Acute Bloody Diarrhea: A Medical Emergency for Patients of All Ages
Lori R. Holtz*Marguerite A. Neill‡
Phillip I. Tarr*
*Division of Gastroenterology and Nutrition, Department of Pediatrics, Washington University School of Medicine, St Louis, Missouri;‡Division of Infectious Diseases,
Department of Medicine, Warren Alpert Medical School, Brown University, Providence, Rhode Island
Acute bloody diarrhea should be considered a medical
emergency. Its causes are frequently serious or action-
able or both and are usually identified. However, acute
bloody diarrhea as a stand-alone clinical presentation
has received little scholarly attention in the past several
decades. Although the range of possible causes of acute
bloody diarrhea is broad, infectious considerations are
paramount and should always be prioritized in the eval-
uation of such patients. History, examination, and lab-
oratory testing should be focused on minimizing time
priate therapy). Strategically chosen tests and imaging,
avoidance of extraneous diagnostic pursuits, and provi-
sion of supportive care while awaiting diagnostic clarity
are central to the adroit management of patients with
acute bloody diarrhea. Diagnostic considerations differ
somewhat between adults and children but have many
elements in common, including the need for vigilance
in detecting Escherichia coli O157:H7 infection. In this
review, we discuss diagnostic approaches (emphasizing
the importance of rapid, accurate, and thorough micro-
biologic investigation) and measures that can be taken
to support patients while awaiting information that
determines the cause of their disease. These topics are
discussed in the context of the medical care that is
available to children and adults with bloody diarrhea in
most institutions in developed nations.
ening and have urgent epidemic control implications in
the community. Even if bloody diarrhea is not infectious
in origin, it could represent illnesses that warrant expe-
ditious diagnosis and treatment. The physician faces
many issues in managing patients with bloody diarrhea
while he or she awaits the results of microbiologic anal-
yses and clinical resolution. Uncovering the cause of this
disorder can be a challenge, because there are many
cute bloody diarrhea is a medical emergency, be-
cause it often signifies disorders that are life threat-
enteric pathogens that cause acute bloody diarrhea and
several noninfectious gastrointestinal illnesses that cause
loose, bloody stools, often interpreted as diarrhea.
Patients with acute bloody diarrhea present diagnostic,
management, and infection control challenges for care-
givers. There is not much literature to guide evaluation of
patients with acute bloody diarrhea. Most patients are
initially seen in primary care settings or emergency de-
partments and subspecialists are rarely involved early in
the course of diagnosis and treatment. In searching articles
published since 2003 using the terms “bloody diarrhea”
and “review,” only 12 published papers1–12and 1 govern-
ment document from Scotland (http://www.documents.
hps.scot.nhs.uk/about-hps/hpn/vtec.pdf) even minimally
begin to address this topic. Subspecialty training rarely
emphasizes management of enteric illnesses, which are
customarily attended to by primary care practitioners,
but gastroenterologists and infectious diseases physi-
cians are often considered to be the local authorities on
In view of the sparse literature available to guide physi-
cians in their management of patients with acute bloody
diarrhea, we present the series of events likely to be encoun-
tered at various stages of illness, based on the assumptions
that health care providers have access to good enteric mi-
crobiology laboratories, can perform routine blood tests
infectious, community-acquired acute bloody diarrhea in
the developed world are caused by Campylobacter, Escherichia
coli O157:H7, and other Shiga toxin-producing E coli
(STEC), Salmonella, Shigella, and Yersinia. E coli O157:H7 is
the most important pathogen to consider in evaluating
patients of any age because of its serious nature (its ability
Abbreviations used in this paper: CT, computerized tomography;
HUS, hemolytic uremic syndrome; IBD, inflammatory bowel disease;
STEC, Shiga toxin-producing Escherichia coli.
© 2009 by the AGA Institute
to cause the hemolytic uremic syndrome [HUS]) and be-
cause adroit prediagnostic management can greatly im-
prove care. E coli O157:H7 infection can be modeled after a
myocardial infarction—the acute event that parallels angina
is the bloody diarrhea, and the subsequent extraintestinal
consequences are largely thrombotic in origin.13–19
The elements discernible at presentation that increase
the likelihood that a child or adult is infected with E coli
O157:H7 are presented in Table 1.
Bloody Diarrhea in Children
Acute bloody diarrhea is a rare event in previously
healthy children. Practitioners in some venues, such as
emergency departments (particularly pediatric emergency
departments), have reasonable expertise in handling these
familiar with the assessment of such patients. It is impor-
tant to create a system to record the history, physical exam-
ination results, and microbiologic evaluation data from
patients with acute bloody diarrhea, so that patients are
evaluated economically, and diagnostic clarity is attained
rapidly. The initial contact with the medical system is
often the best opportunity to optimize management of
patients with bloody diarrhea. Collection of good data
during early stages of illness reduces the need to fill in
gaps at later stages. (Was the stool sent for a good
bacterial work-up on initial presentation? What was the
initial complete blood count?) Rare events require good
systems and vigilance for detection and initial manage-
ment, especially when one considers the high frequency
of episodes of nonbloody, considerably less consequen-
Emergency department experiences have taught
us the elements of patient history that are important to
detect at the initial presentation (Table 2). Some indica-
tion as to whether bloody diarrhea is likely to be bacterial
in origin can be gleaned simply by establishing a timeline
of the illness and its characteristics during that interval.
We find some lines of inquiry to be irrelevant and even
misleading. We do not find it helpful to know if a patient
with acute bloody diarrhea (?1 week duration) has a
family history of inflammatory bowel disease (IBD). Al-
though a family history of IBD does increase a person’s
risk of this disorder, and there is often family concern
Table 1. Elements at Presentation That Suggest a Patient
Is Infected With E coli O157:H7
● Nonbloody diarrhea that becomes bloody after 1–3 days.
● No fever at presentation to medical care.
● Tender abdomen.
● More than 5 stools in the past 24 hours.
● Pain is worse on defecation.
● No, few, or moderate fecal leukocytes (but fecal leukocytes, in
our opinions, have little relevance in this situation).
● Diarrhea, and especially bloody diarrhea, persists during first 8
hours in hospital.
● There is no relative bandemia in the differential white cell count.
Table 2. Key Elements of History That Can Help Differentiate Bacterial Versus Nonbacterial Causes of Bloody Diarrhea in
When was the first loose stool?Diarrhea of ?10 days’ duration is more likely to be infectious in origin than is diarrhea that has
Diarrhea that is initially nonbloody but then becomes bloody within 5 days of onset is reasonably
likely to result from a diagnosable bacterial infection. Diarrhea that is bloody from the first
loose stool, or that becomes bloody after ?1 week of nonbloody diarrhea, is much less likely
to be infectious in origin. Caveat: patients might not know the exact timing of the appearance
of the blood in the stools.
Painless bloody diarrhea is unlikely to be infectious in origin.
This seems especially true in E coli O157:H7 infections, but all bacterial colitis can cause
Rectal prolapse is seen most frequently in shigellosis and E coli O157:H7 infections.
Fever is rare at time of initial evaluation in a medical setting in patients infected with E coli
O157:H7 (although approximately one-half of all patients infected with this pathogen will
report a fever before evaluation).
