CLICK AQUI
Approach to the
patient with nontyphoidal Salmonella in a stool culture
Literature review current through: Jan
2013. | This topic last updated: ene 18, 2013.
INTRODUCTION — Salmonellae are frequently isolated from the stool of
patients with gastroenteritis. In a 2005 survey from FoodNet of
laboratory-confirmed causes of acute foodborne illnesses in ten states in the
United States, nontyphoidal Salmonella was the most commonly isolated pathogen,
occurring in 39 percent of cases, followed by Campylobacter in 34 percent [1].
Salmonella
typhimurium and S. enteritidis are the most frequently isolated serotypes of
Salmonella found in stool cultures in the United States. S. enteritidis, a
common cause of foodborne disease outbreaks [2], is most frequently associated with eggs or egg-containing products [3]. A nationwide epidemic of S. enteritidis infection in the United
States in 1994 was associated with contaminated ice cream [4]; another epidemic was associated with the ingestion of raw eggs [5].
Most
nontyphoidal Salmonella infections occur in otherwise healthy individuals.
However, a variety of host defense alterations result in increased
susceptibility to infection with Salmonella spp, including impaired cellular immunity
due to AIDS [6-8], corticosteroid use [9] or malignancy [10], and alteration in the intestinal flora due to prior antibiotic
therapy [11,12]. A number of other factors may also be important in selected patients
(table 1) [13-25]. These may result in more severe initial infection and more serious
sequelae such as bacteremia, metastatic foci of infection, or prolonged
infection.
The major
issues involved in the therapeutic approach to the finding of nontyphoidal
Salmonella in a stool culture will be reviewed here. A more detailed review of
the microbiology and epidemiology of salmonellosis and the microbiology,
epidemiology, pathogenesis, clinical manifestations, treatment, and prevention
of typhoid fever are presented separately.
CLINICAL MANIFESTATIONS — Gastroenteritis due to Salmonellae is clinically
indistinguishable from gastroenteritis caused by many other pathogens.
Furthermore, enteric infection with nontyphoidal Salmonellae may be clinically
mild or even asymptomatic, which can complicate clinical decisions about
treatment interventions.
The cardinal
features -- nausea, vomiting, fever, diarrhea, and cramping -- usually occur
within 8 to 72 hours of ingesting contaminated food or water. A higher ingested
dose of bacteria correlates with increases in the severity of diarrhea, the
duration of illness, and weight loss [26].
Although there
have been descriptions of clinical characteristics associated with
salmonellosis (eg, "pea soup" diarrhea), there are no clinical
characteristics that reliably distinguish Salmonella infection from other forms
of gastroenteritis.
Invasive disease — Fewer than 5 percent of individuals with documented
Salmonella gastroenteritis develop bacteremia [27,28]. Bacteremia can lead to a variety of extraintestinal manifestations
such as endocarditis, mycotic aneurysm, and osteomyelitis [29]. Two species, S. choleraesuis [30,31] and S. heidelberg [32,33], appear to be more invasive. In addition, antibiotic resistant strains
of S. typhimurium, are associated with a two- to threefold increase in the risk
of bacteremia [28,34].
Course — Nontyphoidal Salmonella gastroenteritis is usually
self-limited. Fever generally resolves within 48 to 72 hours, and diarrhea
within 4 to 10 days [5]. Diarrhea persisting more than 10 days should lead to consideration of
other diagnoses. Mortality rates of 0.5 to 1 percent have been reported in
outbreaks of S. enteritidis, but these are most likely overestimates since
milder cases tend to be unrecognized [28,35].
MICROBIOLOGY — Laboratory isolation of Salmonellae from stool cultures
usually requires a minimum of 48 hours; 72 hours is needed if overnight
enrichment broth incubation is used in addition to primary plating of stool
samples. Salmonellae are gram-negative, facultatively anaerobic
Enterobacteriaceae, which are differentiated from the normal gram-negative
flora of the intestinal tract, in part, by the color of the colonies on
indicator plates. The sensitivity and specificity of single or multiple stool
cultures for diagnosis of salmonellosis are unknown.
