 6 cases (16.6%). In all cases this was due
to failure to find the orifice of Vater’s papilla or
difficulties encountered with its catheterization. In
7 cases (19.4%), incomplete visualization of the
bile and pancreatic ducts were obtained. In 4 pa−
tients (11.1%) ERCP yielded a complete depiction
of the bile and pancreatic ducts, consistent with the
cholangiopancreatogram obtained in MRCP. The
ERCP examination, however, failed to visualize
pathological fluid collection within or around the
pancreas, which was found on MRCP images. In
2 patients (5.5%) ECRP allowed an incomplete
depiction of the bile and pancreatic ducts and
failed to reveal pancreatic pseudocysts visible on
MRCP images (Fig. 1a).
Overall, incomplete cholangiopancreatograms
from ERCP (incomplete visualization of the exam−
ined ducts or failure to show a pancreatic pseudo−
cyst) were obtained for 13 patients (36.1%). In
17 patients (47.2%) ERCP yielded complete visu−
alization of the bile and pancreatic ducts, fully
compatible with the result obtained from MRCP.
MRCP gave a complete image of the bile and
13-sozan.qxd
7/10/09 12:42 PM
Page 300
300
T. SOZAŃSKI, V. SOKOLSKA, K. MOROŃ
b
a
Fig. 1. A 31−year−old patient hospitalized with exacerbated chronic pancreatitis. He had been operated twice because of
post−inflammatory pancreatic pseudocysts. In ERCP, the pancreatic duct was filled selectively only in the head and the
proximal part of the body of the pancreas (a). It was not possible to see bile ducts in ERCP. In MRCP, a non−dilated
pancreatic duct in the head and the proximal part of the body of the pancreas was depicted. A tortuous main pancreatic
duct with cyst−like dilated branches was seen in the tail and the distal part of the body of the pancreas. In the tail of the
pancreas a pseudocyst was revealed. Non−dilated intra− and extrahepatic bile ducts were also depicted (b)
Ryc. 1. 31−letni pacjent przyjęty z powodu zaostrzenia przewlekłego zapalenia trzustki, w przeszłości 2−krotnie
operowany z powodu pozapalnych pseudotorbieli trzustki. W badaniu ERCP selektywnie zakontrastowano przewód
Wirsunga, uwidaczniając go jedynie w obrębie głowy i początkowej części trzonu trzustki (a). W ERCP nie udało się
zakontrastować dróg żółciowych. W wykonanym badaniu MRCP uwidoczniono przewód Wirsunga prawidłowej
średnicy i przebiegu w obrębie głowy i początkowej części trzonu trzustki. W dalszej części trzonu i ogona trzustki
przewód Wirsunga wykazywał kręty przebieg z obecnością licznych torbielowato poszerzonych bocznych
odgałęzień. W trzonie trzustki stwierdzono pseudotorbiel. Uwidoczniono także nieposzerzone drogi żółciowe
wewnątrz− i zewnątrzwątrobowe (b)
pancreatic ducts in all 36 cases (Fig. 1b). In 15 pa−
tients in whom MRCP was performed, the deci−
sion was made not to perform any other invasive
procedures, saving those patients from the risk of
developing iatrogenic complications.
Discussion
The comparison of ERCP and MRCP has been
the subject of numerous studies and discussions
among physicians of various specialties. ERCP
allows obtaining very−high−quality images of the
bile and pancreatic ducts. It is, however, an inva−
sive method burdened with the risk of complica−
tions [6–10]. MRCP yields images of quality com−
parable to that of ERCP but is also an examination
safe for the patient [22, 23]. MRCP does not
require finding and catheterizing Vater’s papilla or
contrast medium administration into the examined
ducts. In the event of difficulty related to bile and
pancreatic duct orifice catheterization, or if both
ducts have separate orifices on the major and
minor Vater’s papilla, when only the bile duct or
the pancreatic duct may be catheterized, MRCP is
a method allowing complete visualization of the
ducts. Moreover, the visualization of individual
sections of the ducts using MRCP does not depend
on the possibility of contrast medium penetration,
as is the case with ERCP. MRCP reveals natural
fluids, such as bile and pancreatic juice. Therefore,
if there is an obstruction preventing contrast medi−
um from penetrating during ERCP, MRCP allows
visualization of the ducts both distally and proxi−
mally to the place of obstruction. As a result, if
ERCP yields incomplete visualization, showing
only part of the ducts, MRCP allows visualization
of the ducts in their entirety [1, 24].