The more prominent and persistent the vomiting, the less likely that the bloody diarrhea has a
More than 5 stools in the 24 hours before evaluation are consistent with bacterial colitis.
When was blood first noted?
Is there abdominal pain?
Is the pain worse at the time of
Is there rectal prolapse?
Is there fever?
Is there vomiting?
How many bowel movements have
there been in the past 24 hours?
If the age is appropriate, diagnosis of intussusception should be considered.
If contacts have bloody diarrhea or a culture-confirmed pathogen, there is an increased likelihood
that the bloody diarrhea is bacterial in origin. However, the absence of a history of ill contacts
does not exclude an infectious cause for the bloody diarrhea.
This should always be noted, but it is difficult to factor answers to this question into a useful
Recent antibiotic usage?
1888 HOLTZ ET AL GASTROENTEROLOGY Vol. 136, No. 6
that bloody diarrhea represents new-onset IBD, patients
with acute bloody diarrhea should first be tested for
infectious colitis. We do not consider it helpful for the
physician to pursue food consumption or exposure his-
tories, because the number of potential infecting vehicles
is large, such exposures are common, sporadic bacterial
enteric infections can be associated with many novel
vehicles,24and failure to cite an exposure in no way alters
the clinical evaluation. However, it is immediately impor-
tant to report any evidence that there is a cluster of
infected individuals to local disease investigation author-
ities, even before a positive culture result is available. In
addition, when a provider learns of a positive culture
result, in most states it is the responsibility of the noti-
fied physician to report the case to the local health
department. Notifications should be made promptly, be-
cause the infecting vehicle might still be in commerce;
source investigations are aided by careful and thorough
food histories taken by professional interviewers as early
as possible. We favor immediate notification of the
health department (even on weekends and holidays), even
though longer time intervals are often permissible by law.
It is important also to get the isolate to the appropriate
disease control authorities for typing.25
Patients with infectious colitis can appear to be well
when observed between spasms of abdominal pain. Such
appearances can engender false assurance that the bloody
diarrhea is not consequential; there is a natural tendency
to consider bloody diarrhea to be merely a variant of
self-limited and considerably less consequential gastro-
enteritis. However, if parents report that their child is in
severe pain, or if the family is anxious about the severity
of the child’s illness, it is important to give such concerns
credence, even if the examiner does not observe discom-
fort during the few minutes of assessment. Such infor-
mation should be used in deciding whether to hospitalize
a patient who is probably not infected with E coli
O157:H7 or to send the child home, despite a low (but
not negligible) risk of HUS or death.
A challenge of the physical examination is to iden-
tify relevant information and to avoid misleading find-
ings. As with all patients with diarrhea, the examiner
should try to address the adequacy of circulating blood
volume (heart rate, blood pressure, and postural changes);
patients with acute bloody diarrhea rarely present in or
near shock. A high fever (?39°C) prompts consideration
of Shigella infection, whereas the absence of fever is typical
of E coli O157:H7 infections.26,27
It is also useful to attempt to detect abdominal ten-
derness and abdominal masses. Children infected with
bacterial diarrheagenic pathogens often have abdominal
tenderness26–28; the absence of abdominal tenderness
provides some assurance that the bloody diarrhea is not
of bacterial origin. Right lower quadrant tenderness is
commonly observed in patients with acute bloody diar-
rhea, possibly because the terminal ileum is frequently
affected in patients with bacterial enteric infections
(“bacterial ileocecitis”).29–32Finding a mass in the abdo-
men, particularly on the right side, in a child raises the
possibility of an intussusception.33
A digital rectal examination in a child with acute
bloody diarrhea is unlikely to provide actionable infor-
mation. Nasogastric tubes, inserted to determine whether
the bloody “diarrhea” really reflects a large volume upper
gastrointestinal hemorrhage, are not necessary if the pa-
tient’s history suggests acute bloody diarrhea.
Laboratory evaluations should be used only to
obtain data that will help manage patients in the short
term, make rapid and correct diagnoses, and optimize
management. Thorough and proficient diagnostic micro-
biology is critical, along with clear written and verbal
communication between the clinician and the laboratory.
This can be a challenge, because it is not always easy to
find the laboratory to which the specimen has been sent.
To identify the bacterial causes of diarrhea, stool
samples are cultured on agar plates or in liquid medium.
Cultures should be tested for pathogens that are likely to be
present in the local community and those that require
medical intervention. Ideally, the cultures should target
Salmonella, Shigella, Campylobacter, and E coli O157:H7.34,35
strategies for children with suspected or definite E. coli
O157:H7 infections are presented in Table 3.
It is important to diagnose E coli O157:H7 infections
rapidly, but the route to finding this pathogen is nuanced
and somewhat intricate. The Centers for Disease Control
and Prevention will soon publish guidance for bacterial
assessment of stools.39In our experience, the abilities to
culture stool specimens in a local laboratory (without
overnight transport) and to examine them twice daily are
ideal; these factors can reduce the time required to obtain
test results by 1 day.
Microbiologists should notify the requesting provider
by telephone if they detect a “presumptive positive” bac-
terial enteric pathogen. We define a presumptive positive
colony of E coli O157:H7 as a nonmucoid colony that
does not ferment on sorbitol MacConkey agar and reacts
with a reagent that detects the O157 lipopolysaccharide.
It is also important to include a control to confirm that
the agglutination is not a nonspecific bacterial pheno-
type. In addition, stool specimens should be tested simul-
taneously for Shiga toxin, by inoculating a broth culture
and performing a toxin assay after overnight culture.
Shiga toxin testing should not replace a complete anal-
May 2009 ACUTE BLOODY DIARRHEA1889
ysis of the culture because substantial difficulties in clin-
ical and public health management arise from conclu-
sions based solely on a positive Shiga toxin test.
Clinicians that receive reports stating that stool cul-
ture results are “negative” should remember that, al-
though enteric culture media for Salmonella and Shigella
might detect Yersinia and Aeromonas (an organism of
uncertain pathogenicity), such media is not optimal for
the detection of these latter 2 genuses. In addition, Sal-
monella/Shigella testing will not identify E coli O157:H7,
other STEC, or Vibrio parahaemolyticus. Thus, when phy-
sicians receive negative test results, they should confirm
that optimal microbiologic testing protocols and re-
agents were used in the analysis.
Testing for Clostridium difficile Infection
It is reasonable to consider C difficile infection in
any patient with acute bloody diarrhea; recent studies
have compelled us to reconsider our approach to diag-
nosing and managing such infections. Specifically, severe
C difficile infections can be acquired in the community,
are not always associated with recent antibiotic use, and
current diagnostic strategies are far from perfect.40–42
Patients infected with C difficile follow a more protracted
course before they seek medical attention than do pa-
tients infected with other bacterial enteric pathogens. A
C difficile diagnosis should not rely on a single test; that is,
at least 2 different antigens should be sought.42–44More-
over, when C difficile test results are positive, especially for
children, therapy should not be administered until it has
been determined that they are not also infected with
Salmonella, Shigella, Campylobacter, or E coli O157:H7. In
situations of coinfections, treatment or nontreatment of
the non-C difficile pathogen should take precedence.
