J Gastroenterol

DOI 10.1007/s00535-017-1359-5

REVIEW

Clinical practice guideline for post-ERCP pancreatitis

Tetsuya Mine1 • Toshio Morizane2 • Yoshiaki Kawaguchi1 • Ryukichi Akashi3 •

Keiji Hanada4 • Tetsuhide Ito5 • Atsushi Kanno6 • Mitsuhiro Kida7 •

Hiroyuki Miyagawa8 • Taketo Yamaguchi9 • Toshihiko Mayumi10 •
Yoshifumi Takeyama11 • Tooru Shimosegawa6

Received: 7 February 2017 / Accepted: 4 June 2017

Ó Japanese Society of Gastroenterology 2017

Abstract implementations, the same timing of tests, and missing data

Background Endoscopic retrograde cholangiopancreatogra- were assessed as well as the indirectness of the study subjects,
phy (ERPC) is used for the diagnosis and treatment of pan- index tests, reference standard, and outcomes. Grading of rec-
creatic and biliary diseases. Post-ERCP pancreatitis (PEP) is a ommendations was determined as strong or weak. In clinical
complication which needs special care and clinical practice practice, the judgment of attending doctors should be more
guideline for this morbidity is also needed. important than recommendations described in clinical practice
Methods The key clinical issues of diagnosis and treatment of guidelines. Gastroenterologists are the target users of this
PEP were listed and checked, and then the clinical questions clinical practice guideline. General practitioners or general
were formulated. PubMed (MEDLINE) and Ichushi-web (Ja- citizens are not supposed to use this guideline. The guideline
panese medical literature) were used as databases. For the study committee has decided to include wide clinical issues such as
of diagnostic test accuracy, items similar to QUADAS-2, i.e., etiological information, techniques of ERCP, the diagnosis,
random selection from a population to which the diagnostic test treatments, and monitoring of PEP in this guideline.
is applied, blinding of index tests and reference tests, com- Results In this concise report, we described ten clinical
pleteness of reference standard, completeness of test questions, recommendations, and explanations pertaining to
risk factors, diagnosis, prognostic factors, treatments, and
preventive interventions in the medical practice for PEP.
& Tetsuya Mine
tetsu-m@is.icc.u-tokai.ac.jp Conclusions We reported here the essence of the clinical
practice guideline for PEP.
1
Tokai University, School of Medicine, Isehara, Japan
2
Japan Council for Quality Health Care, Tokyo, Japan Keywords Endoscopic retrograde
3
Kumamoto City Medical Association Health Care Center, cholangiopancreatography (ERPC)
Post-ERCP
Kumamoto, Japan pancreatitis
Clinical guideline
4
Onomichi General Hospital, Onomichi, Japan
5
Graduate School of Medical Sciences, Kyushu University,
Fukuoka, Japan
Introduction
6
Department of Gastroenterology, Tohoku University, Sendai,
Japan
Endoscopic retrograde cholangiopancreatography (ERPC)
7
is used for diagnosis and treatment of pancreatic diseases
Kitasato University School of Medicine, Sagamihara, Japan
and biliary tract diseases. ERCP needs sophisticated skills.
8
Sapporo-Kosei General Hospital, Sapporo, Japan Post-ERCP pancreatitis (PEP) is a complication which
9
Chiba Cancer Center, Chiba, Japan needs special care and a clinical practice guideline for this
10
University of Occupational and Environmental Health, morbidity has been needed. The Japan Pancreas Society
Kitakyushu, Japan asked a committee to undertake the process of developing
11
Faculty of Medicine, Kinki University Hospital, this clinical practice guideline. We report here the essence
Osakasayama, Japan of the clinical practice guideline.

123
 J Gastroenterol

The population to whom this guideline is meant to apply [1–4]. Qualitative systematic reviews were done evaluating
is adults who receive ERCP because of various diseases or the risk of biases and indirectness of each study. The
morbidities. Gastroenterologists are the target users of this domains of risk of bias were selection bias, performance
clinical practice guideline. General practitioners or general bias, detection bias, attrition bias, and other biases. As for
citizens are not supposed to use this guideline. We expect randomized controlled trials (RCTs), random allocation,
that gastroenterologists should be certified by the Japanese concealment, blinding of healthcare provider and patients,
Society of Gastroenterology or the Japan Gastroentero- blinding of outcome measurers, incomplete outcome data,
logical Endoscopy Society will use this guideline. The intention-to-treat analysis, early stopping of trial, selective
guideline committee decided to include wide clinical issues outcome reporting, and other items were assessed. As for
such as etiological information, techniques of ERCP, the observational studies, the same four domains were assessed
diagnosis, treatments and monitoring of PEP in this but representativeness of study subjects, difference in
guideline. In this concise report, we described all these background factors of groups compared, difference in care,
clinical questions and recommendations, but mostly we appropriateness of outcome measurement, completeness of
omitted statements or explanations for them. follow-up, inadequate confounder adjustments, dose–re-
sponse relationship between risk factors or interventions
and outcomes, assumed confounders which attenuate
Materials and methods effectiveness, and magnitudes of effects were assessed as
items. As for the study of diagnostic test accuracy, items
Members similar to QUADAS-2 [5], i.e., random selection from a
population to which the diagnostic test is applied, blinding
The chairman asked chief doctors of sections at which the of index tests and reference tests, completeness of refer-
yearly frequency of ERCP implementation was high in ence standards, completeness of test implementations, the
hospitals all over Japan to join as a member. The guideline same timing of tests, and missing data were assessed.
development committee consisted of experts in this field Indirectness of study subjects, index tests, reference stan-
and an expert of guideline development methodology. dard, and outcomes were also assessed.
Conflicts of interest were disclosed according to the
guideline of the Japanese Society of Gastroenterology. Formulating recommendations
Costs of literature gathering, meetings, and other logistic
activities were covered by a research fund provided by the The body of evidence was evaluated based on overall risk
Ministry of Health, Labor, and Welfare Japan. During the of biases, overall indirectness, inconsistencies across
period of developing this guideline, no committee members studies, reporting biases (publication biases), and when
were asked or solicited about the development activities by possible imprecision of effect measures was obtained from
other stakeholders. a quantitative systematic review (meta-analysis). Strength
of the body of evidence was categorized into four levels
Topic nomination and evidence search [3]: A, strong; B, moderate; C, weak; and D, very weak.
Strength reflects confidence in effectiveness and shows the
Key clinical issues of diagnosis and treatment and other strength to support a related recommendation. Grading of
issues about PEP were listed and approved by the all recommendations was determined as strong or weak (1 or
members of the committee. Each member undertook a few 2) [3], based on evaluating benefits, harms, and burdens
key clinical issues and formulated clinical questions. These which a patient would receive. Recommendation grades
clinical questions were approved by all members of the were determined by unanimous agreement of the commit-
committee. PubMed (MEDLINE) and Ichushi-web (a tee members.
database of Japanese medical literature) were used as
databases. Medical literature since 1985 was searched for Statements for legal matters
each clinical question. The literature search was performed
by members of the Japan Medical Library Association who It is not anticipated that the recommendation is applied
suggested search queries and presented with search results without considering conditions of individual patients,
in collaboration with the committee members. because conditions such as genetic background, comor-
bidities, severity of the disease, and disease stage vary
Evidence assessment among individuals. It is impossible for the guideline
development committee to take into account individual
Systematic reviews were done for each clinical question conditions of each patient when they formulate recom-
based on a literature set collected for each clinical question mendations. Thus, this clinical practice guideline should

