I

n
t
h
e
C
l
i
n
i
c
In the Clinic
Acute
Pancreatitis
Prevention page ITC5-2
Diagnosis page ITC5-5
Treatment page ITC5-11
Practice Improvement page ITC5-13
Tool Kit page ITC5-14
Patient Information page ITC5-15
CME Questions page ITC5-16
Section Editors
Darren Taichman, MD, PhD
Barbara J. Turner MD, MSED
Sankey Williams, MD
Physician Writers
Kapil Gupta, MD, MPH
Bechien Wu, MD, MPH
The content of In the Clinic is drawn from the clinical information and
education resources of the American College of Physicians (ACP), including PIER
(Physicians Information and Education Resource) and MKSAP (Medical Knowledge
and Self-Assessment Program). Annals of Internal Medicine editors develop In the
Clinic from these primary sources in collaboration with the ACPs Medical Education
and Publishing Division and with the assistance of science writers and physician writ-
ers. Editorial consultants from PIER and MKSAP provide expert review of the content.
Readers who are interested in these primary resources for more detail can consult
http://pier.acponline.org, http://www.acponline.org/products_services/mksap/15/?pr31,
and other resources referenced in each issue of In the Clinic.
CME Objective: To revuew current evidence for the prevention, diagnosis, and
treatment of acute pancreatitis.
The information contained herein should never be used as a substitute for clinical
judgment.
2010 American College of Physicians
Who is at increased risk for acute
pancreatitis?
Of the many causes of acute pancre-
atitis, gallstone disease (approximately
35% to 40% of cases) and excessive
alcohol consumption (approximately
30% of cases) dominate (5) (Table 1).
Gallstone disease is among the most
common disorders in the United
States, affecting an estimated 6.3 mil-
lion men and 14.2 million women 20
to 74 years of age (6). It is difficult to
predict which patients with either
symptomatic or asymptomatic gall-
stones will develop pancreatitis. One
risk factor is the presence of stones in
the common bile duct (choledo-
cholithiasis), especially small stones
(<5 mm) or microlithiasis comprising
stones that measure <2 mm because
they can obstruct the orifice of the
pancreatic duct at the level of the am-
pulla. Pancreatitis occurs when gall-
stones pass into the bile duct and
become trapped at the sphincter of
Oddi, stopping the flow of pancreatic
fluid containing digestive enzymes
into the duodenum. If the blockage
continues, activated enzymes build up
in the pancreas and cause severe in-
flammation. To reduce the risk for
such complications as pancreatitis, pa-
tients with symptomatic gallstones
usually have cholecystectomy and
those with common bile duct stones
have them removed by endoscopic
retrograde cholangiopancreatography
(ERCP), an imaging and therapeutic
technique that combines endoscopy
and fluoroscopy.
Alcohol-related pancreatitis usually
occurs after long-term (>5 years),
heavy alcohol consumption. Risk
increases with the amount of alco-
hol consumed, indicative of a direct
toxic effect on the pancreas when
the alcohol is metabolized. Because
only about 5% of alcoholics develop
pancreatitis, additional unknown
genetic or other factors must in-
crease susceptibility. Smoking to-
bacco may play a role; it has been
reported to accelerate progression
of established alcoholic pancreatitis
(7). One study found an association
between high intake of beer (>14
drinks per week) and pancreatitis,
but not for wine or spirits (8).
Hypertriglyceridemia is another
important risk factor for pancreati-
tis, especially during pregnancy (9).
No clear risk profile can indicate
which patients with elevated
triglycerides will develop pancreati-
tis, but the complication occurs
rarely in the absence of significant
2010 American College of Physicians ITC5-2 In the Clinic Annals of Internal Medicine 2 November 2010
1. Bradley EL. A clinically
based classification
system for acute pan-
creatitis: Summary of
the International
Symposium on Acute
Pancreatitis, Atlanta,
Ga; September 11-13,
1992. Arch
Surg;1993;128:586-90.
[PMID:8489394]
2. Fagenholz PJ, Castillo
CF, Harris NS, Pelletier
AJ, Camargo CA Jr. In-
creasing United
States hospital ad-
missions for acute
pancreatitis, 1988-
2003. Ann Epidemiol.
2007;17:491-7.
[PMID:17448682]
3. Lowenfels AB,
Maisonneuve P, Sulli-
van T. The changing
character of acute
pancreatitis: epidemi-
ology, etiology, and
prognosis. Curr Gas-
troenterol Rep.
2009;11:97-103.
[PMID:19281696]
4. Lankisch PG, Breuer
N, Bruns A, et al. Nat-
ural history of acute
pancreatitis: a long-
term population-
based study. Am J
Gastroenterol.
2009;104:2797-805.
[PMID:19603011]
5. Forsmark CE, Baillie J.
AGA Institute Clinical
Practice and Eco-
nomics Committee.
AGA Institute techni-
cal review on acute
pancreatitis. Gas-
troenterology.
2007;132:2022-44.
[PMID:17484894]
A
cute pancreatitis is an acute inflammatory process of the pancreas
that can occur as an isolated event or relapsing episodes. Acute pan-
creatitis is a heterogeneous disease ranging from minimal pancreatic
inflammation seen in mild interstitial pancreatitis to extensive pancreatic
necrosis and liquefaction of severe attacks. Diagnosis is based on the pres-
ence at least 2 of 3 features: abdominal pain; increased pancreatic enzyme,
amylase, and/or lipase levels to 3 times the upper limit of normal; and im-
aging tests showing characteristic findings of acute pancreatitis (1). Alcohol
and gallstones are the two most common causes, but there are many less
common causes. Acute pancreatitis accounts for more than 200 000 hospi-
tal admissions annually in the United States, and incidence has been in-
creasing (2). The rates of acute pancreatitis per 1000 Americans 40 to 59
years of age are the highest they have been in the past 20 years, and rates
are higher for blacks than for whites. Mortality from acute pancreatitis is
<5% overall, but severe attacks cause longer hospitalization and significant-
ly higher mortality (3). The annual relapse rate of acute pancreatitis ranges
from 0.6% to 5.6%, depending on the cause, and is highest when pancre-
atitis results from alcohol consumption (4).
Prevention
2010 American College of Physicians ITC5-3 In the Clinic Annals of Internal Medicine 2 November 2010
elevations, usually >1000 to
2000 mg/dL (10).
Several drugs have been linked to
development of acute pancreatitis
(Table 1), but the risk is generally
low. In one review, the authors as-
sessed the evidence for specific
drugs causing acute pancreatitis as
well as their clinical presentations
and proposed a classification of
drug-induced pancreatitis (11).
Patients who develop apparent
drug-induced acute pancreatitis
should still be evaluated for other
causes before attributing an
episode to particular medications.
While searching for another, more
common cause of acute pancreati-
tis, the temporal association of
medication use and development
of the episode should be evaluat-
ed. The clinician needs to recog-
nize that drug-induced pancreati-
tis can occur at any point in the
course of a medication regimen,
ranging from at or shortly after
initiation to an idiosyncratic
reaction after prolonged use. It
may be necessary to rechallenge
with the drug if it is critical for
the patients health. In general,
drug-induced acute pancreatitis is
less common than was previously
believed, and without strong evi-
dence for drug-related pancreati-
tis, medications can usually be
continued (12).
6. Everhart JE, Khare M,
Hill M, Maurer KR.
Prevalence and eth-
nic differences in
gallbladder disease in
the United States.
Gastroenterology.
1999;117:632-9.
[PMID:10464139]
7. Yadav D, Whitcomb
DC. The role of alco-
hol and smoking in
pancreatitis. Nat Rev
Gastroenterol Hepa-
tol. 2010;7:131-45.
[PMID:20125091]
8. Kristiansen L, Grn-
baek M, Becker U, Tol-
strup JS. Risk of pan-
creatitis according to
alcohol drinking
habits: a population-
based cohort study.
Am J Epidemiol.
2008;168:932-7.
[PMID:18779386]
9. Ewald N, Hardt PD,
Kloer HU. Severe hy-
pertriglyceridemia
and pancreatitis:
presentation and
management. Curr
Opin Lipidol.
2009;20:497-504.
[PMID:19770656]
10. Yadav D, Pitchumoni
CS. Issues in hyper-
lipidemic pancreati-
tis. J Clin Gastroen-
terol. 2003;36:54-62.
[PMID:12488710]
11. Badalov N, Baradari-
an R, Iswara K, Li J,
Steinberg W, Tenner
S. Drug-induced
acute pancreatitis:
an evidence-based
review. Clin Gas-
troenterol Hepatol.
2007;5:648-61.
[PMID:17395548]
12. Nitsche CJ, Jamieson
N, Lerch MM, Mayer-
le JV. Drug induced
pancreatitis. Best
Pract Res Clin Gas-
troenterol.