Testing for Parasites and Viruses
Except for limited circumstances, there is little or no
justification for testing for diarrheagenic parasites in acute
bloody diarrhea samples from children in most of the de-
veloped world, because the diarrheagenic parasites found in
these regions do not cause colitis. In areas where amebic
dysentery still exists, diagnostic microbiology is often not
available. There is no role for viral testing of stools of
children with acute bloody diarrhea. Rare cases of cytomeg-
alovirus colitis are best diagnosed colonoscopically.
Positive test results for fecal occult blood, leukocytes,
lactoferrin, and calprotectin45sometimes indicate an enteric
infection. However, negative results from these tests cannot
exclude the presence of a pathogen. In any case, such tests
are not warranted in patients with acute bloody diarrhea;
their stool samples should always undergo microbiologic
analysis for bacterial enteric pathogens.
Nonmicrobiologic Laboratory Tests
Patients with severe diarrhea can present challeng-
ing clinical pictures and prompt their physicians to order
many additional studies. However, the results of these
tests can be difficult to interpret and are sometimes
misleading. Therefore, it is best to request a limited yet
strategically chosen panel of tests.
A complete blood count is useful for evaluating pa-
tients with bloody diarrhea. Increased levels of hemoglo-
bin indicate dehydration. Patients with suspected E coli
O157:H7 infections benefit when platelet counts and
hemoglobin levels are monitored, especially those that
progress to HUS. Patients infected with Shigella or Salmo-
nella often have leukocytosis or a relative bandemia. It is
unlikely that a coagulation disorder would manifest as
bloody diarrhea; therefore, coagulation studies do not
provide helpful information. It is important to test for
electrolyte and serum creatinine levels.
If a bacterial pathogen is suspected to be the cause of
bloody diarrhea in infants, the elderly, or immunocom-
Table 3. Management Strategies for Children With
Suspected or Definite E coli O157:H7 Infections
● Infuse ?20 mL/kg intravenously on presentation, assuming that
there is no evidence of cardiopulmonary overload.
● Withhold antibiotics, antimotility agents, narcotics, and
nonsteroidal anti-inflammatory drugs.
● Continue intravenous isotonic crystalloid (normal saline, normal
saline with 5% dextrose, or lactated Ringer’s solution) at
maintenance volume requirements. Do not substitute hypotonic
fluids, such as half normal saline, for isotonic fluids. We
encourage repeated boluses of normal saline (10–20 mL/kg) if
there is any question of diminished urine output, assuming that
the patient is not showing signs of central volume overload.
● It is acceptable to add potassium to the intravenous fluids if the
serum potassium concentration is normal or low.
● Most patients can eat or drink fluids, but there is often poor
intake at the height of the illness.
● Daily laboratory testing should consist of a complete blood
count, electrolytes, and serum urea nitrogen and creatinine
● The patient should be admitted to an institution skilled in the
monitoring of fluid status. Children and the elderly should be
monitored especially closely.
● The daily platelet count, the clinical condition, and the day of
illness are the most important indicators of patient status. The
risk of HUS is greatly diminished when the platelet count rises or
if the platelet count is stable and symptoms are resolved or
resolving. We favor repeating laboratory tests 1 day after
● As the creatinine concentration rises, patients should be
monitored even more closely for hypertension or signs of central
overload. They should be transferred, if necessary, to an
institution where acute renal failure can be managed and
treated. We do not believe that a rising creatinine in a still-
urinating patient is grounds for fluid restriction as long as they
can be monitored assiduously.
● Peripheral edema should not prompt fluid restriction if there is no
● Avoid the antecubital region for phlebotomies, because this site
might be needed for a peripherally inserted central catheter.
Think about this eventuality as soon as the platelet count starts
1890 HOLTZ ET AL GASTROENTEROLOGY Vol. 136, No. 6
promised patients, a blood culture should be performed.
We discourage performing urinalysis if a patient has
bloody diarrhea. Although in evaluating infants with
fever (with or without diarrhea), urine testing is per-
formed to exclude the presence of a urinary tract infec-
tion, it is often difficult to obtain a sufficiently clean
urine sample to make an accurate diagnosis. A contam-
inated urine sample can result in unwarranted adminis-
tration of antibiotics, which are contraindicated for
many causes of bloody diarrhea. A catheterized urine
specimen should not be obtained from a patient with
acute diarrhea, because of the risk of introducing fecal
bacteria into the urinary tract. If there is a concern about
renal injury in a patient who might be infected with E coli
O157:H7, serum creatinine concentrations can be ana-
lyzed to provide information about renal function.
Radiographic Imaging and Surgical
Evaluators of patients with acute bloody diarrhea
are sometimes so concerned about abdominal pain and
tenderness that they consider the possibility of an intra-
abdominal lesion that requires surgery. Imaging is often
used, and in infectious colitis inflammation of the ter-
minal ileum and colon can be observed by ultrasound
scanning or computerized tomography (CT).32,46,47Bar-
ium enemas that show colonic edema (“thumb-printing”
of colonic mucosa) can indicate the presence of an E coli
O157:H7 infection. Dilated bowel loops are nonspecific
and are often observed on plain abdominal radiographs
in many patients with diarrhea.
Even though CT is often obtained in these presenta-
tions, the information obtained from such analysis of
children with acute bloody diarrhea is considerably less
valuable than that from directed examination and careful
collection of the patient’s history; there is scant likeli-
hood of identifying an abnormality on CT scan that
changes the diagnostic or therapeutic strategy. Further-
more, there are potential long-term consequences of ab-
dominal radiation,48especially for children.49We recog-
nize that CT evaluation of the abdomen in adults with
acute bloody diarrhea might have value for those in
whom a diagnosis of ischemic colitis is entertained (see
below), but we believe that this is a inappropriately used
test in childhood.
Diarrhea can often be present in patients with nonin-
fectious appendicitis,50and bacterial gastrointestinal in-
fections can be associated with appendicitis51or colonic
perforation.52,53However, in children, bloody diarrhea
almost never signifies a disorder that warrants surgical
Initial Empiric and Symptomatic Therapy
After evaluating a patient with bloody diarrhea,
determining that vascular compromise is neither present
nor imminent, and receiving negative laboratory test re-
sults, providers must then decide how (and if) to relieve
symptoms and treat the illness. Patients should not be
given narcotics or antimotility agents because of their
complications. This concern is admittedly based on case
reports and association with small studies,54–56but re-
cent reports attesting to the safety of antimotility agents
in patients with bloody diarrhea57,58analyzed treatment
outcomes of ?100 subjects, so the level of risk was not
precisely established. There is particular concern about
giving antimotility agents, including opioid narcotics, to
patients who might be infected with E coli O157:H7,
because of their association with increased risk of HUS,59
the neurologic complications that could arise from
HUS,60,61and the prolongation of illness with a longer
duration of bloody diarrhea.59Intravenous boluses of
normal saline (20–30 mL/kg body weight) are often use-
ful in alleviating abdominal pain.
We discourage the administration of antibiotics to
patients with acute bloody diarrhea unless it is clear that
a pathogen is present that would respond to the therapy.