In a study by Soto et al., MRCP allowed com−
plete visualization of the bile and pancreatic ducts
in all 37 patient after failed or incomplete ERCP.
No abnormalities in the examined ducts were
found in 11 patients, saving them from invasive
procedures (PTC, laparotomy) and reducing the
risk of iatrogenic complications [25]. Similar
results were obtained by Fulcher et al. from ana−
lyzing 27 patients after failed or incomplete ERCP.
In all cases, MRCP yielded valuable diagnostic
13-sozan.qxd
7/10/09 12:42 PM
Page 301
MRCP After Failed or Incomplete ERCP
images that could be the basis for qualifying the
patients for conservative therapy or invasive pro−
cedures [26]. Reinhold et al., analyzing a large
group of 159 patients, found 10 cases of failed
ERCP, corresponding to 6% of patients, and 3
cases of failed MRCP, corresponding to 2% of
patients [27].
The results of the present study demonstrated
that MRCP is a valuable imaging method for eval−
uating patients after failed or inadequate ERCP.
Simultaneously, MRCP, as a noninvasive tech−
nique, may prevent other, invasive procedures,
decreasing the potential number of iatrogenic
complications, shorting hospitalization time, and
thus reducing therapy−associated costs. In a study
301
by Varghese et al., only 12 of 59 patients who
underwent MRCP after failed or inadequate ERCP
were subsequently treated surgically [28].
The present study found 13 cases of incom−
plete bile and pancreatic duct visualization with
ERCP, corresponding to 36.1% of patients with
attempted ERCP. ERCP failed in 6 patients
(16.6%). This high ratio of incomplete and failed
ERCP examination is the result of a diagnostic
scheme adopted by numerous domestic clinics and
hospitals stating that ERCP is the preferred diag−
nostic procedure. MRCP is still often a procedure
reserved for patients in whom other diagnostic
procedures, including ERCP, were insufficient for
making a diagnosis.
References
[1] Fulcher AS, Turner MA, Capps GW: MR Cholangiography: Technical Advances and Clinical Applications.
RadioGraphics 1999, 19, 25–41.
[2] Fulcher AS, Turner MA: MR Pancreatography: A Useful Tool for Evaluating Pancreating Disorders.
RadioGraphics 1999, 19, 5–24.
[3] Macdonald GA, Peduto AJ: Magnetic resonance imaging and diseases of the liver and biliary tract. Part 2.
Magnetic resonance cholangiography and angiography and conclusions. J Gastroenterol Hepatol 2000, 15,
992–999.
[4] McCune WS, Sharp PE, Muscowitz M: Endoscopic cannulation of the ampulla of Vater: a preliminary report.
Ann Surg 1968, 167, 752–755.
[5] Mercer S, Singh S, Paterson I: Selective MRCP in the management of suspected common bile duct stones. HPB
(Oxford) 2007, 9 (2), 125–130.
[6] Binmoeller KF, Schafer TW: Endoscopic Management of Bile Duct Stones. J Clin Gastroenterol 2001, 32 (2),
106–118.
[7] Cohen SA, Siegel JH, Kasmin FE: Complications of diagnostic and therapeutic ERCP. Abdom Imaging 1996,
21, 385–394.
[8] Freeman ML, Nelson DB, Sherman S: Complications of endoscopic biliary sphincterotomy. N Engl J Med 1996,
335, 909–918.
[9] Sheridan MB: Endoscopic retrograde cholangiopancreatography should no longer be used as a diagnostic test:
the case in favour. Digest Liver Dis 2002, 34, 370–374.
[10] Baillie J: Training in advanced pancreaticobiliary endoscopy: why, how, and will we even need ERCP in the
future? Tech Gastrointest Endosc 2004, 6, 100–106.
[11] Barish MA, Soto JA: MR cholangiopancreatography: techniques and clinical applications. AJR 1997, 169,
1295–1303.
[12] Bret PM, Reinhold C: Magnetic resonance cholangiopancreatography. Endoscopy 1997, 29, 472–486.