123
J Gastroenterol

not be used as legal basis to assess the appropriateness of well as for treatment, such as biliary drainage (2-B;
individual medical practice. Table 2).

CQ1-03. Should ERCP be performed

Results for the diagnosis and treatment of chronic
pancreatitis?
CQ1-01. Should ERCP be performed
for the diagnosis of pancreatic tumors? Recommendation ERCP is useful for the diagnosis of
chronic pancreatitis and the treatment of pancreatic stones
Recommendation The diagnostic ability of ERCP is not (2-C).
sufficient for evaluating solid pancreatic tumors. Therefore,
the indication for ERCP should be carefully considered in CQ1-04. Should ERCP be performed
cases involving these tumors. However, ERCP is necessary for the treatment of cholelithiasis?
for the diagnosis of cases with intraductal papillary muci-
nous neoplasms or autoimmune pancreatitis (2-C). Recommendation For the treatment of bile duct stones,
Endoscopic biliary drainage should be selected for cases with ERCP should be performed when the presence of bile duct
obstructive jaundice due to pancreatic tumors (2-B; Table 1). stones is strongly suspected based on results from other
diagnostic methods (2-B).
CQ1-02. Should ERCP be performed ERCP is not recommended for gallbladder drainage in
for the diagnosis? cases of acute cholecystitis with gallbladder stones (2-C).

Recommendation ERCP should be performed for the CQ2-1-1 If there is a pancreatic juice outflow
diagnosis of biliary tract cancer, such as detecting the obstruction, is PEP likely to develop?
spread of carcinoma during a histological diagnosis and for
direct cholangiography, as well as for treatment, such as Recommendation It is considered that PEP is likely to
biliary drainage (1-B). develop if there is a pancreatic juice outflow obstruction.
ERCP for gallbladder cancer should be performed for Therefore, careful monitoring is necessary (2-B).
diagnosis of the histology and progression of the disease, as
CQ2-01-2a Are the time for ERCP, volume
Table 1 Diagnostic performance of ERCP for pancreatic cancer and pressure of contrast medium injected,
Author Number of cases Sensitivity (%) and number of sessions of cannulation related
to PEP?
Gilinsky et al. 117 80
Bakkevold et al. 2082 79 Recommendation The longer duration of obstruction of the
Niederau and Grendell 565 92 pancreatic duct, contrast medium volume and pressure, and
Burtin et al. 68 92 frequent cannulation are believed to cause pancreatic duct
Rosch et al. 184 89 pressure hypertension; these are shown to be related to
onset of PEP. Therefore, these risk factors should be as
minimized as much as possible (2-B).
Table 2 Diagnostic abilities of ERCP-guided cytology and biopsy
for biliary tract cancer
CQ2-01-2b Are pancreatic duct brushing and/
Author Number of cases Sensitivity (%) or intraductal (IDUS) involved in onset of PEP?
Mansfield 16 75
Govil 22 68 Recommendation Pancreatic duct brushing and/or IDUS
Glasbrenner 27 66.7 are thought to be involved in the onset of PEP. Care should
Sugiyama 17 Cytology: 36 be taken when ERCP is performed (2-C).
Biopsy: 71
Vandervoort 10 60
CQ2-01-2c Are dysfunction of the minor papilla
Macken 30 65
and/or Santorini duct involved in the onset of PEP?
Farrell 4 50
Recommendation Dysfunction of the minor papilla and/or
Kitajima 51 Cytology: 71.6
Santorini duct is thought to be involved in the onset of
Biopsy: 65.2
PEP. Thus, cases such as the contrast agent stasis in the

