AJR 2001; 177:619-623
© American Roentgen Ray Society
Imaging Features of Enterohemorrhagic Escherichia coli Colitis
Frank H. Miller1,
Jason J. Ma2 and
Francis J. Scholz3
1
Department of Radiology, Northwestern Memorial Hospital, Northwestern
University Medical School, 676 N. St. Clair St., Ste. 800, Chicago, IL
60611.
2
University of Illinois College of Medicine, 1740 W. Taylor St., Chicago, IL
60612.
3
Lahey Clinic, Medical Center, 41 Mall Rd., Burlington, MA 01805.
Received October 31, 2000;
accepted after revision March 16, 2001.
Address correspondence to F. H. Miller.
Abstract
OBJECTIVE. The purpose of this article is to define and illustrate
the radiologic findings in patients with enterohemorrhagic Escherichia
coli colitis.
CONCLUSION. Although not definitive, imaging studies in conjunction
with an appropriate clinical history can aid in the early diagnosis of E.
coli colitis and exclude surgical conditions. CT is more sensitive than
conventional radiography for detection. Contiguous involvement, including the
transverse colon, was seen in all patients. Because CT is becoming routine in
the initial workup of patients with acute abdominal pain, it is important for
the radiologist to suggest E. coli colitis in the proper setting.
Introduction
Only since the early 1980s has Escherichia coli O157:H7 been
recognized as a cause of hemorrhagic colitis. E. coli colitis is now
identified more frequently, in part because of increased incidence, but also
because of better reporting and laboratory testing. It is important to
diagnose E. coli colitis because it has high associated morbidity and
mortality.
Infected patients characteristically have an afebrile illness that begins
with abdominal cramps and watery diarrhea, which progresses to bloody diarrhea
[1]. Nonspecific abdominal pain
may be the most prominent symptom. Blood leukocytes are usually elevated but
there are few fecal leukocytes
[1], which are often present in
infectious colitis. For this reason, E. coli colitis may not be
suspected. Because most patients are afebrile, other causes of bloody
diarrhea, such as inflammatory bowel disease and ischemic colitis, may also be
clinically suspected. Lack of clinical suspicion, coupled with the need for
special microbiologic tests, frequently results in inappropriate treatment
with antibiotics, steroids, and unnecessary surgical procedures
[2]. Failure to diagnose E.
coli colitis can lead to hemolytic—uremic syndrome and death.
Therefore, radiologists must be familiar with E. coli colitis to
assist in early diagnosis.
Clinical findings have been well described, but there are limited reviews
of the radiologic findings of E. coli colitis in the literature and,
to our knowledge, even fewer reported cases of CT findings
[3]. We undertook a
retrospective review to better define the radiologic findings in patients who
had E. coli colitis and underwent CT, abdominal radiography, or a
barium enema.
Materials and Methods
We conducted a retrospective review of the clinical and radiologic findings
of seven patients with E. coli colitis seen at our instititutions
between 1992 and 1999. The patients were selected from teaching files and via
a computerized search of medical and radiology records. Our study population
included four male and three female patients with a mean age of 50 years
(range, 11-78 years), whose cases were unrelated to epidemics. All patients
were diagnosed with E. coli colitis on the basis of positive stool or
biopsy cultures. Six patients were examined with CT, five with conventional
radiography, and one with a barium enema.
CT was performed on model 9800 CT scanners (General Electric Medical
Systems, Milwaukee, WI). Helical CT was performed on HiSpeed Advantage
scanners (General Electric Medical Systems). Contiguous axial 10-mm-thick
sections were obtained with the 9800 CT scanner. Helical CT of the abdomen and
pelvis was performed with the following parameters: collimation, 7 mm; table
speed, 7 mm/sec; pitch, 1; voltage, 120 kVp; amperage, 210-300 mA. Five of six
patients received IV contrast material (150 mL of diatrizoate meglumine
[Reno-60; Bracco Diagnostics, Princeton, NJ] or iopamidol [Isovue; Bracco
Diagnostics]) at 2 mL/sec. All patients received 200-800 mL of barium sulfate
suspension (Readi Cat; E-Z-EM, Westbury, NY) beginning 1.0-1.5 hr before CT.
Rectal contrast material was not administered.