- Salmonella and Shigella are lactose-negative
organisms that form translucent colonies on MacConkey agar. Less than 1
percent of Salmonellae are lactose-positive, which can rarely cause
difficulties in identification.
- Most nontyphoidal Salmonellae produce hydrogen
sulfide, which causes colonies to appear light greenish with black centers
on Hektoen enteric agar or white with black centers on Salmonella-Shigella
agar.
Enrichment
broths (tetrathionate or selenite) are used to facilitate identification of
Salmonella or Shigella when low numbers of organisms are present. Overnight
incubation in these broths inhibits the growth of E. coli but not Salmonella or
Shigella.
Suspicious
colonies from plates are further evaluated biochemically, and subsequently
confirmed as Salmonella with commercial polyvalent antisera specific for
Salmonella O and Vi antigens. Simple grouping based upon O antigen is usually
reported initially before more complete serotyping is available. S. typhimurium
belongs to group B, and S. enteritidis and S. typhi belong to group D.
Salmonellosis is a reportable illness in the United States, and laboratories
are required to report isolates to state authorities.
TREATMENT CONSIDERATIONS — There are a number of clinical
settings in which therapy is considered after a stool culture for Salmonella
returns positive for nontyphoidal Salmonella. The discussion below will address
recommendations for the following settings:
- Treatment of symptomatic
patients
- Preemptive treatment and
special hosts
- Asymptomatic carriage of
nontyphoidal Salmonellae
- Food handlers and health care workers with
salmonellosis
Symptomatic patients — The cornerstone of therapy of symptomatic individuals with
Salmonella gastroenteritis is replacement of fluids and electrolytes. As noted
above, the illness is usually self-limited; in addition, the vast
majority of Salmonella infections are undiagnosed and are not associated with
complications.
However, the
lack of rapid diagnostic testing methods for enteric pathogens requires that
decisions about therapy be made empirically at the time patients present. The
issue of empiric antimicrobial therapy for infectious diarrhea in general
(including enterotoxigenic E. coli) is discussed elsewhere.
A number of
controlled trials have addressed the role of antimicrobial therapy in otherwise
healthy patients with Salmonella infection [36-38]. A meta-analysis of 12 trials with 767 otherwise healthy individuals
with nontyphoidal Salmonella gastroenteritis found no significant benefit from
antimicrobial therapy on the length of illness, diarrhea, or fever [39]. As a result, we do not recommend antibiotic treatment for
immunocompetent adults or children over 12 months of age with mild to moderate
symptoms of gastroenteritis from Salmonellae.
However,
treatment should be considered in a subset of severely ill immunocompetent
individuals on an individualized basis [38]. These include
patients with:
- Severe diarrhea (more than 9 or 10 stools per
day)
- High fever
- A need for hospitalization
The presence of
bloody diarrhea does not necessarily indicate the need for
antimicrobial treatment. Many patients with Salmonellosis have occult blood
detectable in stool samples, while overtly bloody stools are more likely to be
due to Shigella or enterohemorrhagic E. coli.
For individuals
with the above characteristics, three to seven days of antimicrobial therapy
has generally been recommended. This recommendation is based upon studies
suggesting that antimicrobial treatment in these settings may be associated
with improvement in symptoms and more rapid clinical recovery [40]. Studies of empiric antimicrobial therapy (before cultures are back)
in severe community-acquired diarrhea have also found a reduction of disease
duration by one to two days [41,42].
As a general
rule, the potential for improvement of severe illness and prevention of
complications (eg, bacteremia, focal infection or persisting symptoms) appear
to outweigh the small risks of antibiotic treatment, although this has not been
definitively demonstrated in large, randomized placebo-controlled trials.