123
 J Gastroenterol

main pancreatic duct at ERCP require attention for onset of CQ5-1. Is specific explanation about the incidence
PEP (2-C). and mortality of severe acute pancreatitis essential
in the informed consent for ERCP?
CQ3-01 Is it possible to get risk factors by hearing
an anamnesis? Recommendation The possibility of death caused by dete-
riorating pancreatitis after ERCP should be explained in
Recommendation There are risk factors of PEP, and it’s advance (1-C; Table 3).
recommended to hear a case history of previous diseases
(1-B). CQ5-2. Should it be explained that for diagnostic
purposes ERCP can be substituted by magnetic
CQ3-02 Is there a possibility of PEP when strong resonance cholangiopancreatography (MRCP)?
abdominal pain appears after ERCP?
Recommendation For diagnostic purposes, it is essential to
Recommendation Strong abdominal pain indicates a high explain that ERCP can be substituted by MRCP, except in
probability of PEP and needs to confirm the diagnosis (1- some diseases, such as autoimmune pancreatitis (AIP;
A). 1-B). Pathological examination of bile or pancreatic juice
cytology through ERCP cannot be substituted by MRCP
CQ3-03 Is abdominal palpation useful in diagnosing (1-C).
PEP after ERCP?
CQ6-01. How many hours after examination is
Recommendation Abdominal palpation is useful in diag- the optimal time for early diagnosis of pancreatitis
nosing PEP after ERCP, and it’s recommended (1-C). following ERCP?

CQ-4-01 What are the patient-related risk factors Recommendation Measurement of serum pancreatic
for PEP? Are young age, female sex, history enzymes (P) and measurement of serum pancreatic
of sphincter of Oddi dysfunction (SOD), history enzymes (pP), primarily a serum amylase, 2–6 h after
of PEP, and recurrent pancreatitis predisposing ERCP is recommended (1-B).
factors for PEP?
CQ6-02. Which enzyme is the best and easiest to use
Recommendation Sufficient attention should be paid to for diagnosis of acute pancreatitis after ERCP
suspected SOD, female sex, and history of pancreatitis as in clinical practice?
patient-related risk factors for PEP (2-B). Attention should
be paid to young age, absence of extrahepatic bile duct Recommendation In the diagnosis of pancreatitis after
dilatation, and normal serum bilirubin as patient-related ERCP, measurement of serum amylase is recommended
risk factors for PEP (2-B). when measurement of serum lipase is difficult (1-A). For
the diagnosis of acute pancreatitis, measurement of serum
CQ-4-02 What are the procedure-related risk lipase is useful and preferable if possible (1-A). Urinary
factors for PEP? Are pancreatic sphincterotomy, trypsinogen 2 (dipdisk test) is promising because of its
endoscopic papillary balloon dilation (EPBD), excellent ability to rapidly diagnose acute pancreatitis, but
difficulty in cannulation, and precutting it is not commercially available in Japan, limiting its use
predisposing factors for PEP? for research purposes at this time (1-B).

Recommendation Attention should be paid to precutting CQ6-03. Is measurement of procalcitonin (PCT)

and one or more contrast injections to the pancreatic duct useful to determine the severity of PEP?
as procedure-related risk factors for PEP (2-B). Attention
should be paid to five or more cannulation attempts, pan- Recommendation Procalcitonin (PCT) is considered to be
creatic sphincterotomy, papillary balloon dilation, and highly useful in determining the severity of PEP; however,
residual bile duct stones as procedure-related risk factors at present, it is expensive and requires time to measure.
for PEP (2-B). Attention should be paid to PEP when Therefore, the level of recommendation for PCT as a
performing papillectomy (2-B). general examination is low (1-B).

123
J Gastroenterol

Table 3 Epidemiological Year 2007 2008 2009 2010 2011

survey of pancreatitis following
ERCP Total 11403 13869 14427 16848 18723
Pancreatitis incidences (%) 100 (0.877) 116 (0.836) 170 (1.178) 165 (0.979) 168 (0.897)
Severe pancreatitis incidences (%) 12 (0.105) 13 (0.094) 17 (0.118) 20 (0.119) 27 (0.144)

CQ7-1 Is chest and abdominal X-ray recommended CQ08-02 Does the early assessment of the severity
for the diagnosis of PEP? of PEP contribute to the improvement of mortality
rate and reduction in the incidence
Recommendation When PEP is suspected, chest and of complications?
abdominal X-ray may not be useful for the diagnosis of
PEP (2-C). Recommendation Once the condition is diagnosed as PEP,
we recommend that its severity be immediately assessed
CQ7-2 Are ultrasonography (US), computed and that the assessment be repeated thereafter (1-C).
tomography (CT), and magnetic resonance imaging
(MRI) useful for the diagnosis of PEP? Recommendations as below about treatment of PEP
are in accordance with guidelines for pancreatitis
Recommendation When PEP is suspected, US is recom-
mended (1-B). When US demonstrates poor images, CT is CQ9-01. Does the treatment of PEP with antibacterial
recommended (1-B). MRI may be useful in diagnosing bile agents shorten the treatment period as compared with the
duct stones causing pancreatitis and hemorrhagic necro- nonuse of these drugs?
tizing pancreatitis (2-C). Recommendation In patients with severe disease, the use
of antibacterial agents might reduce the risk of infectious
CQ8-01: Are the severity assessment criteria pancreatic complications such as pancreatic abscess,
by the Ministry of Health, Labor, and Welfare shorten the treatment period, and decrease mortality (2-B).
(MHLW) appropriate for severity assessment Prophylactic treatment with antibacterial agents is basically
of PEP? not necessary in patients with mild disease, but is required
in patients with biliary tract infection (1-A).
Recommendation The severity assessment criteria by the Statement As far as we searched, no study has reported
MHLW are recommended (1-B). on the use of antibacterial agents specifically for the
Statement Of onset, it is not necessarily suitable in treatment of PEP. In general, the treatment of pancreatitis
actual clinical practice. Currently in Japan, therefore, to with antibacterial agents remains highly controversial, and
assess the severity of PEP, we consider it reasonable to use consensus has yet to be obtained. In particular, the need for
the MHLW severity assessment criteria for acute treating mild cases remains unclear, although antibacterial
pancreatitis. agents have been reported to be effective in patients with
As for patients whose conditions are complicated with severe pancreatitis [6–13]. Therefore, patients in whom
infections such as cholangitis or renal impairment, the PEP is expected to progress and become serious should
MHLW severity assessment criteria may lead to overesti- receive antibacterial agents intravenously or via the celiac
mation of severity. Therefore, severity assessment criteria artery or superior mesenteric artery to prevent the pro-
suitable for the pathology of PEP should be established gression of infection of the pancreas and surrounding tissue
(Table 4). to sepsis and multiple organ failure [6–8]. The