2010;24:143-55.
[PMID:20227028].
Table 1. Causes of Acute Pancreatitis
More Common Causes Comments
Gallstones and microlithiasis Most common cause
Alcohol abuse Alcohol-related disease usually occurs only occurs after
>510 y of heavy drinking
Drugs More common in older patients, HIV-positive persons, or in
those receiving immunomodulating agents
ERCP Can be a trigger, particularly if performed by an inexperienced
clinician or if the patient has sphincter of Oddi dysfunction
Hyperlipidemia Usually with extremely elevated triglyceride levels (>1000 mg/dL)
Hypercalcemia Commonly caused by hyperparathyroidism or cancer, can be a
trigger by increasing activation of trypsinogen
Genetic Hereditary, and research has linked gene mutations in cationic
trypsinogen (PRSS1), SPINK1, or CFTR genes with acute and
chronic pancreatitis
Autoimmune pancreatitis Diffuse sausage shaped finding on imaging with rim enhance-
ment or ductal abnormalities.
Infections Includes viruses: mumps, coxsackievirus, cytomegalovirus, varicella,
HSV, HIV; bacteria: Mycoplasma, Legionella, Leptospira, Salmonella;
Parasites: Toxoplasma, Cryptosporidium, Ascaris; and fungi:
Aspergillus
Idiopathic Accounts for approximately 15%20% of cases; causes include
sphincter of Oddi dysfunction, microlithiasis, and biliary sludge;
anatomical abnormalities
Less common causes
Cystic lesions of the pancreas More likely if cysts involve the main duct, such as pancreatic
intraductal papillary mucinous tumor
Cystic fibrosis Rare, occurs when some viable pancreatic tissue remains
Pancreas divisum Controversial as a cause so exclude all other causes first
Pancreatic cancer Focal pancreatitis can indicate an underlying mass
Penetrating peptic ulcer Rare, clue is thickening of the duodenal wall
Postsurgical Such as ischemia related to bypass surgery
Trauma History is usually compelling
Tropical pancreatitis Endemic in some parts of Asia and Africa
Vasculitis Rare even in patients with vasculitis
ERCP = endoscopic retrograde cholangiopancreatography; HSV = herpes simplex virus.
13. Cheng CL, Sherman
S, Watkins JL, et al.
Risk factors for post-
ERCP pancreatitis: a
prospective multi-
center study. Am J
Gastroenterol.
2006;101:139-47.
[PMID:16405547]
14. Saad AM, Fogel EL,
McHenry L, et al.
Pancreatic duct
stent placement
prevents post-ERCP
pancreatitis in pa-
tients with suspect-
ed sphincter of Oddi
dysfunction but nor-
mal manometry re-
sults. Gastrointest
Endosc. 2008;67:
255-61.
[PMID:18028920].
15. Nordback I, Pelli H,
Lappalainen-Lehto R,
et al. The recurrence
of acute alcohol-as-
sociated pancreatitis
can be reduced: a
randomized con-
trolled trial. Gas-
troenterology.
2009;136:848-55.
[PMID:19162029]
2010 American College of Physicians ITC5-4 In the Clinic Annals of Internal Medicine 2 November 2010
An important iatrogenic risk factor
for development of acute pancreati-
tis is ERCP. The risk for acute pan-
creatitis related to ERCP ranges
from 5% to 20%, depending on
physician-related factors, such as
the level of experience performing
the procedure, and patient charac-
teristics, especially sphincter of
Oddi dysfunction and a history of
previous ERCP-related pancreatitis
(13). From a technical standpoint,
the incidence of ERCP-related
pancreatitis seems to be decreased
by placement of a pancreatic duct
stent at the time of ERCP (14).
Careful patient selection and avoid-
ance of ERCP, unless clearly indi-
cated, will decrease the risk for
acute pancreatitis resulting from
this procedure.
Other less common causes of
acute pancreatitis are listed in
Table 1. Rare causes of unex-
plained acute pancreatitis include
cancer; mucinous neoplasm; re-
mote history of trauma; infections
caused by parasites, such as toxo-
plasmosis and cryptosporidiosis;
and viruses (cytomegalovirus,
Epstein Barr virus). Autoimmune
processes leading to pancreatitis,
such as vasculitis and autoimmune
pancreatitis, are well described but
underrecognized.
What behavioral advice should
clinicians give to a patient to
minimize the chance of a repeated
episode of acute pancreatitis?
After a clear cause of acute pan-
creatitis has been identified, ef-
forts should be made not only to
eliminate the cause but also to
provide counseling and education
for the patient about the need to
avoid known risk factors for the
disease. When alcohol consump-
tion has been identified as the
cause, patients should be evaluated
for alcohol abuse or dependence.
The patient should receive inten-
sive counseling about the exigency
of abstaining from alcohol to
avoid repeated episodes of acute
pancreatitis or chronic pancreati-
tis, as well as the other well-
known complications of alcohol
abuse and dependence. In this sit-
uation, one brief alcohol counsel-
ing session will not suffice.
In a randomized, controlled trial of 120
patients hospitalized for a first episode of
alcohol-associated acute pancreatitis, 59
patients received repeated 30-minute al-
cohol reduction and social stressor coun-
seling intervention both before discharge
and after 6-months while the 61 control
participants received only the initial
counseling session (15). Over the next
2 years, significantly fewer recurrent
episodes occurred in the patients with re-
peated alcohol counseling.
In addition, referral to alcohol spe-
cialty treatment will improve absti-
nence, ideally with support to
ensure that the patient receives
this care.
As noted, there are few drugs with
a definite link to acute pancreatitis
(12). Physicians should be alert
for drug-induced pancreatitis in
certain groups, such as the elderly
or patients with HIV infection or
cancer, who often take multiple
medications (16). However, even
when the association seems to be
clear, questions may linger with
regard to whether it was the drug
or the underlying condition for
which the drug was prescribed
that caused the pancreatitis.
Patients who develop acute pan-
creatitis because of hypertriglyc-
eridemia should be counseled
about lifestyle modifications, such
as reducing sugars and unhealthy
fats, and should have aggressive
medical interventions (fibrates or
nicotinic acid) to reduce triglyc-
eride levels to normal. When the
triglyceride level is 500 mg/dL,
the first priority is to prevent
acute pancreatitis by reducing the
level to <500; reducing the risk for
coronary heart disease is a second-
ary goal, according to an expert
panel report (17).
16. Trivedi CD, Pitchu-
moni CS. Drug-in-
duced pancreatitis:
an update. J Clin
Gastroenterol.
2005;39:709-16.
[PMID:16082282]
17. Third Report of the
Expert Panel on De-
tection, Evaluation,
and Treatment of
High Blood Choles-
terol in Adults (ATP
III Final Report). Cir-
culation.
2002;106:3143-421.
[PMID:12485966].
18. Yadav D, Agarwal N,
Pitchumoni CS. A
critical evaluation of
laboratory tests in
acute pancreatitis.
Am J Gastroenterol.
2002;97:1309-18.
[PMID:12094843]
2010 American College of Physicians ITC5-5 In the Clinic Annals of Internal Medicine 2 November 2010
sequestration seen in more severe
cases. Jaundice indicates biliary tree
obstruction. The clinician should
perform a careful abdominal exami-
nation focusing especially on aus-
cultation for bowel sounds, location
of pain, guarding (usually severe),
rebound, and distention. Distention
with absent bowel sounds indicates
ileus. Ecchymosis in the flanks
(Grey-Turner sign) or around the
umbilicus (Cullen sign) are indica-
tive of blood in the abdomen from
pancreatic necrosis. Mental status
impairment is also an indicator of
more severe pancreatitis and may
occur as a result of septicemia, hy-
poxemia, electrolyte imbalance, or
alcohol use. Multiorgan dysfunc-
tion signifies a more severe episode
with complications, such as pancre-
atic necrosis.
A patient with gallstones and a his-
tory of fever, chills, and/or rigors
suggests ascending cholangitis, but
these symptoms may be due only to
the inflammatory process associat-
ed with acute pancreatitis.
What laboratory tests are useful
in the evaluation of acute
pancreatitis?
Elevation of the serum amylase
and/or lipase levels at least three
times the upper limit of normal is a
key component of diagnosing acute
pancreatitis. Measurement of serum
amylase levels has good sensitivity
but low specificity, signifying a high
false-positive rate (18). Other causes
What elements of the history
and examination are helpful in
suggesting a diagnosis of acute
pancreatitis?
The most common presenting
symptom of acute pancreatitis is
abdominal pain, classically de-
scribed as occurring in the upper
abdomen and radiating to the back.