[13] Mitchell DG: MRI Principles. W.B. Saunders, Philadelphia 1999, 191–203.
[14] Masui T, Katayama M, Kobayashi S et al.: Magnetic resonance cholangiopancreatography: comparison of res−
piratory−triggered three−dimensional fast−recovery fast spin−echo with parallel imaging technique and breath−hold
half−Fourier two−dimensional single−shot fast spin−echo technique. Radiat Med 2006 Apr, 24 (3), 202–209.
[15] Tang Y, Yamashita Y, Arakawa A: Pancreaticobiliary Ductal System: Value of Half− Fourier Rapid Acquisition
with Relaxation Enhancement MR Cholangiopancreatography for postoperative Evaluation. Radiology 2000, 215,
81–88.
[16] Reinhold C, Guibaud L, Genin G, Bret PM: MR cholangiopancreatography: comparison between two−dimen−
sional fast spin−echo and three−dimensional gradient−echo pulse sequences. J Magn Reson Imaging 1995, 4, 379–384.
[17] Laubenberger J, Buchert M, Schneider B, Blum U, Hennig J, Langer M: Breath−hold projection magnetic res−
onance cholangiopancreatography (MRCP): a new method for the examination of the bile and pancreatic ducts.
Magn Reson Med 1995, 33, 18–23
[18] Sai J, Ariyama J: MRCP Early Diagnosis of Pancreatobiliary Diseases. Springer−Verlag 2000, 11–62.
[19] Miyazaki T, Yamashita Y, Tsuchigame H, Urata J, Takahashi M: MR cholangiography using HASTE (half−
Fourier acquisition single−shot turbo spin−echo) sequences. AJR 1996, 166, 1297–1303.
[20] Sananes J, Lecesne R, Raymond J, Couzigou P, Laurent FH, Drouillard J: MR cholangiography with HASTE
sequence: a new technique for noninvasive exploration of the biliary tract. Radiology 1995, 197 (P), 343.
[21] HosseinzadehK, Furlan A, Almusa O: 2D thick−slab MR cholangiopancreatography: does parallel imaging with
sensitivity encoding improve image quality and duct visualization? AJR 2008 Jun, 190(6), W327–W334.
13-sozan.qxd
7/10/09 12:42 PM
Page 302
302
T. SOZAŃSKI, V. SOKOLSKA, K. MOROŃ
[22] Pamos S, Benages A, Medina E, Martinez Sanjuan V: Prospective evaluation of magnetic resonance cholan−
giopancreatography in patients with biliary disease: comparative study with conventional ultrasonography and
endoscopic retrograde cholangiopancreatography. Dig Liver Dis 2003, 35, 186–192.
[23] Albert JG, Riemann JF: ERCP and MRCP – when and why. Best Pract Res Clin Gastroenterol 2002, 16 (3),
399–419.
[24] Sahni VA, Mortele KJ: Magnetic resonance cholangiopancreatography: current use and future applications. Clin
Gastroenterol Hepatol 2008 Sep, 6 (9), 967–977.
[25] Soto JA, Yucel EK, Barish MA, Chuttani R, Ferrucci JT: MR cholangiopancreatography after unsuccessful or
incomplete ERCP. Radiology 1996, 199, 91–98.
[26] Fulcher AS, Turner MA, Capps GW, Zfass AM, Baker KM: Half−Fourier RARE MRCP in 300 subjects.
Radiology 1998, 207, 21–32.
[27] Reinhold C, Taourel P, Bret PM, Cortas GA, Mehta SN, Barkun AN, Wang L, Tafazoli F:
Choledocholithiasis: Evaluation of MR Cholangiography for Diagnosis. Radiology 1998, 209, 435–442.
[28] Varghese JC, Farrell MA, Courtney G, Osborne H, Murray FE, Lee MJ: Role of MR cholangiopancreatog−
raphy in patients with failed or inadequate ERCP. AJR 1999, 173, 1527–1533.
Address for correspondence:
Tomasz Sozański
Department of Pharmacology
Wroclaw Medical University
Mikulicza−Radeckiego 2
50−345 Wroclaw
Poland
Tel.: +48 697 083 542, +48 71 784 14 40
E−mail: [email protected]
Conflict of interest: None declared
Received: 2.03.2009
Revised: 25.05.2009
Accepted: 19.06.2009