Table 4 Severity classification of PEP

Mild Moderate Severe

Clinical symptoms of acute pancreatitis Needs 4–10 days of hospitalization Needs at least 10 days of hospitalization
Elevated serum amylase level Cases of hemorrhagic pancreatitis
3 or more times higher than the normal level Necrosis or pseudocyst formation
(24 h after the procedure of ERCP) Needs percutaneous drainage or surgery
Needs emergency hospitalization
Extension of hospitalization by 2–3 days

123
 J Gastroenterol

recommended antibacterial agents are carbapenems such as in whom treatment was started within 48 h than in the
imipenem [14–16] and meropenem [17–19] and new qui- other groups [25] (than in patients in whom treatment was
nolones such as ciprofloxacin [20], which have a broad started more than 72 h after onset).
antibacterial spectrum and good penetration of pancreatic In a national survey of pancreatic regional arterial
tissue. infusion in patients with acute necrotizing pancreatitis, the
mortality rates were significantly lower in patients in whom
CQ9-02 Does the treatment of post-pancreatitis treatment was begun within 48 h after disease onset
ERCP with protease inhibitors shorten (11.9%) than in those in whom treatment was started more
the treatment period as compared with the nonuse than 48 h after onset (23.6%).
of these drugs?

Recommendation In patients with serious pancreatitis, CQ9-03 As compared with conventional infusion
pancreatic regional arterial infusion of protease inhibitors therapy, can high-volume infusion therapy improve
and antibacterial agents can decrease the incidence of treatment outcomes in patients with PEP? Does
infectious complications and mortality rates (2-B). In par- the infusion volume influence treatment outcomes
ticular, this trend is significant in patients in whom treat- in patients with PEP?
ment is begun within 48 h.
Statement At present, a general consensus has yet to be Recommendation In acute pancreatitis, increased vascular
reached concerning the use of protease inhibitors for permeability can lead to the extravasation of plasma
treatment of PEP alone. In principle, however, patients in components into interstitial spaces, resulting in dehydration
whom acute pancreatitis is diagnosed are generally given and shock. Appropriate infusion therapy should, therefore,
protease inhibitors, approved for this indication by be given soon after disease onset to maintain a urine vol-
National Health Insurance. As for the administration ume of at least 0.5 mL/kg/h (1-A). Subsequently, the
method, drug penetration is poorer in pancreatic tissue than infusion volume should be managed while closely moni-
in the liver and kidney. In particular, because acute toring variables such as hemodynamics and urine volume
necrotic pancreatitis is associated with pancreatic circula- to decrease the rates of complications and mortality (1-A).
tory disturbances, drug penetration is considered poor after Statement In patients with pancreatitis, extravasation of
intravenous administration. To solve this problem, a tech- plasma components into interstitial spaces due to increased
nique for arterial infusion therapy has been developed to vascular permeability can cause dehydration and shock.
directly infuse drugs into arteries that supply the pancreas. Therefore, adequate infusion therapy should be begun soon
This method for arterial infusion therapy has been reported after disease onset to maintain a urine volume of at least
to increase concentrations of protease inhibitors and 0.5 mL/kg/h [26]. The results of a national survey con-
antibacterial agents in pancreatic tissue, inhibit the pro- ducted in 2009 by the Research Committee on Pancreatic
gression of pancreatitis, and decrease the risk of infectious Diseases, supported by the MHLW of Japan, showed that
complications [21–23]. the infusion volume given within the first 24 h after start-
In a study comparing pancreatic regional arterial ing infusion therapy was inadequate (less than 50 mL/kg)
infusion of nafamostat (a protease inhibitor) plus imipe- in all patients younger than 60 years who died of severe
nem (an antibacterial agent), arterial infusion of nafa- pancreatitis [27]. As compared with healthy adults who
mostat alone (an antibacterial agent was given require a water intake of 1500–2000 mL per day
intravenously), and intravenous infusion of nafamostat (25–30 mL/kg body weight), patients with early-stage
plus imipenem in patients with acute necrotizing pancre- acute pancreatitis require a 2–5 times higher infusion vol-
atitis, the mortality rates were 6.7, 13.6, and 43.8% and ume (50–150 mL/kg body weight). At this time, extracel-
the incidences of infectious necrosis as a complication lular fluid (acetated Ringer’s solution or lactated Ringer’s
were 0, 22.8, and 50%, respectively. These results solution) is used. In particular, half to one third of the daily
demonstrated that pancreatic regional arterial infusion of infusion volume should be given within the first 6 h. Even
nafamostat plus imipenem was superior to the other elderly patients and those with cardiovascular, pulmonary,
treatments [24]. In another study, continuous regional or renal dysfunction should receive an adequate infusion
arterial infusion of nafamostat plus imipenem was started volume while continuously monitoring the central venous
within 48 h after disease onset, 48–72 h after onset, or pressure and other cardiovascular variables. If necessary,
more than 72 h after onset in patients with acute necro- patients should be transferred to a full-service hospital with
tizing pancreatitis. The incidence of respiratory failure an intensive care unit that can carefully manage cardio-
and the mortality rate were significantly lower in patients vascular condition.