The pain is typically severe and
persistent without alleviating or
relieving factors and is usually asso-
ciated with nausea and vomiting.
When ileus is present, vomiting
reduces the pain associated with
acute pancreatitis only slightly.
In patients with suspected acute
pancreatitis, a detailed history
should address the potential causes
listed in Table 1. Previous cholecys-
tectomy for gallstones in a person
with no or minimal use of alcohol
increases the likelihood of pancre-
atitis due to retained gallstones.
Careful history should assess for
hyperlipidemia, abdominal trauma,
similar previous episodes, or prior
ERCP. A detailed list of medica-
tions must be reviewed, focusing on
the likelihood of a drug being the
cause as well as timing of use (11).
On physical examination, vital
signs including pulse, orthostatic
blood pressure, and respiratory rate
must be performed to evaluate hy-
dration status and indicate the
severity of pancreatitis. Tachycardia
and hypotension represent intravas-
cular depletion secondary to fluid
Prevention... Gallstones and excessive alcohol consumption are the two most
common causes of acute pancreatitis. It is not possible to predict which patients
with these conditions will develop this complication. Removal of gallstones and
alcohol cessation can help prevent recurrences. Other less common causes include
hypertriglyceridemia and side effects of medications, but alcohol and gallstones
should first be ruled out as sole or concurrent causes. Recurrent pancreatitis re-
lated to hypercalcemia is best prevented through treatment of the underlying
cause. Iatrogenic acute pancreatitis due to ERCP can be reduced by careful pa-
tient selection and possibly through placement of a pancreatic duct stent.
CLINICAL BOTTOM LINE
Diagnosis
19. Liu KJ, Atten MJ,
Lichtor T, et al.
Serum amylase and
lipase elevation is as-
sociated with in-
tracranial events. Am
Surg. 2001;67:215-9;
discussion 219-20.
[PMID:11270877]
20. Seno T, Harada H,
Ochi K, et al. Serum
levels of six pancre-
atic enzymes as re-
lated to the degree
of renal dysfunction.
Am J Gastroenterol.
1995;90:2002-5.
21. Manjuck J, Zein J,
Carpati C, Astiz M.
Clinical significance
of increased lipase
levels on admission
to the ICU. Chest.
2005;127:246-50.
[PMID:15653991]
22. Wachter RM, Gold-
man, L, Hollander H.
Hospital Medicine.
Philadelphia: Wolters
Kluwer Health; 2005.
2010 American College of Physicians ITC5-6 In the Clinic Annals of Internal Medicine 2 November 2010
of elevated serum amylase levels in-
clude disorders of salivary glands and
fallopian tubes, intestinal ischemia,
perforated peptic ulcer, and chronic
renal insufficiency. To improve speci-
ficity, the level of the serum amylase
or lipase needs to be three times nor-
mal. Measurement of lipase levels is
more sensitive than that of amylase
levels in acute alcoholic pancreatitis
or when patients present to the
emergency department days after
disease onset because it remains ele-
vated for a longer period. However,
lipase can also be falsely elevated in
cases of renal insufficiency and head
trauma or an intracranial mass as
well as in patients receiving heparin
therapy (through activation of
lipoprotein lipase) (19, 20). Elevated
serum lipase levels are also common
among critically ill patients in the in-
tensive care unit (ICU) (21). Simul-
taneous measurement of amylase and
lipase levels does not seem to im-
prove diagnostic accuracy (18).
No enzyme assay can assess the
severity or cause of an episode of
acute pancreatitis. Serum C-reactive
protein at 48 hours is the best avail-
able laboratory marker of severity.
Liver enzymes should also be rou-
tinely checked. Elevated liver en-
zymes (alanine transaminase) >150
IU/L has a 95% positive predictive
value and a specificity of 96% but
sensitivity of less than 50%; the ac-
curacy of the aspartate transaminase
is similar (22). Elevations can sug-
gest gallstone pancreatitis. Triglyc-
eride levels should be checked
because levels above >1000 mg /dL
can precipitate acute pancreatitis that
is often severe. A low serum calcium
level can cause acute pancreatitis but
may also be a consequence of acute
pancreatitis due to other causes (23).
The presence of leukocytosis on
complete blood count may result
from the acute pancreatic inflam-
mation alone or point to an
underlying infectious process. In-
creased hematocrit and blood urea
nitrogen (BUN) levels may reveal
hemoconcentration, indicative of
fluid sequestration. Early changes
in the serial BUN levels provide the
most useful assessment of response
to initial resuscitation (24).
An acute drop in hemoglobin in an
unstable patient may represent he-
morrhagic pancreatitis. Patients
with pancreatitis may also develop
disseminated intravascular coagu-
lopathy, perhaps due to circulating
pancreatic enzymes or to vascular
injury precipitating consumption of
coagulation factors (25).
What other diagnoses should
clinicians consider in a patient
with possible acute pancreatitis?
The clinical presentation of upper
abdominal pain with nausea, vomit-
ing, and fever has a broad differential
(Table 2). Although peptic ulcer per-
foration often mimics this presenta-
tion, it is distinguished by free air
seen on imaging studies. Acute
cholecystitis, symptomatic choledo-
cholithiasis, and cholangitis are typi-
cally described as causing right upper
quadrant pain but can also present
with epigastric pain similar to that of
acute pancreatitis. Normal serum
amylase and lipase levels as well as
characteristic imaging findings, such
as gallbladder wall thickening
(cholecystitis) or common bile duct
stones (choledocholithiasis), help
differentiate biliary disease from
acute pancreatitis but, as noted, acute
pancreatitis may also present with
gallstone-related biliary obstruction.
Patients with intestinal obstruction
will have abdominal distention, col-
icky abdominal pain, and an obstruc-
tive bowel pattern on imaging. They
may also have elevated serum amy-
lase levels but these levels are usually
lower than those associated with
acute pancreatitis. Mesenteric vascu-
lar obstruction should be suspected
in patients with underlying vascular
or cardiac disease. Pain associated
with nonobstructive mesenteric is-
chemia is usually postprandial, and
on rare occasions an abdominal bruit
may be heard. Table 2 lists additional
23. Schtte K, Malfer-
theiner P. Markers for
predicting severity
and progression of
acute pancreatitis.
Best Pract Res Clin
Gastroenterol.
2008;22:75-90.
[PMID:18206814]
24. Wu BU, Johannes RS,
Sun X, Conwell DL,
Banks PA. Early
changes in blood
urea nitrogen pre-
dict mortality in
acute pancreatitis.
Gastroenterology.
2009;137:129-35.
[PMID:19344722]
25. Saif MW. DIC sec-
ondary to acute
pancreatitis. Clin Lab
Haematol.
2005;27:278-82.
[PMID:16048498]
26. Nichols MT, Russ PD,
Chen YK. Pancreatic
imaging: current and
emerging technolo-
gies. Pancreas.
2006;33:211-20.
[PMID:17003640]
2010 American College of Physicians ITC5-7 In the Clinic Annals of Internal Medicine 2 November 2010
causes of upper abdominal pain that
should be considered in the differen-
tial of acute pancreatitis.
What is the role of imaging
studies in the evaluation of
acute pancreatitis?
Imaging plays an important role in
identify the cause of the attack of
acute pancreatitis and in assessing
severity (26). A plain abdominal ra-
diograph may show nonspecific
signs of acute pancreatitis, such as a
sentinel loop (localized ileus involv-
ing the jejunum), colon cutoff sign
(isolated distention of the trans-
verse colon), duodenal distention
with air and fluid, and pleural effu-
sion localized to the left thorax. In
cases of abdominal distention with
acute pain, the X-ray may reveal
free air showing a perforated viscus
as the cause of pain.
However, the initial imaging study
of choice is ultrasonography of the
right upper quadrant because it is
readily available, noninvasive, inex-
pensive, and relatively sensitive
(95%) for diagnosing gallstone dis-
ease. The presence of gallstones
and/or dilatation of the common bile
duct supports stones as the cause of
acute pancreatitis. However, the dis-
tal common bile duct and pancreas
are frequently obscured by overlying
bowel gas and limit the sensitivity of
ultrasonography for diagnosing gall-
stone-associated pancreatitis.