Although several studies of adults have concluded that
whether or not a pathogen is identified, antibiotic ad-
ministration shortens the disease course, but in each of
those studies the benefits were limited62–65; the time to
first formed stool was approximately 1 day less in the
treated groups than in the control groups. Such a small
benefit should be weighed against the potential conse-
quences: antibiotics are associated with increased risk of
HUS in children59and adults66infected with E coli O157:
H7. Salmonella infection of otherwise healthy persons
does not warrant antibiotic therapy; the empiric use of
antimicrobials might prolong fecal excretion without
substantially improving the symptoms.62,67,68Certainly,
acute bloody diarrhea can be caused by Shigella infection.
In these cases antibiotics are indicated, but in North and
South America, Japan, Europe, and most of Asia, where S.
dysenteriae serotype 1 infections are rare, shigellosis is not
life threatening, so there is sufficient time to wait for
culture results before starting treatment.
Children with acute bloody diarrhea should be admit-
ted to the hospital, and volume should be expanded
parenterally with isotonic crystalloid. This can provide
comfort and most importantly protect against severe
HUS. Early in the course of illness, volume expansion
with isotonic crystalloid, rather than hypotonic fluids,
was associated with a less severe course when HUS devel-
oped.69Hospitalization, by removing the infected patient
from the community, can halt secondary spread through
more consistently observed contact isolation measures.
Werber et al70proposed that there would be considerably
lower rates of secondary transmission of E coli O157:H7
and of associated HUS in households if primary cases
were hospitalized. They also calculated that the isolation
(by hospitalization) of 47 patients (95% confidence inter-
val, 16–78 patients) would prevent 1 case of HUS.70
May 2009 ACUTE BLOODY DIARRHEA1891
Interestingly, this analysis determined that this interven-
tion for E coli O157:H7 infections aimed at preventing
renal failure in contacts is as effective as administering
postexposure antibiotics to contacts of cases with menin-
gococcal meningitis to prevent secondary invasive menin-
Most cases of bloody diarrhea resolve spontane-
ously, whether a causative agent is identified and whether
antibiotics are used. In children, intravenous fluid admin-
istration is stopped, and patients are discharged when it is
clear that they are not infected with E coli O157:H7, when
there is clear clinical improvement, and when the platelet
count increases or stabilizes. This determination can often
be made in advance of the complete stool culture result; it
can take several additional days to exclude the presence of
identification of a pathogen other than E coli O157:H7 that
accounts for the symptoms.
Useful guidelines for the management of patients defi-
nitely or possibly infected with E coli O157:H7 are presented
in Table 4. When specific data are collected in the context of
the day of illness on which the information comes to light,
it is possible to determine the prognosis (Figure 1).
The kinetics of E coli O157:H7 infections, in terms of
resolution or development of HUS, are portrayed in Fig-
Table 4. Guidelines for the Management of Patients Definitely or Possibly Infected with E coli O157:H7
Level of risk Day of illnessPlatelet trend Clinical statusCulture resultManagement
High risk1 UnknownDay to day trend
PendingBy definition on day 1, an inter-day trend is
not established. Occasionally symptoms
resolve extremely quickly, and in this
case we discharge to home after 12
hours of observation, with a follow-up
CBC the day after discharge.
Clinical improvement is always
encouraging, but it is not prudent to
discharge patients until the trend in the
platelet count is known early in illness.
Still ill, or
Still ill, or
Pending, or E coli
Indeterminate risk 5–9Unknown Pending, or E coli
Admit and await trend in platelet count.
Paradoxically, children infected with E
coli O157:H7 who present late in illness
have lower risk of HUS, but if HUS
develops, it has a tendency to be more
In this situation, discharge is appropriate,
with a follow-up platelet count on the
day following discharge. Once E coli
O157:H7 is excluded, it is less
important to obtain a post-discharge
Patients who have persistent or worsening
symptoms should not be discharged.
Causes of bloody other than E coli O157:
H7 are less consequential, and patients
can usually be safely discharged if this
pathogen is excluded.
Patients who have persisent of worsening
symptoms should not be discharged, but
they probably will not develop HUS.
No need for second determination of
Low risk 2–4Risingb
Still ill, or
2–4 Unknown E coli O157:H7
2–4Unknown Still ill, or
E coli O157:H7
Improving Pending, or E coli
Pending, or E coli
Still ill, or
Patients who have persistent severe or
worsening symptoms should not be
discharged, but their risk of HUS is low.
NOTE. This table uses a variety of data (day of illness, trend in platelet count, culture results, and clinical condition) that we find helpful in
assessing a patient’s HUS risk. Management protocols are shown in Figure 1.
aA falling platelet count is a value that is at least 5% lower than the platelet count one day earlier.
bA rising platelet count is a value that is at least 5% higher than the platelet count one day earlier.
cA stable platelet count is within 5% of the platelet count one day earlier.
1892 HOLTZ ET ALGASTROENTEROLOGY Vol. 136, No. 6
Bloody diarrhea is most appropriately defined as
gross blood in liquid or largely liquid stool. Although
some clinicians claim that an infection can be identified
by detecting occult blood in the absence of grossly bloody
stools,71this finding should not prompt the same diag-
nostic and therapeutic measures as clearly bloody diar-
rhea. Sometimes, bleeding is caused by trauma to the
distal gastrointestinal tract during forceful defecations in
acute nonbloody diarrhea. In these situations, the bleed-
ing resolves within hours, and patients can be discharged
after a minimal period of observation and volume expan-
sion (6–12 hours) without use of the resolution algo-
rithm described earlier.
Considerations in Adults
Bloody Diarrhea Versus Gastrointestinal
The evaluation of bloody diarrhea in adults is
not as straightforward as in children because adults
have a higher incidence of noninfectious causes of
bleeding, many of which must be rapidly identified.
Patients with symptoms of ?3 days in duration tend
to be seen in emergency departments or urgent care
centers rather than primary care offices, reflecting the
changing landscape of medical care delivery in the
United States. In these hectic settings, clinicians might
not delve into important details of the patient’s bowel
habits and bleeding history, leading to admission to an
inappropriate service (surgical or medical). The result-
ing delay can pose problems not just for the patient
but also for health care workers, who could become
infected with a pathogen from a misdirected classifi-
cation of a patient.
Reflecting the lack of systematic studies of adults with
bloody stools, most texts and reviews discuss this prob-
lem in sections that include a diverse set of diagnoses
(infectious diarrhea, gastrointestinal bleeding, inflamma-
tory bowel disease, diverticulitis, and cancer). Although
there is little published guidance on this topic, the im-
portance of categorizing adults with bloody stools can-
not be overestimated, because it drives the diagnostic
evaluation and its temporal priorities. The relevant infor-
mation to obtain during initial evaluations of adults with
Figure 1. Management flow chart for patients with bloody diarrhea.
This flowchart suggests strategies for managing patients definitely or
by guidance in Table 4. BUN, blood urea nitrogen; CBC, complete
blood count; IVF, intravenous fluid.
imately 3 days after ingestion of the organism, the patient develops
diarrhea, abdominal pain, fever, and vomiting. The diarrhea becomes
bloody 1–3 days later, rarely on the first day. In 80%–90% of infected
children with positive cultures, visible blood is present in the stools.