Appropriate
antibiotic choices for adults with relatively normal renal function and no
known drug intolerances include a fluoroquinolone (eg, ciprofloxacin 500 mg orally twice daily or levofloxacin 500 mg orally once daily), trimethoprim-sulfamethoxazole (160 mg/800 mg orally twice daily), amoxicillin (500 mg orally thrice daily), or, if intravenous therapy is
required, a third generation cephalosporin (eg, ceftriaxone 1 to 2 g intravenously once daily or cefotaxime 2 g intravenously every eight hours).
Antibiotic
resistance patterns should be taken into account when making treatment
decisions. In 2003, the antimicrobial agents with the highest prevalence of
resistance among non-typhoidal Salmonellae weretetracycline, sulfamethoxazole, ampicillin, streptomycin (all in the range of 13 to 16 percent); these values had not
changed much over the preceding ten years. In contrast, the prevalence of
resistance to fluoroquinolones and third generation cephalosporins is
increasing; between 1996 and 2003, nalidixic acid resistance rose from 0.2
to 2.3 percent, and ceftiofur resistance rose from 0.2 to 4.5 percent [43].
Organisms with
multiple drug resistances are common, highlighting the need for sensitivity
testing.
Because of
their activity against most common gram-negative enteric pathogens and their
favorable side effect profile, the fluoroquinolones are probably the most appropriate
agent for most individuals without contraindications to these medications.
Although increasing data show that fluoroquinolones are safe in children [44], other agents are advisable initially for children or pregnant women
because of possible damage to growing cartilage.
Preemptive treatment — Preemptive therapy in an attempt to prevent complications
of Salmonella intestinal infection, such as bacteremia and metastatic foci of
infection, is considered in some patients, even with less severe infection.
Older adults
and the very young are at increased risk for complications. Neonates with
gastroenteritis, particularly those born prematurely and/or to mothers with
gastroenteritis, are at risk and may not appear toxic or acutely ill [33,45,46]. Salmonella meningitis is a feared complication with a high case
fatality rate that occurs primarily in infants. In most cases, infants less
than three months of age should receive antimicrobial therapy for symptomatic
salmonellosis [47]; in addition, some authorities treat all children less than one to two
years old unless the child is afebrile and clinically improving at the time the
culture results become available [45].
Adults over 50,
particularly, those with known atherosclerotic disease, are at higher risk of
endovascular infection and aortitis if bacteremia develops. Approximately 10
percent of adults over 50, who are identified on the basis of a nontyphoidal
Salmonella bacteremia, will be found to have infective arteritis [48,49]. Those with endovascular or osseous prostheses may reasonably be
considered in this group, and similarly treated even if not severely ill with
gastroenteritis.
The duration of
therapy in these groups at the extremes of age has not been studied. Most
experts would recommend 3 to 10 days of treatment depending upon the specific
clinical situation, suspicion of bacteremia, laboratory findings, and
microbiological data. It seems prudent to limit the exposure to antibiotics in
normal hosts, because of their potential to paradoxically prolong carriage (see
below) and adverse drug reactions, and concerns about antibiotic resistance in
general.
Preemptive
therapy is definitely warranted in immunocompromised patients, including those
with:
- An organ transplant
- AIDS
- Those receiving corticosteroids or other
immunosuppressive drugs
- Sickle cell disease, hemoglobinopathies,
disorders of the reticuloendothelial system, (eg, cirrhosis)
- Cancer or lymphoproliferative disease with
current or recent chemotherapy.
It is
impossible to know how many patients with these conditions who get mild to
moderate gastrointestinal salmonellosis go on to develop bacteremia and
complications. However, it is likely to be a significant percentage and a cause
for worry. In two reports (one from Malawi and the other from Taiwan), 31 and
43 percent of HIV-infected adults (who had not been treated with antiretroviral
therapy) developed recurrent Salmonella bacteremia [50,51].
Although the
above provides a rationale for preemptive therapy, its benefit has not been
documented in a well-designed prospective randomized trial and is unlikely to
be so tested. Similarly, the optimal duration of therapy has not been studied
with respect to persistent carriage, or other complications. Because Salmonellae
persist in the reticuloendothelial system, and these groups of patients have
impaired immune clearance mechanisms (especially those with advanced HIV or
organ transplants), a longer duration of therapy seems advisable.
