CLINICAL MANIFESTATIONS AND COMPLICATIONS
Bacillary dysentery is clinically similar regardless of infecting
serotype. There are some clinical differences, particularly relating
to the greater severity and risk of complications with S. dysenteriae
serotype 1 infection. Ingestion of shigellae is followed by an
incubation period of 12 hr to several days before symptoms
ensue. Severe abdominal pain, high fever, emesis, anorexia, generalized
toxicity, urgency, and painful defecation characteristically
occur. The diarrhea may be watery and of large volume
initially, evolving into frequent, small-volume, bloody mucoid
stools. Most children never progress to the stage of bloody diarrhea,
but some have bloody stools from the outset. Significant
dehydration is related to the fluid and electrolyte losses in feces
and emesis. Untreated diarrhea can last 1-2 wk; only about 10%
of patients have diarrhea persisting for> 10 days. Persistent diarrhea
occurs in malnourished infants, children with AIDS, and
occasionally previously normal children. Even nondysenteric
disease can be complicated by persistent illness.
Physical examination initially shows abdominal distention
and tenderness, hyperactive bowel sounds, and a tender rectum
on digital examination. Neurologic findings are among the most
common extraintestinal manifestations of bacillary dysentery,
occurring in as many as 40% of hospitalized children. Enteroinvasive
E. coli can cause similar neurologic toxicity. Convulsions,
headache, lethargy, confusion, nuchal rigidify, or hallucinations
may be present before or after the onset of diarrhea. The cause
of these neurologic findings is not understood. In the past, these
symptoms were attributed to the neurotoxicity of Shiga toxin,
but it is now clear that this explanation is wrong because the
organisms isolated from children with Shigella-related seizures
are usually not Shiga toxin producers. Seizures sometimes occur
when little fever is present, suggesting that simple febrile convulsions
do not explain their appearance. Hypocalcemia or hyponatremia
may be associated with seizures in a small number of
patients. Although symptoms often suggest central nervous
system infection and cerebrospinal fluid pleocytosis with minimally
elevated protein levels can occur, meningitis due to shigellae
is rare. Based on animal studies, it has been suggested that
proinflammatory mediators, including TNF-a and interleukinl~,
nitric oxide, and corticotropin-releasing hormone, all play
a role in the enhanced susceptibility to seizures caused by S.
dysenteriae.
The most common complication of shigellosis is dehydration.
Inappropriate secretion of antidiuretic hormone with profound
hyponatremia can complicate dysentery, particularly when S.
dysenteriae is the etiologic agent. Hypoglycemia and proteinlosing
enteropathy are common. Other major complications
include sepsis and disseminated intravascular coagulation, particularly
in very young, malnourished children. Given that shigellae
penetrate the intestinal mucosal barrier, these events are
surprisingly uncommon.
Shigella and sometimes other gram-negative enteric bacilli are
recovered from blood cultures in 1-5% of patients in whom
blood cultures are taken; because patients selected for blood
cultures represent a biased sample, the risk of bacteremia in
unselected cases of shigellosis is presumably lower. Bacteremia is
more common with S. dysenteriae serotype 1 than with other
shigellae; the mortality rate is high (-20%) when sepsis occurs.
Neonatal shigellosis is rare. Neonates might have only lowgrade
fever with mild, non bloody diarrhea. However, complications
occur more commonly than in older children and include
septicemia, meningitis, dehydration, colonic perforation, and
toxic megacolon.
S. dysenteriae serotype 1 infection is commonly complicated
by hemolysis, anemia, and HUS. This syndrome is caused by
Shiga toxin-mediated vascular endothelial injury. E. coli that
produce Shiga toxins (e.g.,E. coli 0157:H7, E. coli 0111:NM,
E. coli 026:Hl1) also cause HUS (Chapter 512).
Rectal prolapse, toxic megacolon or pseudomembranous colitis
(usually associated with S. dysenteriae), cholestatic hepatitis, conjunctivitis,
iritis, corneal ulcers, pneumonia, arthritis (usually
2-5 wk after enteritis), reactive arthritis, cystitis, myocarditis, and
vaginitis (typically with a blood-tinged discharge associated with
S.flexneri) are uncommon events. Although rare, surgical complications
of shigellosis can be severe; the most common are intestinal
obstruction and appendicitis with and without perforation.