123
J Gastroenterol

CQ10-01 Do protease inhibitors prevent PEP? Are CQ10-02 Do nonsteroidal anti-inflammatory drugs
there differences according to drug, dosing regimen, (NSAIDs) prevent PEP? Are there differences
or dose? according to dose?

Recommendation The results of individual RCTs have Recommendation NSAIDs (transanal administration of
demonstrated that protease inhibitors may reduce the indomethacin or diclofenac 50 or 100 mg) have a pro-
incidence of PEP; thus, they are generally used in clinical phylactic effect on PEP (2-B).
practice at present (1-C). The results of a recent meta- Transanal administration of indomethacin or diclofenac
analysis have shown that protease inhibitors such as 50 or 100 mg immediately before or after ERCP is rec-
gabexate and ulinastatin did not reduce the incidence of ommended; however, the dose of NSAIDs in the Japanese
PEP, and thus, evidence of the prevention of PEP was population needs to be examined in the future (2-B).
lacking (2-C). Five RCTs have exhibited that nafamostat Statement A meta-analysis of four RCTs in 2008 and
mesilate has a prophylactic effect on PEP; however, further 2009 [44–47] has examined the transanal administration
studies are needed (2-B). of indomethacin or diclofenac 100 mg (indomethacin
Statement A meta-analysis of 18 RCTs in 2011 has administration immediately before ERCP in two RCTs
reported that protease inhibitors cannot prevent PEP [28]. and diclofenac administration immediately after ERCP in
Several studies have been conducted to examine whether two RCTs). It has reported that the prevention of PEP was
there are differences according to the type of protease significant in the NSAIDs group (relative risk [RR] 0.36,
inhibitors (gabexate mesilate, ulinastatin, and nafamostat 95% CI 0.22–0.60, number needed to treat [NNT] 15),
mesilate). and there were no adverse events associated with
With regard to gabexate mesilate, six high-quality RCTs NSAIDs.
[29–34] reported no difference in the incidence of PEP A meta-analysis of nine RCTs in Japan in 2013 [47] has
between the control group (6.3%) and the gabexate mesi- demonstrated the efficacy of NSAIDs (summary
late group (4.5%). In addition, a meta-analysis of four RR = 0.58, 95% CI = 0.44–0.76). Furthermore, subgroup
RCTs [35] and a meta-analysis of five RCTs [36] have analysis has also exhibited a prophylactic effect for both
demonstrated that gabexate mesilate has no prophylactic indomethacin and diclofenac.
effect on PEP (OR 0.67, 95% CI 0.31–1.47), hyperamy- A recent overseas meta-analysis [48] has concluded that
lasemia, or abdominal pain. transanal administration of NSAIDs has the most superior
In a meta-analysis of studies of dosing regimens of prophylactic effect (summary RR = 0.37, 95%
gabexate mesilate, [37] neither short-term administration CI = 0.21–0.59).
over 6 h or less nor long-term administration over 12 h or The prophylactic effect of NSAIDs (transanal adminis-
more was correlated with a prophylactic effect on PEP. tration of indomethacin or diclofenac 50 or 100 mg) has
Another meta-analysis of two RCTs [38] has concluded been reported in many studies with a high evidence level
that long-term administration of gabexate mesilate had no [49]; however, NSAID administration is not covered by
prophylactic effect on PEP. health insurance in Japan. Since NSAIDs cause adverse
A subsequent meta-analysis [39] of seven RCTs in 2013 reactions, their use as prophylactic treatment, including
has also reported that gabexate mesilate is not effective. their doses, in all ERCP patients, needs to be examined
There were four RCTs of ulinastatin; [40–43] in studies further.
of ulinastatin versus control, pre-ERCP administration of
ulinastatin 150,000 U showed a prophylactic effect [41], CQ10-03 Does somatostatin prevent PEP?
whereas post-ERCP administration of ulinastatin
100,000 U had no prophylactic effect [43]. In comparative Recommendation Evidence of the efficacy of somatostatin
studies of gabexate mesilate in Japan, [40, 42] no statisti- on post-ERCP prevention is poor. Somatostatin may pre-
cally significant difference was noted between among uli- vent PEP depending on dosing regimens (12-h adminis-
nastatin 450,000 U (high dose), ulinastatin 150,000 U (low tration of high-dose somatostatin and bolus
dose), and gabexate mesilate 900 mg [40] or between uli- administration). However, it is recommended to use
nastatin 150,000 U and gabexate mesilate 600 mg [42]. somatostatin only in research setting (2-B).
Consequently, it was concluded that ulinastatin has no
prophylactic effect on PEP. CQ10-04 Do steroids prevent PEP?
Similarly, a study of six RCTs [39] has concluded that
ulinastatin had no efficacy. Recommendation Evidence of a prophylactic effect of
A study of five RCTs [39] has concluded that nafamostat steroids on PEP is poor, and it is recommended that ster-
mesilate has a prophylactic effect on PEP. oids not be administered (2-A).