Most experts
would treat such patients longer initially (eg, 14 days or even longer [weeks
to months]), in an effort to prevent persistent or relapsing infection. These
patients should be evaluated carefully for bacteremia, and treatment decisions
made in an individualized fashion. Consultation with an expert in infectious
diseases is recommended. The high tissue and intracellular concentrations
achieved by fluoroquinolones may make them particularly effective (eg, ciprofloxacin 500 mg orally twice daily or levofloxacin 500 mg orally once daily); other agents such as trimethoprim-sulfamethoxazole or beta-lactams would also be appropriate choices.
Pregnant women — Pregnant women acquire nontyphoidal Salmonella infection at
rates similar to the general population [52,53]. Pregnancy does not necessarily increase severity of illness, although
there are case reports of pregnant women with severe nontyphoidal Salmonella
infection, and fetal loss in the setting of disseminated infection is possible
in any trimester [54-56]. Pregnant woman with fever and gastroenteritis symptoms should be
evaluated for salmonellosis as well as other enteric bacterial infections
(including Campylobacter and Listeria).
Routine
treatment of salmonellosis in pregnancy is not necessary, particularly in the
setting of mild illness early in pregnancy; antibiotics may not speed
resolution or mitigate symptoms, and may prolong Salmonella carriage.
Antibiotic
therapy for treatment of salmonellosis in pregnancy is appropriate for febrile
patients with severe disease, especially in the setting of bacteremia or
infection near term. Women with peripartum infection may shed organisms during
childbirth with high risk for neonatal transmission of infection, although severity
of illness does not necessarily correlate with persistence of infection or
duration of shedding [57]. If severe illness occurred early in pregnancy, there is no role for
additional treatment at the time of delivery.
Prophylaxis of
neonates may be appropriate in some situations to prevent complications of
systemic neonatal salmonellosis, including meningitis [58,59]. In addition, outbreaks of salmonellosis in NICUs and maternity wards
neonates have been described [60,61].
Infectious
disease consultation is advisable to assist with management and follow-up
evaluation; these must be individualized depending on clinical circumstances
including the patient, the organism, and presence of antibiotic resistance.
Asymptomatic carriage — Asymptomatic convalescent excretion of Salmonella bacilli
is extremely common after either symptomatic or asymptomatic nontyphoidal
Salmonella infection [62]. The influence of antimicrobial therapy on Salmonella carriage is
controversial. The conflicting data are due in part to differences in culture
techniques, culture intervals, infecting strains and duration of follow-up. Nevertheless, several general statements can be made:
- The median duration of excretion is
approximately five weeks for all age groups, with S. typhimurium being
more rapidly cleared than other serotypes [62]
- The median duration of excretion is
approximately seven weeks in patients less than five years of age, 2.6
percent of whom excrete Salmonellae for more than one year [62]
- The duration of excretion may be longer after
symptomatic than asymptomatic infection
- Intermittent shedding is common, so a single
negative culture is not that reassuring.
- Fecal cultures are generally considered the
"gold standard" for diagnosis of Salmonella gastroenteritis and
should not be difficult to obtain if people are having diarrhea. However,
rectal swabs were moderately useful (sensitivity 64 percent and
specificity 90 percent) compared with fecal cultures in an inpatient study
of a live attenuated S. typhimurium vaccine [63].
Given these
data, routine follow-up cultures are not recommended after uncomplicated
Salmonella gastroenteritis in immunocompetent patients.
Studies
performed in the prequinolone era demonstrated that antimicrobial therapy of
Salmonella gastroenteritis prolonged the duration of Salmonella excretion. This
paradoxical finding has been thought to be due to the deleterious effect of
antibiotics on normal intestinal flora, which protects against colonization
with enteric pathogens. Prolongation of carriage is not related to the
emergence of organisms resistant to the antimicrobial agent used, which is a
relatively rare event.