On average, severity of illness and risk of death are least with
disease caused by S. sonnei and greatest with infection by S.
dysenteriae type 1. Risk groups for severe illness and poor outcomes
include infants; adults >50 yr; children who are not breastfed;
children recovering from measles; malnourished children and
adults; and patients who develop dehydration, unconsciousness,
or hypo- or hyperthermia or have a history of convulsion when
first seen. Death is a rare outcome in well-nourished older children.
Multiple factors contribute to death in malnourished children
with shigellosis, including illness in the first year of life,
altered consciousness, dehydration, hypothermia, thrombocytopenia,
anemia, hyponatremia, renal failure, hyperkalemia hypoglycemia,
bronchopneumonia, and bacteremia.
The rare syndrome of severe toxicity, convulsions, extreme
hyperpyrexia, and headache followed by brain edema and a
rapidly fatal outcome without sepsis or significant dehydration
(Ekiri syndrome or "lethal toxic encephalopathy") is not well
understood.
DIFFERENTIAL DIAGNOSIS
Although clinical features suggest shigellosis, they are insufficiently
specific to allow confident diagnosis. Infection by Campylobacter
jejuni, Salmonella spp, enteroinvasive E. coli, Shiga
toxin-producing E. coli (e.g. E. coli 0157:H7), Yersinia enterocolitica,
Clostridium difficile, and Entamoeba histolytica as well
as inflammatory bowel disease can cause confusion.
DIAGNOSIS
Presumptive data supporting a diagnosis of bacillary dysentery
include the finding of fecal leukocytes (usually >50 or 100 PMNs
per high power field, confirming the presence of colitis), fecal
blood, and demonstration in peripheral blood of leukocytosis
with a dramatic left shift (often with more bands than segmented
neutrophils). The total peripheral white blood cell count is usually
5,000-15,000 cells/mm3, although leukopenia and leukemoid
reactions occur.
Culture of both stool and rectal swab specimens optimizes the
chance of diagnosing Shigella infection. Culture media should
include MacConkey agar as well as selective media such as
xylose-lysine deoxycholate (XLD) and SS agar. Transport media
should be used if specimens cannot be cultured promptly. Appropriate
media should be used to exclude Campylobacter spp and
other agents. Studies of outbreaks and illness in volunteers show
that the laboratory is often not able to confirm the clinical suspicion
of shigellosis even when the pathogen is present. Multiple
fecal cultures improve the yield of Shigella. The diagnostic inadequacy
of cultures makes it incumbent on the clinician to use
judgment in the management of clinical syndromes consistent
with shigellosis. Use of polymerase chain reaction (PCR) analysis
of stool for specific genes such as ipaH, lIirF, or lIirA can detect
cases not diagnosed by culture, but it is usually available only in
research laboratories. In children who appear to be toxic, blood
cultures should be obtained, especially in very young or malnourished
infants because of their increased risk of bacteremia.
TREATMENT
As with gastroenteritis from other causes, the first concern in a
child with suspected shigellosis should be for fluid and electrolyte
correction and maintenance (Chapter 332). Drugs that retard
intestinal motility (e.g., diphenoxylate hydrochloride with atropine
[Lomotil) or loperamide [Imodium)) should not be used
because of the risk of prolonging the illness.
Nutrition is a key concern in areas where malnutrition is
common. A high-protein diet during convalescence enhances
growth in the following 6 mo. A single large dose of vitamin A
(200,000 IU) lessens severity of shigellosis in settings where
vitamin A deficiency is common. Zinc supplementation (20 mg
elemental zinc for 14 days) has been shown to significantly
decrease the duration of diarrhea, improve weight gain during
recovery and immune response to the Shigella, and decrease diarrheal
disease in the subsequent 6 mo in malnourished children.
The next concern is a decision about the use of antibiotics.