One experienced gastrointestinal radiologist retrospectively reviewed all
the films. Another researcher independently reviewed the medical records and
pathology reports. Abdominal radiographs were examined to determine the
bowel-gas pattern, including bowel dilatation, presence of thumbprinting, and
evidence of ileus or ascites. CT scans were analyzed to determine the
distribution of bowel involvement, degree of wall thickening, presence of a
double-halo or target sign (two or three concentric rings), pericolonic fluid
or stranding, pneumatosis, free air, and intraabdominal fluid. The bowel wall
was considered thickened if it measured more than 3 mm in diameter when
partially distended. After the abdominal radiographs and CT scans were
reviewed, the colonoscopic findings were correlated with the imaging
findings.
Results
All seven patients had bloody diarrhea and abdominal pain. Their WBCs
ranged from 11,100 to 23,300/mm3. Five patients were afebrile
(Table 1).
In all six patients who underwent CT examinations, findings showed abnormal
colonic wall thickening. The wall thickness ranged from 3 to 20 mm (average,
10 mm). The transverse colon was involved to a variable extent. There was
contiguous involvement without skip lesions. One patient had pancolitis (Fig.
1A,1B).
One patient had wall thickening limited to the transverse and descending
colons. One patient had diffuse thickening sparing the cecum, and another
patient had diffuse involvement sparing the sigmoid colon (Fig.
2A,2B).
One patient had involvement of the cecum, ascending colon, and transverse
colon sparing the descending and sigmoid colons
(Fig. 3). In one patient, wall
thickening was isolated to the cecum, ascending colon, and the terminal ileum,
with transverse colon involvement to a minimal degree. This patient was the
only patient with small-bowel involvement
(Fig. 4). A target sign of the
wall of the colon was seen in three (50%) of the six patients (Fig.
1A,1B,2A,2B,3).
The target sign was not seen in the patient without IV contrast material. Four
patients (67%) had pericolonic stranding (Figs.
1A,1B,2A,2B,3,4).
Four patients had mild to moderate amounts of intraabdominal fluid; in only
one patient could this fluid be accounted for by cirrhosis. Colonoscopy
performed on three patients revealed edematous and erythematous mucosa in the
same segments identified on CT in those patients.
View larger version (112K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A. —53-year-old febrile man with diarrhea and lower abdominal
pain of 2 days' duration. Patient had bloody and mucous stools of 24-hr
duration. Contrast-enhanced CT scan shows pancolitis with thickened ascending
colon and descending colon with target sign (solid arrows). Note
pericolonic inflammatory changes (open arrows).
|
|
View larger version (153K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B. —53-year-old febrile man with diarrhea and lower abdominal
pain of 2 days' duration. Patient had bloody and mucous stools of 24-hr
duration. CT scan reveals sigmoid colon (solid arrow) as less
thickened than rest of colon. Note pelvic fluid (open arrow). After
CT examination, patient remembered having eaten a largely uncooked meal
several days before admission.
|
|
View larger version (117K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B. —63-year-old woman with bloody diarrhea of 3 days' duration
and suspected bowel obstruction. Contrast-enhanced CT scan shows bowel wall
thickening involving ascending (open arrow), transverse (solid
straight arrow), and descending colons with pericolonic inflammation
(curved arrows). Note target sign of wall of ascending colon
(open arrow).
|
|
View larger version (126K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3. —54-year-old afebrile woman presented with abdominal pain and
bloody diarrhea. CT scan shows thickened colon with target sign (solid
arrows) predominately involving ascending and transverse colons. Note
pericolonic inflammatory changes (open arrow).
|
|
View larger version (145K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4. —21-year-old afebrile man with history of gastroesophageal
reflux and dull right lower quadrant pain. He was suspected to have
appendicitis. Contrast-enhanced CT scan shows inflammation of cecum
(curved solid arrow) and terminal ileum (straight solid
arrow) without evidence of appendicitis. Note pericolonic inflammatory
changes (open arrow). CT findings are similar to typhlitis. Patient
had eaten Chinese food before symptoms and Escherichia coli colitis
developed.
|
|
Abdominal radiographs were obtained in five patients. Thumbprinting was
seen in one patient with involvement of the transverse colon and splenic
flexure (Fig. 2A). One patient
had an adynamic ileus. The remaining three patients did not have dilated
bowel. Three patients had abdominal radiographs that showed a gasless abdomen.
In the patient who under-went a barium enema, submucosal edema was seen in the
cecum and transverse colon with thickened haustra
(Fig. 5).
View larger version (141K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5. —11-year-old boy with diarrhea and bloody stool.