In this case, a contrast-enhanced,
thin-sliced, triple-phase computed
tomography (CT) scan provides an
Table 2. Differential Diagnosis of Acute Pancreatitis
Disease Characteristics Findings
Perforated viscus, especially peptic ulcer Sudden onset of pain that increases over 30-60 min Intraperitoneal air present
Acute cholecystitis and biliary colic Epigastric or right upper quadrant pain that radiates Liver enzymes often elevated; ultrasonography
to right shoulder or shoulder blade may show thickened gallbladder, pericholecystic
fluid
Intestinal obstruction Constant colicky pain Obstructive pattern can be seen on CT scan or
abdominal series
Mesenteric vascular occlusion Classic triad is postprandial abdominal pain, Discrepancy between symptoms (severe pain) and
weight loss, and abdominal bruit examination (benign abdominal examination)
Dissecting aortic aneurysm Sudden onset; pain may radiate to the lower
extremities
Renal colic Flank pain radiates to the genitals; dysuria may Urinalysis with active sediment
be present
Myocardial infarction Upper abdominal or chest pain Electrocardiography usually abnormal
Connective tissue disorders with vasculitis Acute pancreatitis can be due to vasculitis Other signs of vasculitis usually present (skin,
joint, eye, and kidney involvement)
Appendicitis Pain may start in epigastrium or periumbilical then Ultrasonography and and CT aid in diagnosis
migrate to right lower quadrant
Ectopic pregnancy Sudden onset of pain; menstrual abnormalities Rapid drop in hematocrit and intraperitoneal
often precede pain pelvic fluid on imaging should raise suspicion
Pneumonia Fever, malaise, and other respiratory symptoms Changes on physical examination of the chest
(dyspnea, cough, sputum production, chest pain) and abnormalities on chest X-ray possibly due to
usually present ARDS or pleural effusion
AP = acute pancreatitis; ARDS = acute respiratory distress syndrome; CT = computed tomography; HCT = hematocrit.
27. Lankisch PG, Struck-
mann K, Assmus C,
Lehnick D, Maison-
neuve P, Lowenfels
AB. Do we need a
computed tomogra-
phy examination in
all patients with
acute pancreatitis
within 72 h after ad-
mission to hospital
for the detection of
pancreatic necrosis?
Scand J Gastroen-
terol. 2001;36:432-6.
[PMID:11336171]
28. Arvanitakis M, Del-
haye M, De Maerte-
laere V, et al. Com-
puted tomography
and magnetic reso-
nance imaging in
the assessment of
acute pancreatitis.
Gastroenterology.
2004;126:715-23.
[PMID:14988825]
29. Makary MA, Duncan
MD, Harmon JW, et
al. The role of mag-
netic resonance
cholangiography in
the management of
patients with gall-
stone pancreatitis.
Ann Surg.
2005;241:119-24.
[PMID:15621999]
30. Liu CL, Lo CM, Chan
JK, et al. Detection of
choledocholithiasis
by EUS in acute pan-
creatitis: a prospec-
tive evaluation in
100 consecutive pa-
tients. Gastrointest
Endosc. 2001;54:325-
30. [PMID:11522972]
31. Lai R, Freeman ML,
Cass OW, Mallery S.
Accurate diagnosis
of pancreas divisum
by linear-array endo-
scopic ultrasonogra-
phy. Endoscopy.
2004;36:705-9.
[PMID:15280976]
2010 American College of Physicians ITC5-8 In the Clinic Annals of Internal Medicine 2 November 2010
excellent image of the pancreas and
can identify choledocholithiasis or
other causes of abdominal pain. CT
scanning can also be useful to assess
the severity of the pancreatitis and in
identifying complications, such as
necrosis (infected or not), pseudocyst
formation, and diffuse pancreatic flu-
id collection (27). However, early in
the course of disease a CT scan may
not show signs of pancreatitis or its
associated complications. In addi-
tion, intravenous contrast may accel-
erate renal injury. When these fac-
tors are a concern, magnetic
resonance imaging (MRI) offers an
alternative at greater cost to diagnose
and evaluate the severity of acute
pancreatitis (28).
Among newer but even more costly
imaging modalities, the noninvasive
magnetic resonance cholangiopan-
creatography (MRCP) has high
sensitivity (>90%) for choledo-
cholithiasis and can identify other
anatomical abnormalities, such as
pancreas divisum, pancreatic duct
abnormalities, and mucinous neo-
plasm in the pancreas (29). It can
be useful to exclude the presence of
a retained stone or debris if there is
a high index of clinical suspicion.
Secretin-enhanced MRI is useful
for evaluating pancreatic function
and anatomy when the patient is
suspected of having underlying
chronic pancreatitis. However, since
secretin stimulates pancreatic secre-
tion, it should not be obtained dur-
ing an acute episode because it
could worsen the disease.
Endoscopic ultrasonography is both
sensitive and specific in identifying
small (e.g., 5 mm) gallstones in
the bile ducts (30) and can identify
anatomical abnormalities of the
pancreas. Although it is more inva-
sive than MRI, it can detect smaller
stones and can be used when MRI
is not possible (e.g., in critically ill
patients or when it is contraindicat-
ed, such as in patients with a car-
diac pacer) (31, 32).
Which factors help to predict the
prognosis of a patient with acute
pancreatitis?
Patients should be stratified by risk
for severe morbidity and death relat-
ed to acute pancreatitis because of
the diseases protean manifestations,
unpredictable course, and the need
to identify persons who require in-
tensive care. The Atlanta Classifica-
tion of Acute Pancreatitis was
developed in 1992 to standardize
what was a heterogeneous set of cri-
teria to diagnose the disease and to
assess severity (1). However, because
of a changing understanding of the
pathophysiology and epidemiology
of acute pancreatitis, in 2008 a revi-
sion was proposed to the Atlanta
Classification (still being reviewed
with final approval expected by
2011) that recognizes 2 phases of the
disease that were not appreciated by
the original classification (see the
Box) (33). First, there is a peak in
mortality usually within the first
week of onset and another 2 to 6
weeks after onset. In the first week,
the severity of the disease is usually
reflected by the extent of organ fail-
ure. After that, mortality can be pre-
dicted more by the presence of
pancreatic necrosis and infection.
Therefore, when a patient first
presents, clinicians need to be alert
to the possibility of organ failure
(34). As expected, progression from
single to multiorgan failure is a pre-
dictor of increased mortality (35).
Coagulopathy bodes poorly for pa-
tients as indicated by platelet count
<100 000/mm
3
, fibrinogen <100
mg/dL, and fibrin split products
>80 g/mL. Similarly, low serum
calcium levels ( 7.5 mg/dL) carry
a poor prognosis.
The Atlanta symposium also
identified the development of lo-
cal complications (necrosis and
abscess and pseudocyst formation)
as indicative of severe acute pan-
creatitis. Pancreatic necrosis is
demonstrated by poor perfusion
and nonenhancement on CT scan
Atlanta Classification of Acute
Pancreatitis*
Severe acute pancreatitis
Organ failure (systolic blood pressure
<90 mm Hg, PaO
2
<60 mm Hg, creatinine level
>2 mg/dL, gastrointestinal bleeding
> 500 mL/24 h)
Local complications (pancreatic
necrosis, pseudocyst, or abscess)
3 Ranson criteria.
Mild acute pancreatitis
Minimal organ dysfunction
Uneventful recovery
Lacks features of severe acute
pancreatitis.
Notes: Consider determining APACHE II
score and measuring C-reactive protein
levels. Be aware of limited accuracy of
severity prediction.
*From reference 33.
2010 American College of Physicians ITC5-9 In the Clinic Annals of Internal Medicine 2 November 2010
inflammation. When both organ
failure and infected pancreatic
necrosis are present, relative risk
for mortality doubles (45).
A variety of other classifications
have been developed to assess the
severity of acute pancreatitis early
in the course of disease (Table 3),
including Ranson criteria; the
Acute Physiology and Chronic
Health Evaluation (APACHE-II
of more than 3 cm or >30% of the
pancreas (but these dimensions
are being debated) (39). A
pseudocyst is a fluid collection
within the pancreas, separated by
a nonepithelized wall, that devel-
ops over a period of weeks (by
definition >4 wk). Infection of
either pancreatic necrosis or a
pseudocyst can lead to abscess for-
mation. Pancreatic fluid can also
extravasate as a result of the
Table 3. Clinical Criteria for Determining Severity of Acute Pancreatitis
Classification Predictors Outcomes Based on Score Comments
Ranson criteria Admission measurements Mortality: 0%3% for <3 criteria, Scoring on admission and at 48 h after
(1 point each): Age >55 years; 11%15% for 35 criteria, presentation; limited predictive power reported in
leukocyte count >16 000/mm
3
; and 40% for 6 criteria (36) meta-analysis (37)
glucose >200 mg/dL; LDH
>350 U/L; AST >250 U/L; fluid
sequestration >6 L
Measurement at 48 h (1 point
each): HCT decrease of 10
volume %; BUN increase of
5 mg/dL; calcium <8 mg/dL;
PaO
2
<60 mm Hg; base deficit
>4 mEq/L
Acute Physiology and Chronic Daily: Based on diverse variables, Mortality: <4% for a score <8, Requires data usually only available when
Health Evaluation including age, physiology, and 11%18% for a score 8 (39) patient is in ICU; an increasing APACHE-II
(APACHE) II scoring system long-term health; equation score in the first few days of hospitalization
available at www.sfar.org/ indicates worsening severity whereas the oppo-
scores2/apache22.html#calcul; site indicates improvement (38); APACHE-II
Adding BMI to APACHE-II and APACHE-III have a similar performance
(the APACHE-O score)
increases discrimination (1 point
added for BMI 26-30; 2 points
for BMI >30) (38)
Modified Glasgow prognostic At 48 h after admission (1 point Severe episode: score 3 within Takes 48 h to complete; similar performance to
criteria (Imrie scoring system) each): PaO
2
< 60 mm Hg/7.9 kPa; 48 h APACHE-III (40)
age >55 y; neutrophils (WBC
>15); calcium <2 mmoL/L; renal
function: urea >16 mmoL/L;
enzymes LDH >600 IU/L, AST
>200 IU/L; albumin <32 g/L
(serum); blood glucose level
>10 mmol/L
Bedside Index for Severity in Within 24 h after presentation Mortality: <1% in the lowest Assessed at 24 h; prognostic accuracy similar to
Acute Pancreatitis (BISAP) (1 point each): BUN >25 mg/dL; risk group and >20% in the other scoring systems (42)
score impaired mental status; systemic highest risk group.