123
 J Gastroenterol

CQ10-05 Does pancreatic duct stent placement and 0.38 (95% CI 0.19–0.76), respectively. The success rate
prevent PEP? Does the prophylactic effect differ of deep cannulation of the bile duct was improved according
according to the diameter, length, and form to both reports. A meta-analysis of 12 RCTs in the 2012
of pancreatic duct stents? Cochrane review [70] has shown that WGC, which signifi-
cantly increases the cannulation rate and reduces the risk of
Recommendation Prophylactic pancreatic duct stent PEP, is the most appropriate first-line cannulation technique.
placement is recommended in patients at high risk for PEP Although these results have been obtained in overseas
(2-A). The use of a straight-type, 5-Fr pancreatic sponta- meta-analyses, the results of multicenter studies in Japan
neous dislodgement stent is recommended. Sufficient [57, 58] have demonstrated that WGC does not decrease
attention should be paid to stent dislodgement; when the the incidence of PEP or increase the cannulation success
stent is not dislodged spontaneously, it needs to be endo- rate as compared with the conventional method. The two
scopically removed (2-A). techniques are used differently, depending on judgments of
Statement Two meta-analyses [50, 51] have shown that operators at each institution, at present.
pancreatic duct stents have a prophylactic effect in patients
at high risk for PEP.
A meta-analysis of eight RCTs and non-RCTs in 2011 Discussion
[52] has revealed that pancreatic duct stents have a pro-
phylactic effect on both PEP and hyperamylasemia in Clinical studies about PEP are sparse because the number
patients at high risk for PEP. of study participants is limited and effectiveness of
A meta-analysis of RCTs published in Japan in 2007 potential treatments is not so high that small-size studies
[53] has shown that pancreatic spontaneous dislodgement can prove. Further difficulty to conduct valid studies is that
stents significantly prevented PEP, regardless of the pres- patients who needs ERCP are heterogeneous in etiology of
ence of risk factors for PEP (the PEP incidence in the stent individual disease, location of pathologic involvement,
and no-stent groups was 3.2 and 13.6%, respectively; clinical stage, etc. Therefore, in many cases, we had to
P = 0.019). make recommendations on weak evidence with moderate
An RCT comparing 5-Fr and 3-Fr pancreatic duct stents to high uncertainty.
[54, 55] has revealed that the prophylactic effect on PEP We extracted key information from our full version of
was comparable, with a lower frequency of stent placement clinical practice guidelines about PEP. We described
failure with 5-Fr stents. clinical questions and recommendations but we could not
If the pancreatic duct stent is not dislodged within provide a statement part for the all recommendations due to
5–10 days after ERCP, it may induce pancreatitis; there- limited space. However, we think that the clinical questions
fore, endoscopic stent removal is recommended [54, 56]. and recommendations are useful for care of patients who
need ERCP.
CQ10-06 Does wire-guided cannulation (WGC)
prevent PEP?
Conclusion
Recommendation In deep cannulation of the bile duct,
WGC is expected to reduce the incidence of PEP and We presented with recommendations pertaining to risk
increase the cannulation success rate as compared with factors, diagnosis, prognostic factors, treatments, and pre-
conventional contrast-enhanced imaging. WGC is consid- ventive interventions in the medical practice for PEP.
ered in deep cannulation of the bile duct (2-B).
Compliance with ethical standards
Recent studies in Japan have reported that WGC does
not reduce the incidence of PEP or increase the cannulation Conflict of interest The authors declare that they have no conflict of
success rate as compared with the conventional method; interest.
the two are used differently, depending on the judgments of
operators at each institution, at present.
Statement Some reports state that there is no difference in
the incidence of PEP between the conventional contrast- References
enhanced imaging method and WGC, whereas a meta-
analysis of four RCTs [67–69] has concluded that the inci- 1. IOM (Institute of Medicine). Finding What Works in Health
Care: Standards for Systematic Reviews. Washington, DC: The
dence of PEP was lower with WGC than with conventional National Academies Press. 2011. (http://www.iom.edu/Reports/
contrast-enhanced imaging. Analyses of five RCTs [67] and 2011/Finding-What-Works-in-Health-Care-Standards-for-Sys
seven RCTs [69] yielded ORs of 0.23 (95% CI 0.13–0.41) tematic-Reviews/Standards.aspx).