When bloody diarrhea first develops, the patient has a normal platelet
count, creatinine concentration, and packed-cell volume, with no red-
systems are done early in the illness, there is evidence that thrombin
generation is increased, fibrin deposition is occurring, and plasminogen
activation is suppressed.13The platelet count is the first abnormality to
appear and is used to monitor vascular injury in this disorder. Patients
need to be monitored only until it is apparent that their platelet count is
stable or rising (Table 3). Reprinted with permission from Lancet
May 2009 ACUTE BLOODY DIARRHEA1893
bloody stools, for use in prioritizing diagnostic options,
is presented in Table 5.
Causes of Acute Bloody Diarrhea in Adults
Causes of acute bloody diarrhea are numerous in
adults (Table 6); physicians are faced with the task of
making the correct diagnosis. Age is a key component in
stratifying probable causes; diverticulitis, for example is
unusual in patients younger than 50 years,72whereas
inflammatory bowel disease is more common in this age
group.73Passage of frankly bloody stools without mucus
and without pain suggests anatomic gastrointestinal
bleeding, such as angiodysplasia.
A combination of abdominal pain, cramping, and
stools mixed with blood and mucus suggest that a pa-
tient has colitis but does not distinguish infectious colitis
from idiopathic, inflammatory, or other causes. Patients
older than 50 years with this combination of symptoms,
along with a low-grade fever and left lower quadrant
tenderness, are likely to have diverticulitis,72whereas in
younger patients IBD73and infectious colitis figure more
prominently as diagnostic considerations. Ischemic coli-
tis is a concern in the elderly and in patients with vascu-
lar disease or hypercoagulable states, but colonic isch-
emia can occur idiopathically in younger persons,
especially in long-distance runners.74However, it is im-
portant to exclude infection in all such situations. Most
notably, E coli O157:H7 infections can be misdiagnosed
as noninfectious forms of colitis.75Finding a pathogen
can cause a frameshift to a patient’s care by clarifying a
hitherto confusing diagnostic picture.
Our appreciation of the epidemiology of C difficile infec-
tion in the developed world is changing rapidly. Unexpected
severe and fulminant disease results from infection with a
hypervirulent strain, North American Pulse Field Type 1,
which is widespread in North America76,77and western
Europe. Severe C difficile infections have been reported to
occur in nontraditional risk groups, including healthy per-
sons in the community, in persons without antimicrobial
exposure, and in pregnant women.78,79Bloody diarrhea has
been described in some of these patients.
A patient’s travel history should be sought, but travel-
er’s diarrhea is most frequently caused by enterotoxigenic
E coli and is usually not bloody.80In obtaining a travel
history relevant to bacterial colitis, the health care pro-
vider should inquire about locations visited in the 2
weeks before symptom onset; this time frame encom-
passes the typical incubation periods of these pathogens.
Typhoid fever is uncommon in developed nations, and
many clinicians are unfamiliar with the disease; hemor-
rhage from a necrotic Peyer’s patch can occur as a com-
plication in ?10% of patients with typhoid fever, but
they are usually ill for ?2 weeks.81Amoebiasis82and
schistosomiasis83can present with visible blood in the
stools many months after leaving an endemic area; so, if
these are considerations, a longer and more detailed
retrospective inquiry should be made.
A history of food consumption, particularly of com-
mon foods such as produce, meat, and poultry, is rarely
Table 5. Initial Symptom Evaluation in Adult Patients with
● Character of stool: loose, formed, mixed with mucus
● Stool caliber and volume
● Character of blood: hematochezia, melena, clots, mixed with
● Bleeding intermittent or with every stool
● Bleeding in relation to defecation
● Frequency of bowel movements
● Location, radiation
● Character: cramping, dull ache
● Persistent or intermittent, intensity
First or previous similar episode
Health status, comorbid conditions
● Antimicrobial use within preceding month
● Raw shellfish
● Raw milk
● Congregate living: dormitory, assisted living, nursing home
Table 6. Causes of Acute Bloody Stools in Adults
● Bacteria: Campylobacter jejuni, Salmonella, E coli O157:H7
and selected STEC, V. parahaemolyticus, Shigella, Yersinia,
Aeromonas, C difficile
● Viruses: Cytomegalovirus
● Parasites: Entamoeba histolytica, Schistosomiasis
Inflammatory bowel disease
Anatomic gastrointestinal bleeding
● Gastric ulcer
● Nonsteroidal anti-inflammatory drugs
● Blood dyscrasias (multiple myeloma)
● Meckel’s diverticulum
● Stercoral ulceration
1894 HOLTZ ET AL GASTROENTEROLOGY Vol. 136, No. 6
useful without comparison to a control group. Occasion-
ally, eliciting a history of consumption of unusual foods
such as raw shellfish or raw milk products might prompt
consideration of Vibrio parahaemolyticus or E coli O157:H7
infections, respectively. Among the infectious causes of
bloody diarrhea in the United States, E coli O157:H7
predominates as the most common bacterial pathogen
isolated,84yet only 57% of laboratories in sites of the
Foodborne Diseases Active Surveillance Network check
for this pathogen on a routine basis.37Clinicians who are
unsure about the practices of their microbiology labora-
tory should simply specify that stool culture from adults
include testing for E coli O157:H7, in addition to Salmo-
nella, Shigella, and Campylobacter.
Colorectal cancer is common; as many as 40% of po-
tentially resectable cancers present with hematochezia or
melena.85,86Patients without fever, who have minimal
abdominal pain, are older in age, and who had an initial
loose stool containing blood should be examined for
colorectal cancer. Adults with bloody stools, vomiting,
and imaging results that suggest a bowel obstruction
might have intussusception. Although much more com-
mon in children, intussusception is not exclusively a
pediatric consideration. In adults, intussusception is a
surgical emergency that can be caused by benign or
Practical Approaches to Adults with Bloody
Patients with severe illness who are immunocom-
to have an anatomic disorder and warrant hospital admis-
sion for close monitoring. Patients admitted to hospital
they are not infected with an enteric pathogen.
Stools with pus or mucus and blood should be sent for
bacterial culture; adult stool samples should be handled
with the same considerations as those of children. Tests
for fecal ova and parasites should be performed if bacte-
rial cultures and C difficile toxin assays are negative and if
the patient has lived or visited areas where amoebiasis or
schistosomiasis is endemic. In patients known to have
IBD, microbiologic analyses should be performed if they
have flares of unusual severity or do not respond to their
usual treatment, because enteric infections can compli-
cate this chronic condition.88
Older patients with moderate generalized abdominal
cramping and bloody stools present conundrums; it can
be difficult to rank order infectious colitis, ischemic co-
litis, diverticulitis, and colon cancer as possible causes.