Although some authorities recommend withholding antibacterial
therapy because of the self-limited nature of the infection, the
cost of drugs, and the risk of emergence of resistant organisms,
there is a persuasive logic in favor of empirical treatment of all
children in whom shigellosis is strongly suspected. Even if not
fatal, the untreated illness can cause a child to be quite ill for
weeks; chronic or recurrent diarrhea can ensue. Malnutrition can
develop or worsen during prolonged illness, particularly in children
in developing countries. The risk of continued excretion and
subsequent infection of family contacts further argues against the
strategy of withholding antibiotics.
Shigella species have variable antimicrobial susceptibility. In
general, S. flexneri tends to be more resistant than S. boydii.
There are major geographic variations in antibiotic susceptibility
of shigellae. In most developing countries and in some industrialized
countries such as the USA, Shigella strains are often resistant
to ampicillin and trimethoprim-sulfamethoxazole (TMP-SMX).
Therefore, these drugs should usually not be used for empirical
treatment of suspected shigellosis. Oral ampicillin (100 mg!
kg!24 hr orally, divided 4 times/day) or TMP-SMX (10 mg!
kg!24 hr orally of the TMP component in 2 divided doses) may
be used if the strain is known to be susceptible (e.g., in an outbreak
due to a defined strain). Amoxicillin is less effective than
ampicillin for treatment of ampicillin-sensitive strains. Ceftriaxone
(50 mg!kg!24 hr as a single daily dose IV or 1M) can be used
for empirical therapy, especially for small infants.
The oral 3rd-generation cephalosporin cefixime can also be
used. Oral 1st and 2nd-generation cephalosporins are inadequate
as alternative drugs despite in vitro susceptibility. Nalidixic acid
(55 mg!kg!24 hr orally divided 4 times/day) is also an acceptable
alternative drug when available. Azithromycin (12 mg!kg!24 hr
orally for the first day, followed by 6 mg/kg!24 hr for the next 4
days) has proven to be an effective alternative drug for shigellosis.
Ciprofloxacin (30 mg!kg!24 hr divided into 2 doses) used to be
a back-up drug to treat shigellosis but is now the drug of choice
recommended by WHO for all patients with bloody diarrhea,
irrespective of their ages.
Although quinolones have been reported to cause arthropathy
in immature animals, the risk of joint damage in children appears
to be minimal and is outweighed by the value of these drugs for
treatment of this potentially life-threatening disease. However,
some experts recommend that these agents be reserved for seriously
ill children with bacillary dysentery due to an organism that
is suspected or known to be resistant to other agents, because
overuse of quinolones promotes development of resistance to
these drugs. Treatment in general is for a 5-day course.
Treatment of patients in whom Shigella infection is suspected
on clinical grounds of should be initiated when they are first
evaluated. Stool culture is obtained to exclude other pathogens
and to assist in antibiotic changes should a child fail to respond
to empirical therapy. A child who has typical dysentery and who
responds to initial empirical antibiotic treatment should be continued
on that drug for a full 5-day course even if the stool culture
is negative. The logic of this recommendation is based on the during adult volunteer infection studies. In a child who fails to
respond to therapy of a dysenteric syndrome in the presence of
initially negative stool culture results, additional cultures should
be obtained and the child should be re-evaluated for other possible
diagnoses.
PREVENTION
PREVENTION
Numerous measures have been recommended to decrease the risk
of Shigella transmission to children. Mothers should be encouraged
to prolong breast-feeding of infants. Families and day care
personnel should be educated in proper handwashing techniques
and encouraged to wash hands after using the toilet, changing
diapers, or engaging in preparation of foods. They should be
taught how to manage potentially contaminated materials such
as raw vegetables, soiled diapers, and diaper-changing areas.
Children with diarrhea should be excluded from child care facilities.
Children should be supervised when handwashing after they
use the toilet. Caretakers should be informed of the risk of transmission
if they prepare food when they are ill with diarrhea.
Families should be educated regarding the risk of swallowing
contaminated water from ponds, lakes, or untreated pools.
There is not yet a vaccine that is effective for preventing infection
by Shigella. Several candidate vaccines are under development,
mostly against S. flexneri. Measles immunization can substantially reduce the incidence and severity of diarrheal diseases,
including shigellosis. Every infant should be immunized
against measles at the recommended age.