inflammatory response syndrome
(see text for definition); age
>60 y; presence of pleural
effusion (41)
Modified CT severity index CT scan assessment of Correlated with length of stay Studied in small patient populations
pancreatic inflammation and and clinical complications (44)
necrosis, plus assessment of
extrapancreatic complications (43)
AST = aspartate transaminase; BMI = body mass index; BUN = blood urea nitrogen; CT = computed tomography; HCT = hematocrit; ICU = intensive care unit;
LDH = lactate dehydrogenase; WBC = white blood cells.
32. Sedlack R, Affi A,
Vazquez-Sequeiros E,
Norton ID, Clain JE,
Wiersema MJ. Utility
of EUS in the evalua-
tion of cystic pancre-
atic lesions. Gas-
trointest Endosc.
2002;56:543-7.
[PMID:12297771]
33. Acute Pancreatitis
Classification Work-
ing Group. Revision
of the Atlanta classi-
fication of acute
pancreatitis (3rd re-
vision). www.pan-
creasclub.com/re-
sources/AtlantaClass
ification.pdf. Ac-
cessed 16 Septem-
ber 2010.
34. Johnson CD, Abu-
Hilal M. Persistent or-
gan failure during
the first week as a
marker of fatal out-
come in acute pan-
creatitis. Gut.
2004;53:1340-4.
[PMID:15306596]
35. Brown A, Orav J,
Banks PA. Hemocon-
centration is an early
marker for organ fail-
ure and necrotizing
pancreatitis. Pan-
creas. 2000;20:367-
72. [PMID:10824690]
36. Blum T, Maison-
neuve P, Lowenfels
AB, Lankisch PG. Fa-
tal outcome in acute
pancreatitis: its oc-
currence and early
prediction. Pancre-
atology. 2001;1:237-
41. [PMID:12120201]
37. De Bernardinis M, Vi-
oli V, Roncoroni L,
Boselli AS, Giunta A,
Peracchia A. Discrim-
inant power and in-
formation content of
Ransons prognostic
signs in acute pan-
creatitis: a meta-ana-
lytic study. Crit Care
Med. 1999;27:2272-
83. [PMID:10548220]
38. Banks PA, Freeman
ML. Practice Parame-
ters Committee of
the American Col-
lege of Gastroen-
terology. Practice
guidelines in acute
pancreatitis. Am J
Gastroenterol.
2006;101:2379-400.
[PMID:17032204]
39. Johnson CD, Toh SK,
Campbell MJ. Com-
bination of APACHE-
II score and an obe-
sity score
(APACHE-O) for the
prediction of severe
acute pancreatitis.
Pancreatology.
2004;4:1-6.
[PMID:14988652]
2010 American College of Physicians ITC5-10 In the Clinic Annals of Internal Medicine 2 November 2010
and III) scale; the Modified Glas-
gow prognostic criteria (also known
as the Imrie scoring system); Bed-
side Index for Severity in Acute
Pancreatitis (BISAP) score; and the
Modified CT Severity Index.
It is important to note that neither
serum amylase nor lipase levels are
predictive of the severity of acute
pancreatitis. On the other hand,
C-reactive protein has been widely
used to predict the severity of acute
pancreatitis (18), and in critically ill
patients, it has been shown to be
associated with increased risk for
organ failure and death (46). Pro-
calcitonin has been associated with
pancreatic infection and can be
used as an indicator of the need for
fine-needle aspiration of pancreatic
necrosis (23).
What are the indications for
hospitalization and for intensive
care for a patient with acute
pancreatitis?
Patients with acute pancreatitis
should be hospitalized until they have
been observed for a sufficient period
to evaluate disease severity and pro-
gression. Essential management
includes aggressive intravenous fluid
resuscitation with no fluids or solids
by mouth. Patients often require pain
management with intravenous med-
ications, typically opiates are used and
must be monitored for side effects,
such as respiratory depression. Stable
patients having a mild episode who
have a history of multiple episodes
can sometimes be managed on an
outpatient basis.
Such tests as ERCP are usually done
in an inpatient setting when indicat-
ed. Severe acute pancreatitis requires
close inpatient monitoring, and the
patient should be transferred to ICU
if organ failure develops (47). In eld-
erly patients with underlying cardio-
vascular disease, aggressive fluid
resuscitation should be administered
in an ICU and may require a central
venous catheter for more accurate
fluid monitoring. Transfer to a spe-
cialized monitored unit, although
not necessarily an ICU, should be
considered for patients with high
body mass index (>30), decreased
urine output < 50 mL/h, tachycardia
(pulse rate > 120 beats/min), signs of
encephalopathy, and/or need for ad-
ditional narcotics (39).
Diagnosis... In acute pancreatitis, diagnosis is based on the presence at least 2 of 3
features: abdominal pain; increased pancreatic enzyme, amylase, and/or lipase levels
to 3 times the upper limit of normal; and imaging tests showing characteristic
findings of acute pancreatitis. Ultrasonography of the right upper quadrant may re-
veal stones or biliary duct dilatation and CT scan can be useful to assess for pancre-
atic edema, necrosis, or pseudocyst formation. MRI offers an alternative but is more
costly. Assessing the severity of the attack of acute pancreatitis using clinical labora-
tory parameters; imaging; and standard measurements, such as APACHE-II, BISAP, CT
severity index, or Ranson criteria, can guide management. Acute pancreatitis should
be managed in the inpatient setting with rare exceptions and patients with organ
failure or severe comorbid conditions should be treated in the ICU.
CLINICAL BOTTOM LINE
Treatment
can experience a significant loss of
intravascular volume due to third
spacing and increased permeability
from release of inflammatory media-
tors. Compromised intravascular vol-
ume can lead to decreased perfusion
How should clinicians manage
fluids in a patient with acute
pancreatitis?
Fluid resuscitation is a critical com-
ponent of management of patients
with acute pancreatitis because they
40. Chatzicostas C,
Roussomoustakaki
M, Vlachonikolis IG,
et al. Comparison of
Ranson, APACHE II
and APACHE III scor-
ing systems in acute
pancreatitis. Pan-
creas. 2002;25:331-5.
[PMID:12409825].
41. Blamey SL, Imrie CW,
ONeill J, Gilmour
WH, Carter DC. Prog-
nostic factors in
acute pancreatitis.
Gut. 1984;25:1340-46.
42. Wu BU, Johannes RS,
Sun X, et al. The ear-
ly prediction of mor-
tality in acute pan-
creatitis: a large
population-based
study. Gut.
2008;57:1698-703.
[PMID:18519429] 43.
Papachristou GI,
Muddana V, Yadav D,
et al. Comparison of
BISAP, Ransons,
APACHE-II, and CTSI
scores in predicting
organ failure, com-
plications, and mor-
tality in acute pan-
creatitis. Am J
Gastroenterol.
2010;105:435-41;
quiz 442.
[PMID:19861954]
44. Mortele KJ, Wiesner
W, Intriere L, et al. A
modified CT severity
index for evaluating
acute pancreatitis:
improved correla-
tion with patient
outcome. AJR Am J
Roentgenol.
2004;183:1261-5.
[PMID:15505289]
45. Petrov MS,
Shanbhag S,
Chakraborty M, et al.