123
J Gastroenterol

2. Graham R, Mancher M, Wolman DM, et al. Committee on 21. Kadokawa Y, Takeda K, Sunamura M, et al. Effectiveness of
Standards for Developing Trustworthy Clinical Practice Guide- continuous arterial infusion of protease inhibitors for experi-
lines; Institute of Medicine: Clinical Practice Guidelines We Can mental acute pancreatitis caused by the creation of a blind duo-
Trust. (http://www.nap.edu/catalog/13058/clinical-practice-guide denal loop. J Jpn Soc Gastroenterol. 1990;87:1444–50.
lines-we-can-trust). 22. Hayashi J, Kawarada Y, Isaji S, et al. Therapeutic effects of
3. Schünemann H, Bro_zek J, Guyatt G, et al. GRADE handbook for continuous itraarterial antibiotics infusion in preventing pancre-
grading quality of evidence and strength of recommendations. atic infection in experimental acute necrotizing pancreatitis.
Updated October 2013. The GRADE Working Group, 2013. Pancreas. 1996;13:184–92.
(www.guidelinedevelopment.org/handbook). 23. Mikami Y, Takeda K, Matsuda K, et al. Rat experimental model
4. Minds Guideline Center, Japan Council for Quality Health Care. of regional arterial infusion of protease inhibitor and its effects on
Minds handbook for clinical practice guideline development severe acute pancreatitis. Pancreas. 2005;30:248–53.
2014. 2014. 24. Takeda K, Matsuno S, Sunamura M, et al. Continuous regional
5. Whiting PF, Rutjes AW, Westwood ME, et al. QUADAS-2 arterial infusion of protease inhibitor and antibiotics in acute
Group: QUADAS-2: a revised tool for the quality assessment of necrotizing pancreatitis. Am J Surg. 1996;171:394–8 (level 3b).
diagnostic accuracy studies. Ann Intern Med. 2011;155:529–36. 25. Takeda K, Yamauchi J, Shibuya K, et al. Benefit of continuous
6. Villatoro E, Bassi C, Larvin M. Antibiotics therapy for prophy- regional arterial infusion of protease inhibitor and antibiotics in
laxis against infection of pancreatic necrosis in acute pancreatitis. the management of acute necrotizing pancreatitis. Pancreatology.
Cochrane Database Syst Rev. 2006;4:CD002941. 2001;1:668–73 (level 3b).
7. Working UK. Party on acute pancreatitis: UK guidelines for 26. Consensus Revision Committee on Early Diagnosis and Treat-
management of acute pancreatitis. Gut. 2005;54(Suppl 3):31–9. ment of Acute Pancreatitis. Revised guidelines for the early
8. Forsmark CE, Baillie J. AGA Institute Clinical Practice and diagnosis and treatment of acute pancreatitis by the Consensus
Economics Committee, AGA Institute Governing Board, AGA Revision Committee, third edition. Pancreas. 2011;26:651–83.
Institute technical review on acute pancreatitis. Gastroenterology. 27. Mine T, Akashi R, Igarashi Y, et al. Ministry of Health, Labour
2007;132:2022–44. and Welfare Grant-in-Aid for Scientific Research. Research on
9. Golub R, Siddiqi F, Pohl D. Role of antibiotics in acute pan- measures for intractable disease. Survey on Intractable Diseases.
creatitis: a meta-analysis. J Gastrointest Surg. 1998;2:496–503. 2008 Technical report on survey of intractable pancreatic disease
10. Sharma VK, Howden CW. Prophylactic antibiotics administra- 2009; 77–78.
tion reduces sepsis and mortality in acute necrotizing pancreatitis: 28. Seta T, Noguchi Y. Protease inhibitors for preventing compli-
a meta-analysis. Pancreas. 2001;22:28–31. cations associated with ERCP: an updated meta-analysis. Gas-
11. de Vries AC, Besselink MGH, Buskens E, et al. Randomized trointest Endosc. 2011;73(4):700–706.e1-2.
controlled trials of antibiotic prophylaxis in severe acute pan- 29. Andriulli A, Clemente R, Solmi L, et al. Gabexate or somato-
creatitis: relationship between methodological quality and out- statin administration before ERCP in patients at high risk for
come. Pancreatology. 2007;7:531–8. post-ERCP pancreatitis: a multicenter, placebo-controlled, ran-
12. Bai Y, Gao J, Zou D, et al. Prophylactic antibiotics cannot reduce domized clinical trial. Gastrointest Endosc. 2002;56:488–95.
infected pancreatic necrosis and mortality in acute necrotizing 30. Andriulli A, Solmi L, Loperfido S, et al. Prophylaxis of ERCP-
pancreatitis: evidence from meta-analysis of randomized con- related pancreatitis: a randomized, controlled trial of somatostatin
trolled trials. Am J Gastroenterol. 2008;103:104–10. and gabexate mesylate. Clin Gastroenterol Hepatol. 2004;2:713–8.
13. Akshintala VS, Hutfless SM, Colantuoni E, et al. Systematic 31. Benvenutti S, Zancanella L, Piazzi L, et al. Prevention of post-ERCP
review with network meta-analysis: pharmacological prophylaxis pancreatitis with somatostatin versus gabexate mesylate: a random-
against post-ERCP pancreatitis. Aliment Pharmacol Ther. ized placebo controlled multicenter study. Dig Liv Dis. 2006;38:S15.
2013;38:1325–37. 32. Cavallini G, Tittobello A, Frulloni L, et al. Gabexate for the
14. Pederzoli P, Bassi C, Vesentini S, et al. A randomized multi- prevention of pancreatic damage related to endoscopic retrograde
center clinical trial of antibiotics prophylaxis of septic compli- cholangiopancreatography. Gabexate in digestive endoscopy—
cations in acute necrotizing pancreatitis with imipenem. Surg Italian Group. N Engl J Med. 1996;335:919–23.
Gynecol Obstet. 1993;176:480–3. 33. Manes G, Ardizzone S, Lombardi G, et al. Efficacy of postpro-
15. Schwarz M, Isenmann R, Meyer H, et al. Antibiotic use in cedure administration of gabexate mesylate in the prevention of
necrotizing pancreatitis. Results of controlled study. Dtsch Med post-ERCP pancreatitis: a randomized, controlled, multicenter
Wochenschr. 1997;122:356–61. study. Gastrointest Endosc. 2007;65:982–7.
16. Nordback I, Sand J, Saaristo R, et al. Early treatment with 34. Xiong GS, Wu SM, Zhang XW, et al. Clinical trial of gabexate in
antibiotics reduces the need for surgery in acute necrotizing the prophylaxis of post-endoscopic retrograde cholangiopancre-
pancreatitis—a single-center randomized study. J Gastrointest atography pancreatitis. Braz J Med Biol Res. 2006;39:85–90.
Surg. 2001;5:113–8. 35. Zheng M, Chen Y, Yang X, et al. Gabexate in the prophylaxis of
17. Manes G, Rabitti PG, Menchise A, et al. Prophylaxis with mer- post-ERCP pancreatitis: a meta-analysis of randomized con-
openem of septic complications in acute pancreatitis: a random- trolled trials. BMC Gastroenterol. 2007;7:6–13.
ized, controlled trial versus imipenem. Pancreas. 2003;27:e79–83. 36. Andriulli A, Leandro G, Federici T, et al. Prophylactic admin-
18. Manes G, Uomo I, Menchise A, et al. Timing of antibiotics istration of somatostatin or gabexate does not prevent pancreatitis
prophylaxis in acute pancreatitis: a controlled randomized study after ERCP: an updated meta-analysis. Gastrointest Endosc.
with meropenem. Am J Gastroenterol. 2006;101:1348–53. 2007;65:624–32.
19. Isennman R, Runzi M, Kroon M, et al. Prophylactic antibiotic 37. Rudin D, Kiss A, Wetz RV, et al. Somatostatin and gabexate for
treatment in patients with predicted severe acute pancreatitis: a post-endoscopic retrograde cholangiopancreatography pancreati-
placebo-controlled, double-blind trial. Gastroenterology. tis prevention: meta-analysis of randomized placebo-controlled
2004;126:997–1004. trials. J Gastroenterol Hepatol. 2007;22:977–83.
20. Grag PK, Khanna S, Bohidar NP, et al. Incidence, spectrum and 38. Tsujino T, Komatsu Y, Isayama H, et al. Ulinastatin for pan-
antibiotic sensitivity pattern of bacterial infections among creatitis after endoscopic retrograde cholangiopancreatography: a
patients with acute pancreatitis. J Gastroenterol Hepatol. 2001;16: randomized, controlled trial. Clin Gastroenterol Hepatol.
1055–9. 2005;3:376–83.