Single-contrast barium enema shows thumbprinting suggesting submucosal edema
involving the cecum (solid arrows) and transverse colon (open
arrows).
|
|
Discussion
E. coli is primarily a food-borne pathogen that can be found
worldwide. Many outbreaks in the United States have been associated with the
consumption of contaminated and undercooked ground beef
[1]. The incidence of infection
seems to peak during the summer and autumn. Severe illness occurs mostly in
children and elderly individuals
[4]. The spectrum of illness
associated with E. coli colitis is broad and may range from
asymptomatic infection to nonbloody diarrhea, hemorrhagic colitis,
hemolytic—uremic syndrome, or death. The incubation period after
ingestion of contaminated food is 4-8 days
[1,
5]. The infective dosage
required to produce symptomatic infection is low. E. coli produces
Shiga-like toxins and adherence factors that inhibit protein synthesis and
cause epithelial cell death. Because patients with E. coli colitis
are at increased risk for hemolytic—uremic syndrome, it is crucial to
detect this organism promptly. Young and old patients are at greatest risk for
developing hemolytic—uremic syndrome (2-7%)
[4]. At our institution, a
78-year-old man who developed hemolytic—uremic syndrome did not receive
IV contrast material because he had renal disease. IV contrast material should
not be administered, especially late in the course of E. coli
colitis, if hemolytic—uremic syndrome is a possibility.
The differential diagnosis for E. coli colitis is extensive
because the symptoms are often nonspecific. Considerations include
appendicitis, ischemic colitis, and inflammatory bowel disease, as well as
infectious colitis due to Salmonella, Shigella, or
Campylobacter organisms; Clostridium difficile; or
Cytomegalovirus. Therefore, a carefully obtained history of food consumption
is essential. Stool specimens from all patients with acute onset of bloody
diarrhea should be cultured for E. coli. Many laboratories do not
routinely examine for this organism or toxin.
Imaging tests generally are not required for the diagnosis of E.
coli colitis. The diagnosis is typically made from clinical findings and
suggested by a history of ingestion of uncooked or undercooked meat. In our
patients, the diagnosis was more difficult because the onset was sporadic and
not associated with epidemics. Imaging studies are particularly useful when
the clinical findings are nonspecific. Cultures are often reserved for
patients who are unresponsive to therapy. At a minimum, imaging studies can
help exclude certain pathologic processes and prevent inappropriate surgical
or medical treatment. Two of seven patients in our study were suspected to
have appendicitis, which CT helped exclude. At least three patients were
prescribed antibiotics on admission; these were abruptly discontinued when
E. coli colitis was diagnosed. Increased radiologist awareness of
E. coli colitis will expand the differential diagnosis in patients
with suspected pseudomembranous colitis and prevent inappropriate antibiotics
or surgical exploration for appendicitis or intussusception.
In our patients with E. coli colitis, CT showed features of
colitis that included colonic wall thickening, often with a target sign and
pericolonic stranding. The six patients who had CT examinations showed colonic
wall thickening, the most sensitive finding in our patients. The distribution
of involvement was variable, ranging from isolated segmental to diffuse
involvement of the colon. In our patients, as in other studies
[6], the transverse colon was
most frequently involved. All seven patients had contiguous involvement. The
target sign was a useful finding of colitis seen in 60% of the five patients
who received IV contrast material. The target sign is a sign of inflammation
or edema involving the wall of the bowel. Although nonspecific, this finding
suggests an inflammatory or infectious, not a malignant, cause. In the six
patients, ascites (67%) and pericolonic stranding (67%) were also fairly
sensitive associated findings that should increase suspicion for acute
inflammation. These features can be seen with other colitides, including
pseudomembranous colitis, typhlitis, ischemic colitis, inflammatory bowel
disease, and other infectious colitides.
Conventional radiographs were not as sensitive as CT. Only two (40%) of the
five patients who were examined with conventional radiography showed
abnormalities on radiographs. One patient had an adynamic ileus. The other
patient showed signs of inflammation including thumbprinting. The relatively
low incidence of ileus and thumbprinting in our study may relate to the small
number of patients, differences in the severity of the disease, or ages of the
patients compared with other studies. Conventional radiographs are useful to
look for signs of obstruction or ileus and bowel wall edema. A barium enema
may show nonspecific thumbprinting and ulcerations
[7,
8]
(Fig. 5). Sonography used to
diagnose appendicitis has also been reported to diagnose E. coli
ileocecitis. The findings include nonspecific bowel wall thickening
[9].