Organ failure and in-
fection of pancreatic
necrosis as determi-
nants of mortality in
patients with acute
pancreatitis. Gas-
troenterology. 2010
Jun 9. [Epub ahead
of print]
[PMID:20540942]
46. Lobo SM, Lobo FR,
Bota DP, et al. C-re-
active protein levels
correlate with mor-
tality and organ fail-
ure in critically ill pa-
tients. Chest.
2003;123:2043-9.
[PMID:12796187]
47. Zhang XP, Wang L,
Zhou YF. The patho-
genic mechanism of
severe acute pancre-
atitis complicated
with renal injury: a
review of current
knowledge. Dig Dis
Sci. 2008;53:297-306.
[PMID:17597411]
48. Frossard J-L, Steer
ML, Pastor CM.
Acute pancreatitis.
Lancet.
2008;371:143-52.
[PMID:18191686]
2010 American College of Physicians ITC5-11 In the Clinic Annals of Internal Medicine 2 November 2010
However, studies comparing naso-
gastric with nasojejunal feeding
have not shown significant differ-
ences in outcomes, but larger com-
parative studies are required before
practice recommendations are
changed.
A meta-analysis of 6 studies showed a
lower incidence of infections, reduced
surgical intervention and shorter hospital
stay in patients with acute pancreatitis
receiving nasojejunal feeding (50). In one
study, nasojejunal feeding was associat-
ed with a shorter hospital stay and fewer
complications than parental nutrition
(sepsis, 4% vs. 33%; metabolic complica-
tions, 15% vs. 52%, respectively), and a
savings of $2362 (51).
An ongoing National Institutes of
Health multi-center trial, called the
Study of Nutrition in Acute Pan-
creatitis (SNAP), is evaluating na-
sogastric vs. nasojejunal feeding.
Difficulty placing or maintaining a
nasojejunal tube also requires par-
enteral nutrition. Notably, parenter-
al nutrition may be required in
some critically ill patients as well as
those with severe ileus.
What other supportive care may be
beneficial for acute pancreatitis?
Oxygen may reduce the acute res-
piratory distress syndrome that
can occur in the early stages of
acute pancreatitis. Pain manage-
ment is another key aspect of
treatment. Due to the severity of
pain with acute pancreatitis and
the inability to take pills, par-
enteral narcotic analgesics are es-
sential. Opiates are usually admin-
istered every 2 to 4 hours. A
patient-controlled analgesia pump
offers an alternative when boluses
of pain medications provide inad-
equate pain control. Morphine has
been theoretically implicated in
increasing pressure in the sphinc-
ter of Oddi and potentially de-
creasing pancreatic and biliary
flow into the small bowel lumen,
but this has not been confirmed in
clinical studies. Meperidine, mor-
phine and hydromorphone are
of the pancreas and such complica-
tions as pancreatic necrosis and renal
failure. Fluid administration should
be guided by vital signs, urine out-
put, and change in hematocrit at ad-
mission, 12 hours, and 24 hours
(39). Increasing hematocrit or BUN
is a poor prognostic sign and indi-
cates worsening severity.
How should clinicians manage the
nutritional needs of a patient
with acute pancreatitis?
In mild acute pancreatitis, nutri-
tional support is not necessary.
Once pain has diminished along
with nausea and vomiting, oral nu-
trition can be started. It begins with
clear liquids and clinical monitoring
for change in pain and symptoms of
nausea and vomiting. Resolution of
imaging findings and normalization
of amylase and lipase may not occur
for up to a week, so the diet should
be advanced based on how the pa-
tient feels. There is no clear consen-
sus about fat restriction.
Patients with moderate or severe
pancreatitis must usually abstain
from solids and liquids for several
days to weeks. Although mortality
rates do not differ substantially be-
tween parenteral and enteral nutri-
tion, the latter has been shown to
reduce the rate of infection, surgical
interventions, and noninfectious
complications (48). United King-
dom guidelines for management of
acute pancreatitis recommend en-
teral nutrition for all patients with
severe acute pancreatitis, but state
that the nasogastric route is pre-
ferred for feeding because it is ef-
fective in 80% of cases (49).
However, the nasojejunal route is
increasingly being used in the
United States. Although tube
placement is more difficult, enteral
feeding beyond the ligament of
Treitz may decrease the risk for in-
fectious complications that can oc-
cur with feeding methods in which
the small bowel can be affected by
edema and permeability from in-
flammatory mediators is increased.
49. UK Working Party on
Acute Pancreatitis.
UK guidelines for the
management of
acute pancreatitis.
Gut 2005;54(Suppl
III):iii1-iii9.
50. Louie BE, Noseworthy
T, Hailey D, Gramlich
LM, Jacobs P,
Warnock GL. 2004
MacLean-Mueller
prize enteral or par-
enteral nutrition for
severe pancreatitis: a
randomized con-
trolled trial and
health technology as-
sessment. Can J Surg.
2005;48:298-306.
[PMID:16149365]
51.Abou-Assi S, Craig K,
OKeefe SJ.
Hypocaloric jejunal
feeding is better
than total parenteral
nutrition in acute
pancreatitis: results
of a randomized
comparative study.
Am J Gastroenterol.
2002;97:2255-62.
[PMID:12358242]
52. Villatoro E, Mulla M,
Larvin M. Antibiotic
therapy for prophy-
laxis against infec-
tion of pancreatic
necrosis in acute
pancreatitis.
Cochrane Database
Syst Rev. 2010 May
12;5:CD002941.
[PMID:20464721]
53. Carter CR, McKay CJ,
Imrie CW. Percuta-
neous necrosectomy
and sinus tract en-
doscopy in the man-
agement of infected
pancreatic necrosis:
an initial experience.
Ann Surg.
2000;232:175-80.
[PMID:10903593]
54. Connor S, Ghaneh P,
Raraty M, Sutton R,
Rosso E, Garvey CJ,
et al. Minimally inva-
sive retroperitoneal
pancreatic necrosec-
tomy. Dig Surg.
2003;20:270-7.
[PMID:12748429].
55. van Santvoort HC,
Besselink MG, Bakker
OJ, et al. A step-up
approach or open
necrosectomy for
necrotizing pancre-
atitis. N Engl J Med.
2010;362:1491-502.
[PMID:20410514]
2010 American College of Physicians ITC5-12 In the Clinic Annals of Internal Medicine 2 November 2010
more commonly used narcotics for
pain control in acute pancreatitis.
What is the role of antibiotics in
the management of patients with
acute pancreatitis?
Antibiotics are not currently rec-
ommended for mild interstitial
pancreatitis or even for moderate to
severe pancreatitis with sterile
necrosis. Studies of prophylactic
administration of antibiotics to de-
crease infectious complications
have been largely nonsupportive.
A recent Cochrane review found no benefit
of antibiotics to prevent infection of pan-
creatic necrosis or mortality, with the pos-
sible exception of the -lactam imipenem,
that was associated with a significant de-
crease in pancreatic infection (52). The re-
viewers concluded that better-designed
studies would be required before antibiotic
prophylaxis could be recommended.
However, antibiotics are definitely re-
quired to treat ascending cholangitis,
infected pancreatic necrosis, or an in-
fected pseudocyst. When the patient
seems to be septic or infection is sus-
pected, a fever workup should be con-
ducted with cultures and a chest
X-ray. If needed, CT-guided needle
aspiration of a necrotic area of the
pancreas should be cultured for bacte-
ria and fungi. If the workup is nega-
tive, continue antibiotics if the patient
has septicemia, organ failure, or 30%
necrosis of the pancreas (5).
For an infected necrotic pancreas,
the choice of antibiotic is guided by
the culture. For gram-negative or-
ganisms, options include imipenem,
meropenem, ofloxacin, or
ciprofloxacin with metronidazole, or
a third-generation cephalosporin
with metronidazole. Patients with
infected pancreatic necrosis should
be closely observed to assess for re-
sponse and surgical debridement
should be considered when the pa-
tient does not improvemortality is
high if this disorder is not treated
aggressively (49).
There are multiple approaches for
debridement but no consensus on
which one is best. Open surgical
debridement has been a standard,
but debridement with a percuta-
neous nephroscope offers an alter-
native (53, 54).
A recent multicenter study randomly as-
signed 88 patients with necrotizing pan-
creatitis and suspected or confirmed in-
fected necrotic tissue to primary open
necrosectomy or a step-up treatment ap-
proach (percutaneous drainage followed
by minimally invasive retroperitoneal
necrosectomy if needed) (55). Major com-
plications or death occurred in 69% of pa-
tients in the open necrosectomy group vs.
40% in the step-up treatment group.
Case reports have also described
endoscopic transgastric endoscopic
debridement of an infected area of
necrosis in selected patients who
are poor surgical candidates (56).
This approach should be consid-
ered in advanced centers with ex-
pertise in these techniques.