These patients almost invariably undergo an abdominal
CT scan, which includes administration of oral contrast
that might make subsequent stool specimens difficult to
analyze by microscopic or microbiologic evaluation. For
this reason it is best to obtain specimens for bacterial
cultures and C difficile toxin studies before imaging stud-
ies are performed. This is particularly important if the
patient is a member of a group of ill persons in a con-
gregate living facility.89If a stool specimen is not avail-
able shortly after the patient’s initial evaluation, a deep
rectal swab specimen can be obtained and sent for bac-
terial culture analysis, but it will not suffice for C difficile
An initial assessment of the cause of an illness is
usually based on the patient’s age and demography, the
characterization of the bloody stools, and the patient’s
health history and medication use. The goals of the
physical examination are to determine the severity of
illness (fever, hypotension) and to learn details about
associated pain (location, rebound, and ileus). The next
diagnostic steps are governed by the hierarchy of causes;
physicians must choose among CT scanning of the ab-
domen, colonoscopy, or angiography.90Plain films have
limited usefulness in the evaluation of bloody stools in
adults. CT scans, to be most helpful, should be per-
formed with oral as well as intravenous contrast; if pro-
found ileus is present, patients can be rescanned a few
hours later. Although CT scans do not establish a specific
cause, they contribute information that can be used to
fully assess adults with bloody diarrhea, such as anatomic
localization and extent of bowel involvement (diverticu-
litis or bowel-wall thickening), complications such as
perforation (free air) or pneumatosis coli, and occasion-
ally, vascular thrombosis.
Many individual features of the patient’s illness and ini-
tial evaluation factor into decisions that affect how and
when endoscopy is performed. For patients with an infec-
might have colon cancer, ischemic colitis, or IBD, colonos-
copy is a valuable diagnostic procedure. A colonoscopic
biopsy can detect infections that are unexpected (eg, schis-
cytomegalovirus colitis); both are treatable entities. In addi-
tion, the biopsy can differentiate acute self-limited (proba-
bly infectious) colitis from IBD.91–93However, the diagnos-
tic yield from a biopsy should be considered against the risk
of perforation; a biopsy can often be deferred for several
days to diminish this risk.
Acute bloody diarrhea is a challenging diagnostic
problem in children and adults. Acute bloody stools have
different spectrums of causes between adults and chil-
dren, but there are overlapping causes (infectious colitis
and less frequently intussusception). It is imperative to
identify patients with infectious causes, so that they are
appropriately treated with antimicrobials and so that
infection control measures can be put in place. It is also
important to remember that patients with acute bloody
diarrhea could be infected with E coli O157:H7, so rapid
and thorough enteric microbiology, especially bacterial,
evaluations are priorities. Adult patients with painful
May 2009 ACUTE BLOODY DIARRHEA1895
bloody stools present additional challenges; gastroenter-
ologists must establish a correct diagnosis that mini-
mizes risk and delay. However, they must also consider
broad diagnostic categories that include diverticulitis,
colon cancer, anatomic bleeding, and ischemic colitis. A
patient’s age and presentation details, health history, and
risk factors can guide initial assessments, aided by
thoughtful use of other methods, including CT scan,
colonoscopy, and angiography. For patients of all ages
acute bloody diarrhea is a medical emergency that war-
rants serious medical investigation.
1. Murphy MS. Management of bloody diarrhoea in children in pri-
mary care. BMJ 2008;336:1010–1015.
2. Goldsmid JM, Leggat PA. The returned traveller with diarrhoea.
Aust Fam Physician 2007;36:322–327.
3. Razzaq S. Hemolytic uremic syndrome: an emerging health risk.
Am Fam Physician 2006;74:991–996.
4. Karch H, Tarr PI, Bielaszewska M. Enterohaemorrhagic Esche-
richia coli in human medicine. Int J Med Microbiol 2005;295:
5. Welinder-Olsson C, Kaijser B. Enterohemorrhagic Escherichia coli
(EHEC). Scand J Infect Dis 2005;37:405–416.
6. O’Ryan M, Prado V, Pickering LK. A millennium update on pedi-
atric diarrheal illness in the developing world. Semin Pediatr
Infect Dis 2005;16:125–136.
7. Wieland T. Travellers in medical practice [in German]. Praxis
(Bern 1994) 2005;94:233–237.
8. Cleary TG. The role of Shiga-toxin-producing Escherichia coli in
hemorrhagic colitis and hemolytic uremic syndrome. Semin Pedi-
atr Infect Dis 2004;15:260–265.
9. Uhlen S, Toursel F, Gottrand F. Treatment of acute diarrhea:
prescription patterns by private practice pediatricians [in French].
Arch Pediatr 2004;11:903–907.
10. Lawson JM. Update on Escherichia coli O157:H7. Curr Gastroen-
terol Rep 2004;6:297–301.
11. Gore JI, Surawicz C. Severe acute diarrhea. Gastroenterol Clin
North Am 2003;32:1249–1267.
12. Pollock KG, Stewart A, Beattie TJ, et al. From diarrhoea to hae-
molytic uraemic syndrome—when to seek advice. J Med Micro-
13. Chandler WL, Jelacic S, Boster DR, et al. Prothrombotic coagu-
lation abnormalities preceding the hemolytic-uremic syndrome.
N Engl J Med 2002;346:23–32.
14. Tarr PI, Gordon CA, Chandler WL. Shiga-toxin-producing Esche-
richia coli and haemolytic uraemic syndrome. Lancet 2005;365:
15. Tsai HM, Chandler WL, Sarode R, et al. von Willebrand factor and
von Willebrand factor-cleaving metalloprotease activity in Esche-
richia coli O157:H7-associated hemolytic uremic syndrome. Pe-
diatr Res 2001;49:653–659.
16. Inward CD, Howie AJ, Fitzpatrick MM, Rafaat F, Milford DV, Taylor
CM. Renal histopathology in fatal cases of diarrhoea-associated
haemolytic uraemic syndrome. British Association for Paediatric
Nephrology. Pediatr Nephrol 1997;11:556–559.
17. Richardson SE, Karmali MA, Becker LE, Smith CR. The histopa-
thology of the hemolytic uremic syndrome associated with vero-
cytotoxin-producing Escherichia coli infections. Hum Pathol
18. Riella MC, George CR, Hickman RO, et al. Renal microangiopathy
of the hemolytic-uremic syndrome in childhood. Nephron 1976;
19. Habib R. Pathology of the hemolytic uremic syndrome. In: Kaplan
BS, Trompeter RS, Moake JL, eds. Hemolytic uremic syndrome
and thrombotic thrombocytopenic purpura. New York, NY: Marcel
20. Jones TF, McMillian MB, Scallan E, et al. A population-based
estimate of the substantial burden of diarrhoeal disease in the
United States; FoodNet, 1996–2003. Epidemiol Infect 2007;
21. Monto AS, Koopman JS, Longini IM, Isaacson RE. The Tecumseh
study, XII: enteric agents in the community, 1976–1981. J Infect
22. Fox JP, Hall CE, Cooney MK, Luce RE, Kronmal RA. The Seattle
virus watch; II, objectives, study population and its observation,
data processing and summary of illnesses. Am J Epidemiol
23. Herikstad H, Yang S, Van Gilder TJ, et al. A population-based
estimate of the burden of diarrhoeal illness in the United States:
FoodNet, 1996–7. Epidemiol Infect 2002;129:9–17.