The clinical history is critical to exclude other colitides that may mimic
E. coli colitis. For example, generally pseudomembranous colitis
occurs after the use of antibiotics, and typhlitis occurs in immunosuppressed
patients. In contrast, E. coli colitis occurs a few days after
ingestion of contaminated food products, generally in immunocompetent
patients. At least three patients specifically reported consumption of
possibly tainted meat, but only after CT was performed. Other infectious
colitides can usually be differentiated on the basis of stool cultures.
Cultures are especially helpful for patients who are uncertain of their
history of food ingestion. However, a negative stool culture does not
eliminate E. coli as a cause. E. coli is rapidly cleared and
may be missed if not tested within 6 days of the onset of symptoms
[10]. Stool cultures are
insensitive to lower bacterial counts, and the excretion in the stool can be
intermittent [11].
No specific treatment exists for E. coli colitis. Supportive
measures are the mainstays of therapy. Antibiotics have not proven to be
effective. Antimotility drugs and narcotics are contraindicated because they
may delay clearance of the pathogen and may even increase the risk of
developing hemolytic—uremic syndrome
[12]. Because E. coli
colitis is highly contagious, patients should be advised to maintain hygienic
practices and avoid group settings. Public health authorities should be
notified so that they can initiate epidemiologic investigations and prevent
transmission.
In conclusion, E. coli is a relatively unrecognized cause of
colitis that mimics other colitides but can rapidly progress to
hemolytic—uremic syndrome, a life-threatening condition. Although not
definitive, imaging studies can be extremely useful for the timely diagnosis
of E. coli colitis and to exclude other conditions. As CT becomes
more routine in the initial workup of patients with gastrointestinal
complaints, it is important to suggest E. coli infection in the
correct clinical setting, because the radiologist may be the first to suspect
it.
References
-
Riley LW, Remis RS, Helgerson SD, et al. Hemorrhagic colitis
associated with a rare Escherichia coli serotype. N Engl J
Med 1983;308:681
-685[Abstract]
-
Griffin PM, Ostroff SM, Tauxe RV, et al. Illnesses associated with
Escherichia coli O157:H7 infections: a broad clinical spectrum.
Ann Intern Med
1988;109:705
-712
-
Ferrer MD, Pallardo Y, Nogues E. Enterohemorrhagic colitis caused
by Escherichia coli: CT findings. (letter)
AJR
1997;169:912[Medline]
-
Griffin PM, Tauxe RV. The epidemiology of infections caused by
Escherichia coli O157:H7, other enterohemorrhagic E. coli,
and the associated hemolytic—uremic syndrome. Epidemiol
Rev 1991;13:60
-98[Free Full Text]
-
Ryan CA, Tauxe RV, Hosek GW, et al. Escherichia coli
O157:H7 diarrhea in a nursing home: clinical, epidemiological, and
pathological findings. J Infect Dis
1986;154:631
-638[Medline]
-
Shortsleeve MJ, Wilson ME, Finklestein M, Gardner RC. Radiologic
findings in hemorrhagic colitis due to Escherichia coli O157:H7.
Gastrointest Radiol
1989;14:341
-344[Medline]
-
Kawanami T, Bowen AD, Girdany BR. Enterocolitis: prodrome of the
hemolytic—uremic syndrome. Radiology
1984;151:91
-92[Abstract/Free Full Text]
-
Peterson RB, Meseroll WP, Shrago GG, Gooding CA. Radiographic
features of colitis associated with hemolytic—uremic syndrome.
Radiology
1976;118:667
-671[Abstract]
-
Tarr PI, Weinberger E, Hatch EI Jr, Christie DL. Bacterial
ileocecitis caused by Escherichia coli O157:H7. J Pediatr
Gastroenterol Nutr
1992;14:261
-263[Medline]
-
Wells JG, Davis BR, Wachsmuth IK, et al. Laboratory investigation
of hemorrhagic colitis outbreaks associated with a rare Escherichia
coli serotype. J Clin Microbiol
1983;18:512
-520[Abstract/Free Full Text]
-
Tarr PI. Escherichia coli O157:H7: clinical, diagnostic,
and epidemiological aspects of human infection. Clin Infect
Dis 1995;20:1
-10[Medline]
-
Cimolai N, Carter JE, Morrison BJ, Anderson JD. Risk factors for
the progression of Escherichia coli O157:H7 enteritis to
hemolytic—uremic syndrome. J Pediatr
1990;116:589
-592[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