123
 J Gastroenterol

39. Ueki T, Otani K, Kawamoto K, et al. Comparison between uli- endoscopic pancreatic spontaneous dislodgement stent. Clin
nastatin and gabexate mesylate for the prevention of post-endo- Gastroenterol Hepatol. 2007;5:1339–46.
scopic retrograde cholangiopancreatography pancreatitis: a 50. Chahal P, Tarnasky PR, Petersen BT, et al. Short 5Fr vs long 3Fr
prospective, randomized trial. J Gastroenterol. 2007;42:161–7. pancreatic stents in patients at risk for post-endoscopic retrograde
40. Yoo JW, Ryu JK, Lee SH, et al. Preventive effects of ulinastatin cholangiopancreatography pancreatitis. Clin Gastroenterol
on post-endoscopic retrograde cholangiopancreatography pan- Hepatol. 2009;7(8):834–9.
creatitis in high-risk patients: a prospective, randomized, pla- 51. Fehmi SMA, Schoenfeld PS, Scheiman JM, et al. 5 Fr prophy-
cebo-controlled trial. Pancreas. 2008;37:366–70. lactic pancreatic stents are easier to place and require fewer guide
41. Dai HF, Wang XW, Zhao K. Role of nonsteroidal anti-inflam- wires than 3 Fr stents. Gastrointest Endosc. 2008;67:AB328–9.
matory drugs in the prevention of post-ERCP pancreatitis: a 52. Smith MT, Sherman S, Ikenberry SO, et al. Alterations in pan-
meta-analysis. Hepat Pancreat Dis Int. 2009;8:11–6. creatic ductal morphology following polyethylene pancreatic
42. Elmunzer B, Waljee A, Elta G, et al. A meta-analysis of rectal stent therapy. Gastrointest Endosc. 1996;44:268–75.
NSAIDs in the prevention of post-ERCP pancreatitis. Gut. 53. Bailey AA, Bourke MJ, Williams SJ, et al. A prospective ran-
2008;57:1262. domized trial of cannulation technique in ERCP: effects on
43. Zheng M-H, Xia H, Chen Y-P. Rectal administration of NSAIDs technical success and post-ERCP pancreatitis. Endoscopy.
in the prevention of post-ERCP pancreatitis: a complementary 2008;40:296–301.
meta-analysis. Gut. 2008;57:1632. 54. Katsinelos P, Paroutoglou G, Kountouras J, et al. A comparative
44. Akshintala VS, Hutfless SM, Colantuoni E, et al. Systematic study of standard ERCP catheter and hydrophilic guide wire in
review with network meta-analysis: pharmacological prophylaxis the selective cannulation of the common bile duct. Endoscopy.
against post-ERCP pancreatitis. Aliment Pharmacol Ther. 2008;40:302–7.
2013;38(11–12):1325–37. 55. Kawakami H, Maguchi H, Mukai T, et al. A multicenter,
45. Otsuka T, Kawazoe S, Nakashita S, et al. Low-dose rectal prospective, randomized study of selective bile duct cannulation
diclofenac for prevention of post-endoscopic retrograde cholan- performed by multiple endoscopists: the BIDMEN study. Gas-
giopancreatography pancreatitis: a randomized controlled trial. trointest Endosc. 2012;75:362–72.
J Gastroenterol. 2012;47(8):912–7. 56. Kobayashi G, Fujita N, Imaizumi K, et al. Wire-guided biliary
46. Singh P, Das A, Isenberg G, et al. Does prophylactic pancreatic cannulation technique does not reduce the risk of post-ERCP
stent placement reduce the risk of post-ERCP acute pancreatitis? pancreatitis: multicenter randomized controlled trial. Dig Endosc.
A meta-analysis of controlled trials. Gastrointest Endosc. 2013;25:295–302.
2004;60(4):544–50. 57. Cheung J, Tsoi KK, Quan W-L, et al. Guidewire versus con-
47. Andriulli A, Forlano R, Napolitano G, et al. Pancreatic duct stents ventional contrast cannulation of the common bile duct for the
in the prophylaxis of pancreatic damage after endoscopic retro- prevention of post-ERCP pancreatitis: a systematic review and
grade cholangiopancreatography: a systematic analysis of bene- meta-analysis. Gastrointest Endosc. 2009;70:1211–9.
fits and associated risks. Digestion. 2007;75(2–3):156–63. 58. Tse F, Yuan Y, Moayyedi P, et al. Guidewire-assisted cannula-
48. Choudhary A, Bechtold ML, Arif M, et al. Pancreatic stents for tion of the common bile duct for the prevention of post-endo-
prophylaxis against post-ERCP pancreatitis: a meta-analysis and scopic retrograde cholangiopancreatography (ERCP) pancreatitis.
systematic review. Gastrointest Endosc. 2011;73(2):275–82. Cochrane Database Syst Rev. 2012;12:CD009662.
49. Sofuni A, Maguchi H, Itoi T, et al. Prophylaxis of post-endo-
scopic retrograde cholangiopancreatography pancreatitis by an

123