24. Denno DM, Keene WE, Hutter CM, et al. A tri-county comprehen-
sive assessment of risk factors for sporadic bacterial reportable
enteric infections in children. J Infect Dis 2009;199:467–476.
25. Karama M, Gyles, CL. Methods for genotyping verotoxin-produc-
ing Escherichia coli. Zoonoses Public Health 2009;56 (in press).
26. Klein EJ, Stapp JR, Clausen CR, et al. Shiga toxin-producing
Escherichia coli in children with diarrhea: a prospective point-of-
care study. J Pediatr 2002;141:172–177.
27. Wong CS, Jelacic S, Habeeb RL, Watkins SL, Tarr PI. The risk of
the hemolytic-uremic syndrome after antibiotic treatment of Esch-
erichia coli O157:H7 infections. N Engl J Med 2000;342:1930–
28. Klein EJ, Boster DR, Stapp JR, et al. Diarrhea etiology in a
Children’s Hospital Emergency Department: a prospective cohort
study. Clin Infect Dis 2006;43:807–813.
29. Tarr PI, Weinberger E, Hatch EI Jr, Christie DL. Bacterial ileoce-
citis caused by Escherichia coli O157:H7. J Pediatr Gastroenterol
30. Puylaert JB, Vermeijden RJ, van der Werf SD, Doornbos L, Kou-
mans RK. Incidence and sonographic diagnosis of bacterial ileo-
caecitis masquerading as appendicitis. Lancet 1989;2:84–86.
31. Puylaert JB, Van der Zant FM, Mutsaers JA. Infectious ileocecitis
caused by Yersinia, Campylobacter, and Salmonella: clinical,
radiological and US findings. Eur Radiol 1997;7:3–9.
32. Miller FH, Ma JJ, Scholz FJ. Imaging features of enterohemor-
rhagic Escherichia coli colitis. AJR Am J Roentgenol 2001;177:
33. Klein EJ, Kapoor D, Shugerman RP. The diagnosis of intussus-
ception. Clin Pediatr (Phila) 2004;43:343–347.
34. Tarr P, Hecht G, Bass D. Bacterial, viral, and toxic causes of
diarrhea, gastroenteritis, and anorectal infections. In: Yamada T,
Alpers DH, Kaplowitz N, Kalloo A, Owyang C, Powell DW, eds.
Textbook of gastroenterology. 5th ed. Oxford, United Kingdom:
Blackwell Publishing (in press).
35. Bennett WE Jr, Tarr P. Enteric infections and diagnostic testing.
Curr Opin Gastroenterol 2009;25:1–7.
36. Boyce TG, Pemberton AG, Wells JG, Griffin PM. Screening for
Escherichia coli O157:H7—a nationwide survey of clinical labo-
ratories. J Clin Microbiol 1995;33:3275–3277.
37. Voetsch AC, Angulo FJ, Rabatsky-Ehr T, et al. Laboratory practices
for stool-specimen culture for bacterial pathogens, including
Escherichia coli O157:H7, in the FoodNet sites, 1995–2000.
Clin Infect Dis 2004;38(Suppl 3):S190–S197.
38. Hennessy TW, Marcus R, Deneen V, et al. Survey of physician
diagnostic practices for patients with acute diarrhea: clinical and
public health implications. Clin Infect Dis 2004;38(Suppl 3):S203–
39. Centers for Disease Control and Prevention. Diagnosis of Shiga
toxin-producing Escherichia coli by clinical diagnostic laborato-
ries, STEC Clinical Diagnostics Working Group (in press).
1896 HOLTZ ET AL GASTROENTEROLOGY Vol. 136, No. 6
40. Bartlett JG, Gerding DN. Clinical recognition and diagnosis of
Clostridium difficile infection. Clin Infect Dis 2008;46(Suppl 1):
41. Kelly CP, LaMont JT. Clostridium difficile—more difficult than
ever. N Engl J Med 2008;359:1932–1940.
42. Planche T, Aghaizu A, Holliman R, et al. Diagnosis of Clostridium
difficile infection by toxin detection kits: a systematic review.
Lancet Infect Dis 2008;8:777–784.
43. Ticehurst JR, Aird DZ, Dam LM, Borek AP, Hargrove JT, Carroll KC.
Effective detection of toxigenic Clostridium difficile by a two-step
algorithm including tests for antigen and cytotoxin. J Clin Micro-
44. Reller ME, Lema CA, Perl TM, et al. Yield of stool culture with
isolate toxin testing versus a two-step algorithm including stool
toxin testing for detection of toxigenic Clostridium difficile. J Clin
45. Huicho L, Campos M, Rivera J, Guerrant RL. Fecal screening
tests in the approach to acute infectious diarrhea: a scientific
overview. Pediatr Infect Dis J 1996;15:486–494.
46. Puylaert JB. Imaging and intervention in patients with acute right
lower quadrant disease. Baillieres Clin Gastroenterol 1995;9:
47. Morgan J, Bell M, Sadler MA. CT imaging of acute E. coli-related
colitis. Emerg Radiol 2007;14:187–189.
48. Hall EJ, Brenner DJ. Cancer risks from diagnostic radiology. Br J
49. Brenner DJ, Hall EJ. Computed tomography—an increasing
source of radiation exposure. N Engl J Med 2007;357:2277–
50. Sakellaris G, Tilemis S, Charissis G. Acute appendicitis in pre-
school-age children. Eur J Pediatr 2005;164:80–83.
51. Nussinovitch M, Shapiro RP, Cohen AH, Varsano I. Shigellosis
complicated by perforated appendix. Pediatr Infect Dis J 1993;
52. Kravitz GR, Smith K, Wagstrom L. Colonic necrosis and perfora-
tion secondary to Escherichia coli O157:H7 gastroenteritis in an
adult patient without hemolytic uremic syndrome. Clin Infect Dis
53. Helms M, Simonsen J, Molbak K. Foodborne bacterial infection
and hospitalization: a registry-based study. Clin Infect Dis 2006;
54. Brown JW. Toxic megacolon associated with loperamide therapy.
55. Eronen M, Putkonen H, Hallikainen T, Vartiainen H. Lethal gas-
troenteritis associated with clozapine and loperamide. Am J Psy-
56. Smith GS, Blaser MJ. Fatalities associated with Campylobacter
jejuni infections. JAMA 1985;253:2873–2875.
57. Taylor DN, Sanchez JL, Candler W, Thornton S, McQueen C,
Echeverria P. Treatment of travelers’ diarrhea: ciprofloxacin plus
loperamide compared with ciprofloxacin alone. A placebo-con-
trolled, randomized trial. Ann Intern Med 1991;114:731–734.
58. Murphy GS, Bodhidatta L, Echeverria P, et al. Ciprofloxacin and
loperamide in the treatment of bacillary dysentery. Ann Intern
59. Bell BP, Griffin PM, Lozano P, Christie DL, Kobayashi JM, Tarr PI.
Predictors of hemolytic uremic syndrome in children during a
large outbreak of Escherichia coli O157:H7 infections. Pediatrics
60. Cimolai N, Carter JE. Bacterial genotype and neurological com-
plications of Escherichia coli O157:H7-associated haemolytic
uraemic syndrome. Acta Paediatr 1998;87:593–594.