When should clinicians consider
consultation with a gastro-
enterologist, a surgeon, or an
interventional radiologist?
For patients who have mild acute
pancreatitis with a known cause,
consultation is usually unneces-
sary. However, if the cause is un-
clear or pancreatitis tends to recur,
a gastroenterology consult may be
useful. In patients with more se-
vere attacks, gastroenterology
consultation can assist with man-
agement and monitoring for
complications. Further, when gall-
stone pancreatitis is suspected,
consultation for ERCP may be
necessary, as noted below.
When a patient develops necrotizing
pancreatitis or abscesses or pseudo-
cysts, or pancreatic fluid collection is
necessary, both a surgeon and a gas-
troenterologist should be consulted.
These patients usually require a team
approach because surgical, endoscop-
ic, and percutaneous drainage meth-
ods should be considered. Endoscopic
drainage of pseudocysts has been as-
sociated with better outcomes (57).
Because of limited data on endoscopic
56. Gupta K, Freeman
ML. Disconnected
pancreatic duct with
pancreas necrosis,
treated with trans-
gastric debridement
and pancreatic duct
stent. Clin Gastroen-
terol Hepatol.
2010;8:e51.
[PMID:20005979]
57. Seewald S, Ang TL,
Teng KC, Soehendra
N. EUS-guided
drainage of pancre-
atic pseudocysts, ab-
scesses and infected
necrosis. Dig Endosc.
2009;21 Suppl 1:S61-
5. [PMID:19691738]
58. Maniatis P, Delis S,
Fagrezos D, et al. The
interventional radio-
logical procedures of
the infections of
pancreas. Infect Dis-
ord Drug Targets.
2010;10:5-8.
[PMID:20180752]
59. Petrov MS, van
Santvoort HC,
Besselink MG, et al.
Early endoscopic ret-
rograde cholan-
giopancreatography
versus conservative
management in
acute biliary pancre-
atitis without
cholangitis: a meta-
analysis of random-
ized trials. Ann Surg.
2008;247:250-7.
[PMID:18216529]
60. Loveday BPT, Srini-
vasa S, Vather R, et
al. High quantity and
variable quality of
guidelines for acute
pancreatitis: a sys-
tematic review. Am J
Gastroenterol.
2010;105:1466-76.
61. Nathens AB, Curtis
JR, Beale RJ et al.
Management of the
critically ill patient
with severe acute
pancreatitis. Crit
Care Med.
2004;32:2524-36.
2010 American College of Physicians ITC5-13 In the Clinic Annals of Internal Medicine 2 November 2010
drainage of pancreatic necrosis but
especially when infected, surgical in-
tervention may be required. An inter-
ventional radiology consultation may
be useful for percutaneous CT-guided
catheter drainage of infected pancre-
atic pseudocysts (58). A trial of per-
cutaneous drainage followed by
minimally invasive retroperitoneal
necrosectomy, if necessary, versus sur-
gery with open necrosectomy found
that the minimally invasive approach
had fewer major complications or
death (55). When gallstones are pres-
ent, patients need to be evaluated by a
surgeon for cholecystectomy.
What are the indications for ERCP?
Presence of a retained bile duct
stone seen on imaging is an indi-
cation for ERCP to perform bil-
iary sphincterotomy and stone
removal. Urgent ERCP is indi-
cated if cholangitis is suspected.
In the absence of these criteria,
studies have found that early
ERCP was associated with in-
creased complications.
A meta-analysis of 7 randomized trials
found no significant reduction in overall
complications or mortality from early
ERCP in patients with predicted mild or se-
vere acute biliary pancreatitis without
acute cholangitis (59).
Several studies have shown an ad-
vantage of ERCP in pancreatitis
from obstructive biliary disease. Pa-
tients presenting with complicated
acute pancreatitis due to a disrupted
pancreatic duct with a leak may also
benefit from ERCP and placement
of a pancreatic duct stent.
What is the role of patient
education in the management of
acute pancreatitis?
Patient education plays an impor-
tant role in preventing recurrent
acute pancreatitis. Lifestyle meas-
ures, such as cessation of alcohol
consumption, are critical. Educa-
tion about the risks of certain
medications if implicated in the
initial episode should also be pro-
vided with careful monitoring.
Dietary modification and adher-
ence to lipid-lowering medica-
tions are both necessary in
patients with pancreatitis due to
hypertriglyceridemia.
What do professional
organizations recommend with
regard to the care of patients
with acute pancreatitis?
A recent summary has assessed the
quality of 30 clinical guidelines re-
garding management of acute pan-
creatitis that were published between
1988 and 2008 (60). Among the
more recent U.S. clinical guidelines,
those from the American Thoracic
Society (2004), American College of
Gastroenterology (2006), and the
American Gastroenterological Asso-
ciation (2007) earned high quality
scores (5, 39, 61). The Box summa-
rizes the most recent 2 guidelines.
Treatment... Aggressive fluid resuscitation is the most important approach to manage
acute pancreatitis. Appropriate pain control and supportive care with oxygen supple-
mentation are additional basic measures. In patients with a prolonged course, enteral
nutrition is preferred to parenteral nutrition whenever possible. Antibiotics are only
recommended for a documented infectious process or if there is 30% necrosis. If the
cause of acute pancreatitis is unclear or ERCP reveals retained gallstones, consultation
by a gastroenterologist is indicated. A team approach with both a gastroenterologist
and surgeon is indicated for patients with organized necrosis (sterile or infected),
pseudocyst, or abscess. In some cases, an interventional radiologist should be consult-
ed for specialized diagnostic and therapeutic imaging techniques.
CLINICAL BOTTOM LINE
Practice
Improvement
I
n
t
h
e
C
l
i
n
i
c
Tool Kit
In the Clinic
Acute
Pancreatitis
PIER Module
www.pier.acponline.org
Access the PIER module on acute pancreatitis. PIER modules provide evidence-
based, updated information on current diagnosis and treatment in an electronic
format designed for rapid access at the point of care.
The PIER module on acute pancreatitis includes two tables to help guide diagnosis.
Patient Information
www.annals.org/intheclinic/toolkit-acutepancreatitis.html
Access the Patient Information material that appears on the following page
for duplication and distribution to patients.
http://digestive.niddk.nih.gov/ddiseases/pubs/pancreatitis/
Access information on pancreatitis from the NIDDKs National Digestive
Diseases Information Clearinghouse.
www.nlm.nih.gov/medlineplus/ency/article/000287.htm
www.nlm.nih.gov/medlineplus/spanish/ency/article/000287.htm
Access information on acute pancreatitis in English and Spanish from the
National Library of Medicines Medline Plus.
www.gastro.org/patient-center/digestive-conditions/pancreatitis
Access Understanding Pancreatitis from the American Gastroenterological
Association.
Clinical Guidelines
www.acg.gi.org/physicians/guidelines/AcutePancreatitis.pdf
The American College of Gastroenterology published practice guidelines in
acute pancreatitis in 2006.
www.gastrojournal.org/article/S0016-5085%2807%2900592-6/fulltext
The American Gastroenterological Association Institute published a Medical Position
Statement on the management of acute pancreatitis in 2007.
The PIER module on acute pancreatitis includes two tablesto help guide diagnosis.
Quality Measures
There are currently no Centers for Medicare & Medicaid Services quality
measures for acute pancreatitis.
2 November 2010 Annals of Internal Medicine In the Clinic ITC5-14 2010 American College of Physicians
American College of Gastroenterology Guidelines 2006
Diagnosis of acute pancreatitis requires 2 of the following 3 criteria: abdominal pain, amylase and/or lipase levels 3 times upper limit of
normal, and/or CT scan findings of acute pancreatitis.
Obesity, older age, and organ failure on admission; APACHE II score 8 and/or increasing in first 48 hours; and/or hematocrit 44 suggest
severe acute pancreatitis.
CT scan with intravenous contrast and C-reactive protein > 150 mg/L help to identify necrotizing pancreatitis.
Initial management includes aggressive intravenous hydration and supplemental oxygen. Patients with organ failure require ICU monitoring.
Prolonged illness requires nutritional support. Enteral is preferred to total parenteral nutrition when possible.
Antibiotics are not necessary for necrotizing pancreatitis, even with organ failure. If infection is suspected, CT-guided needle aspiration and
culture are recommended.
MRCP and endoscopic ultrasonography can identify choledocholithiasis. If bile duct stone is confirmed, urgent ERCP is recommended.
For complex necrotizing pancreatitis, pseudocyst, or infected necrosis, interventional options include open or laparoscopic surgery or
percutaneous or endoscopic drainage. Individualize management to the patient and available expertise.