61. Cimolai N, Morrison BJ, Carter JE. Risk factors for the central
nervous system manifestations of gastroenteritis-associated he-
molytic-uremic syndrome. Pediatrics 1992;90:616–621.
62. Wistrom J, Jertborn M, Ekwall E, et al. Empiric treatment of acute
diarrheal disease with norfloxacin. A randomized, placebo-con-
trolled study. Swedish Study Group. Ann Intern Med 1992;117:
63. Goodman LJ, Trenholme GM, Kaplan RL, et al. Empiric antimicro-
bial therapy of domestically acquired acute diarrhea in urban
adults. Arch Intern Med 1990;150:541–546.
64. Dryden MS, Gabb RJ, Wright SK. Empirical treatment of severe
acute community-acquired gastroenteritis with ciprofloxacin. Clin
Infect Dis 1996;22:1019–1025.
65. Noguerado A, Garcia-Polo I, Isasia T, et al. Early single dose
therapy with ofloxacin for empirical treatment of acute gastroen-
teritis: a randomised, placebo-controlled double-blind clinical
trial. J Antimicrob Chemother 1995;36:665–672.
66. Dundas S, Todd WT, Stewart AI, Murdoch PS, Chaudhuri AK,
Hutchinson SJ. The central Scotland Escherichia coli O157:H7
outbreak: risk factors for the hemolytic uremic syndrome and
death among hospitalized patients. Clin Infect Dis 2001;33:
67. Thielman NM, Guerrant RL. Clinical practice. Acute infectious
diarrhea. N Engl J Med 2004;350:38–47.
68. Neill MA, Opal SM, Heelan J, et al. Failure of ciprofloxacin to
eradicate convalescent fecal excretion after acute salmonellosis:
experience during an outbreak in health care workers. Ann Intern
69. Ake JA, Jelacic S, Ciol MA, et al. Relative nephroprotection during
Escherichia coli O157:H7 infections: association with intrave-
nous volume expansion. Pediatrics 2005;115:e673–e680.
70. Werber D, Mason BW, Evans MR, Salmon RL. Preventing house-
hold transmission of Shiga toxin-producing Escherichia coli O157
infection: promptly separating siblings might be the key. Clin
Infect Dis 2008;46:1189–1196.
71. Siegel D, Cohen PT, Neighbor M, et al. Predictive value of stool
examination in acute diarrhea. Arch Pathol Lab Med 1987;111:
72. Jacobs DO. Clinical practice. Diverticulitis. N Engl J Med 2007;
73. Loftus EV Jr. Clinical epidemiology of inflammatory bowel dis-
ease: incidence, prevalence, and environmental influences. Gas-
74. Greenwald DA, Brandt LJ, Reinus JF. Ischemic bowel disease in
the elderly. Gastroenterol Clin North Am 2001;30:445–473.
75. Hunt CM, Harvey JA, Youngs ER, Irwin ST, Reid TM. Clinical and
pathological variability of infection by enterohaemorrhagic (Vero
cytotoxin producing) Escherichia coli. J Clin Pathol 1989;42:847–
76. McDonald LC, Killgore GE, Thompson A, et al. An epidemic, toxin
gene-variant strain of Clostridium difficile. N Engl J Med 2005;
77. Loo VG, Poirier L, Miller MA, et al. A predominantly clonal multi-
institutional outbreak of Clostridium difficile-associated diarrhea
with high morbidity and mortality. N Engl J Med 2005;353:2442–
78. Severe Clostridium difficile-associated disease in populations
previously at low risk—four states, 2005. MMWR Morb Mortal
Wkly Rep 2005;54:1201–1205.
79. Surveillance for community-associated Clostridium difficile—
Connecticut, 2006. MMWR Morb Mortal Wkly Rep 2008;57:34–
80. Steffen R. Epidemiology of traveler’s diarrhea. Clin Infect Dis
81. Parry CM, Hien TT, Dougan G, White NJ, Farrar JJ. Typhoid fever.
N Engl J Med 2002;347:1770–1782.
82. Haque R, Huston CD, Hughes M, Houpt E, Petri WA Jr. Amebiasis.
N Engl J Med 2003;348:1565–1573.
83. Elliott DE. Schistosomiasis. Pathophysiology, diagnosis, and
treatment. Gastroenterol Clin North Am 1996;25:599–625.
84. Slutsker L, Ries AA, Greene KD, Wells JG, Hutwagner L, Griffin
PM. Escherichia coli O157:H7 diarrhea in the United States:
May 2009 ACUTE BLOODY DIARRHEA1897
clinical and epidemiologic features. Ann Intern Med 1997;126: Download full-text
85. Speights VO, Johnson MW, Stoltenberg PH, Rappaport ES, Hel-
bert B, Riggs M. Colorectal cancer: current trends in initial clinical
manifestations. South Med J 1991;84:575–578.
86. Steinberg SM, Barkin JS, Kaplan RS, Stablein DM. Prognostic
indicators of colon tumors. The Gastrointestinal Tumor Study
Group experience. Cancer 1986;57:1866–1870.
87. Stechman M, Safranek P, Purkayastha S. Intussusception in a
young adult with bloody diarrhoea. Int J Surg 2004;2:121–122.
88. Babyatsky MW, Keroack MD, Blake MA, Rosenberg ES, Mino-
Kenudson M. Case records of the Massachusetts General Hos-
pital. Case 35-2007. A 30-year-old man with inflammatory bowel
disease and recent onset of fever and bloody diarrhea. N Engl
J Med 2007;357:2068–2076.
89. Levine WC, Smart JF, Archer DL, Bean NH, Tauxe RV. Foodborne
disease outbreaks in nursing homes, 1975 through 1987. JAMA
90. Brandt LJ. Bloody diarrhea in an elderly patient. Gastroenterology
91. Surawicz CM, Belic L. Rectal biopsy helps to distinguish acute
self-limited colitis from idiopathic inflammatory bowel disease.
92. Surawicz CM. The role of rectal biopsy in infectious colitis. Am J
Surg Pathol 1988;12(Suppl 1):82–88.
93. Surawicz CM, Haggitt RC, Husseman M, McFarland LV. Mucosal
biopsy diagnosis of colitis: acute self-limited colitis and idio-
pathic inflammatory bowel disease. Gastroenterology 1994;107:
Received December 15, 2008. Accepted February 17, 2009.
Address requests for reprints to: Phillip I. Tarr, MD, Division of
Gastroenterology and Nutrition, Department of Pediatrics,
Washington University School of Medicine, 660 S Euclid Avenue,
Campus Box 8208, St Louis, MO 63110. e-mail: Tarr@kids.wustl.edu;
fax: (314) 286-2895.
The authors thank Beth Wolf for assistance with this manuscript,
Dr Eileen Klein for her invaluable insight, Dr Robert Kennedy for
helpful review of our recommendations, Barb Kowalcyk for her
advice and perspective, and our many patients and their families
who have taught us much about enteric infections.
Conflicts of interest
The authors disclose no conflicts.
1898 HOLTZ ET AL GASTROENTEROLOGY Vol. 136, No. 6