American Gastroenterological Association Guidelines 2007
Clinical presentation, increased serum amylase and lipase levels, and imagingespecially contrast-enhanced CT scanassist in diagnosis of
acute pancreatitis.
Organ failure and local pancreatic complications help to assess severity. Progressive organ failure predicts increased mortality.
ICU monitoring is recommended if severe comorbidity or severe disease is diagnosed on the basis of imaging or APACHE II score 8.
Identify the cause through imaging and laboratory studies, including liver enzyme and triglyceride levels.
Specialized imaging, such as endoscopic ultrasonography, is needed when choledocholithiasis is a concern before proceeding with ERCP.
Reserve ERCP for when less invasive methods are unavailable.
Supportive care should include fluid resuscitation, supplemental oxygen, correction of electrolyte abnormalities, and pain control.
Consider nutritional support with preference for enteral nutrition over total parenteral nutrition.
Reserve antibiotic prophylaxis for patients with > 30% of necrosis. Use CT-guided aspiration to guide antibiotic selection.
Base management of pseudocysts, fluid collections, and necrosis on symptoms and available expertise.
In the Clinic
Annals of Internal Medicine
P
a
t
i
e
n
t

I
n
f
o
r
m
a
t
i
o
n
THINGS YOU SHOULD
KNOW ABOUT ACUTE
PANCREATITIS
What is acute pancreatitis?
The pancreas is a gland that lies behind the stomach
and produces fluid that goes into the small intestine to
break down food.
Acute pancreatitis occurs when something blocks the
flow of this fluid or attacks the tissues of the pancreas.
Severe acute cases can be fatal.
What are symptoms of acute
pancreatitis?
Severe, constant pain in the upper abdomen may
spread to the back.
Nausea and vomiting can occur.
Sweating, fast heart rate, and fever can occur.
Who gets acute pancreatitis?
People with gallstone disease or who use alcohol
heavily are at risk.
Other, less common causes include some medications,
injury to the pancreas, high triglyceride levels (often
checked with cholesterol), and pancreas deformities
from birth.
Men are at higher risk than women.
How is it diagnosed?
Your doctor will ask you about risk factors for acute
pancreatitis, review your medications, and examine
your stomach area as well as check your vital signs.
Your doctor will order blood tests of enzymes from
the pancreas among other tests and do X-ray or
ultrasonography studies.
Common tests used to diagnose acute pancreatitis
include ultrasonography, computed tomography (CT)
scan, and endoscopic retrograde cholangiopancrea-
tography (ERCP), which examines the pancreas
through a tube inserted down the throat into the
stomach and pancreas.
How is it treated?
Hospitalization is necessary for nearly all patients
with acute pancreatitis. Sometimes intensive care is
needed.
While in the hospital and under physician care, you
may need to stop eating for a few days to rest the
pancreas.
Pain medications and intravenous fluids are often
needed.
Treatment may be needed for the underlying cause,
such as for alcohol use or surgery to clear a bile duct
blocked by a gallstone.
It is important to avoid anything that can cause
pancreatitis after an episode, such as alcohol, certain
medications, or foods that increase triglyceride levels
in order to prevent the disease from coming back.
For More Information
http://digestive.niddk.nih.gov/ddiseases/pubs/pancreatitis/
http://digestive.niddk.nih.gov/ddiseases/pubs/ercp/
National Institute of Diabetes and Digestive and Kidney Diseases
information on acute pancreatitis and ERCP.
www.nlm.nih.gov/medlineplus/ency/article/000287.htm
www.nlm.nih.gov/medlineplus/spanish/ency/article/000287.htm
Information on acute pancreatitis in English and in Spanish from
the National Library of Medicines MEDLINE Plus.
4.
3.
CME Questions
2 November 2010 Annals of Internal Medicine In the Clinic ITC5-16 2010 American College of Physicians
A 42-year-old woman is evaluated in the
emergency department for acute onset
of epigastric pain that radiates to the
back and is associated with nausea and
vomiting. The patient had previously
been healthy and has no history of
alcohol or tobacco use. Her only
medication is an oral contraceptive pill.
On physical examination, the
temperature is 37.2 C (99 F), the blood
pressure is 158/90 mm Hg, the pulse rate
is 101/min, and the respiration rate is
20/min. There is no scleral icterus or
jaundice. The abdomen is distended with
mid-epigastric tenderness without
rebound or guarding and with hypoactive
bowel sounds. The results of laboratory
studies are follows: leukocyte count,
13 500/L (13.5 10
9
/L); aspartate
aminotransferase, 131 U/L; alanine
aminotransferase, 567 U/L; bilirubin
(total), 1.1 mg/dL (18.8 mol/L); amylase,
824 U/L; lipase, 1432 U/L.
Radiography of the abdomen shows mild
ileus.
Which of the following is the most
appropriate next step in the evaluation
of this patient?
A. CT scan of the abdomen and pelvis
B. Endoscopic retrograde
cholangiopancreatography
C. Esophagogastroduodenoscopy
D. Ultrasonography of the abdomen
A 34-year-old woman is evaluated for
continued severe mid-epigastric pain
that radiates to the back, nausea, and
vomiting 5 days after being hospitalized
for acute alcohol-related pancreatitis.
She has not been able eat or drink and
has not had a bowel movement since
being admitted.
On physical examination, the
temperature is 38.2 C (100.8 F), the
blood pressure is 132/84 mm Hg, the
pulse rate is 101/min, and the respiration
rate is 20/min. There is no scleral icterus
or jaundice. The abdomen is distended
and diffusely tender with hypoactive
bowel sounds. The results of laboratory
studies are follows: leukocyte count, 15
400/L (15.4 10
9
/L); aspartate
aminotransferase, 189 U/L; alanine
aminotransferase, 151 U/L; bilirubin
(total), 1.1 mg/dL (18.8 mol/L); amylase,
388 U/L; lipase, 924 U/L.
CT scan of the abdomen shows a
diffusely edematous pancreas with
multiple peripancreatic fluid collections,
and no evidence of pancreatic necrosis.
Which of the following is the most
appropriate next step in the
management of this patient?
A. Enteral nutrition by nasojejunal
feeding tube
B. Intravenous imipenem
C. Pancreatic dbridement
D. Parenteral nutrition
A 68-year-old man with a history of
alcoholism is evaluated in the emergency
department for a 7-month history of
diarrhea during which he has noted an
increased volume of stool and decreased
consistency. He has had intermittent
abdominal pain but not severe enough to
prevent him from eating or drinking. He
is not taking any medications.
On physical examination, he is afebrile;
the blood pressure is 108/72 mm Hg, the
pulse rate is 80/min, and the respiration
rate is 16/min. The abdomen is soft with
mild periumbilical tenderness but no
distention. The results of laboratory
studies are follows: aspartate
aminotransferase, 155 U/L; alanine
aminotransferase, 88 U/L; alkaline
phosphatase, 96 U/L; bilirubin (total),
1.1 mg/dL (18.8 mol/L); amylase, 65
U/L; lipase, 70 U/L.
CT scan of the abdomen shows
calcifications but no mass. There is fat in
the stool.
Which of the following is the most
appropriate management for this
patient?
A. Fiber
B. Cholestyramine
C. Loperamide
D. Pancreatic enzymes
A 51-year-old man is evaluated for an
8-month history of mid-epigastric pain
that is worse after eating, six to eight
bowel movements a day usually
occurring after a meal, and loss of 6.8 kg
(15 lb) over the past 6 months. The
patient drinks six to eight cans of beer a
day. He takes no medications.
On physical examination, the patient is
thin (BMI 21) and has normal bowel
sounds, mid-epigastric tenderness, but
no evidence of hepatosplenomegaly or
masses. Rectal examination reveals
brown stool that is occult blood
negative. The remainder of the
examination is normal. Plain radiograph
of the abdomen shows a normal bowel
gas pattern and is otherwise normal. The
results of laboratory studies are follows:
leukocyte count, 6800/L (6.8 10
9
/L);
platelet count, 69 000/L (69 10
9
/L);
fasting plasma glucose, 104 mg/dL
(5.77 mmol/L); aspartate amino -
transferase, 191 U/L; alanine amino -
transferase, 82 U/L; amylase, 122 U/L;
lipase, 289 U/L.
Which of the following tests is most
likely to establish the diagnosis in this
patient?
A. Colonoscopy
B. CT scan of the abdomen
C. Measurement of serum
antiendomysial antibodies
D. Stool for leukocytes, culture, ova,
and parasites
1.
2.
Questions are largely from the ACPs Medical Knowledge Self-Assessment Program (MKSAP, accessed at
http://www.acponline.org/products_services/mksap/15/?pr31). Go to www.annals.org/intheclinic/
to complete the quiz and earn up to 1.5 CME credits, or to purchase the complete MKSAP program.