In a 1980 study
of children and infants treated with five days of ampicillin, amoxicillin, or placebo for uncomplicated gastroenteritis, 53 percent who received
antibiotics had bacteriologic relapse versus none who received placebo [64]. Of those who relapsed, 38 percent had recurrent diarrhea.
It was hoped
that the fluoroquinolones would be more effective in eliminating convalescent
carriage. This hypothesis has been evaluated in two, small, randomized,
placebo-controlled treatment studies of outbreaks of Salmonella gastroenteritis
using norfloxacin (400 mg twice daily for seven days) [65] or ciprofloxacin (750 mg twice daily for 14 days) [66]. Neither of the treatment regimens in these studies decreased or
significantly prolonged the duration of carriage. In contrast, a deleterious
effect of norfloxacin was noted in the patients with salmonellosis in the
Swedish study described above [38]. In this study, the incidence of negative cultures at 7 to 12 days was
significantly lower in the norfloxacin group (18 versus 49 percent with
placebo). The median time to negative cultures was also prolonged by
norfloxacin (50 versus 23 days).
In summary,
brief early courses of antibiotics, including the fluoroquinolones, are not effective
in shortening the duration of carriage of nontyphoidal Salmonellae. However,
antimicrobial therapy or prophylaxis does appear to be effective to prevent
infection or suppress excretion of large numbers of organisms acutely. Drug
treatment may occasionally be useful in limiting epidemics in closed settings
in which full compliance with infection control measures may be difficult, such
as pediatric wards [67] or long-term care facilities [68].
Paradoxically, long-term
antimicrobial therapy may be of some utility in patients who have prolonged or
chronic carriage. Chronic carriage of nontyphoidal Salmonella is defined as the
shedding of a Salmonella species for more than one year, as documented by an
initial positive culture of a stool sample obtained at least one month after
resolution of the acute illness and repeat positive cultures [69]. Prolonged carriage appears to occur rarely (0.2 to 0.6 percent in one
study) [70] in otherwise healthy subjects but may be increased in older
individuals, women, and those with biliary tract abnormalities. If chronic
carriage is documented, attempted eradication may rarely be needed for
employment or other social reasons like daycare attendance or the presence of a
severely immunosuppressed family member (see below).
The optimal
approach to chronic carriage of nontyphoidal Salmonella is poorly studied [62]. Drawing upon the experience with treating carriage of Salmonella
typhi, four to six weeks of antimicrobial therapy is a reasonable approach,
although many patients may have side effects related to treatment. When
attempting eradication of carriage we use a fluoroquinolone (eg, ciprofloxacin 500 mg orally twice daily,levofloxacin 500 mg orally once daily, or norfloxacin 400 mg orally twice daily) for four weeks. Alternative regimens,
which have successfully eradicated S. typhi in some patients, include trimethoprim-sulfamethoxazole (160 mg/800 mg orally twice daily) for three months or ampicillin or amoxicillin (3 to 5 g orally in four divided doses) for six weeks (depending
upon sensitivities of the isolate).
Guidelines for
evaluating the efficacy of treatment of carriage have been suggested and
include follow-up cultures six months after completing therapy [69]. In addition, suppressive prophylactic therapy may be warranted in
HIV-positive individuals, particularly, in those with a septicemic primary
illness or low CD4 cell counts, in order to prevent relapse [71].
Food handlers and health care workers — Large
numbers of asymptomatic transient Salmonella excreters are undoubtedly employed
in the health care and food industries worldwide for the following reasons:
- The initial episode of Salmonella
gastroenteritis may be clinically mild and remain undiagnosed
- The median duration of Salmonella excretion
after gastroenteritis is five weeks [62]
- Excretion is frequently
episodic [62,72]
Nevertheless,
only two percent of 566 outbreaks of salmonellosis in the United Kingdom were
traced to specific infected food handlers [73]. Furthermore, transmission from asymptomatic food handlers has rarely
been documented [74,75]. One of these outbreaks in Amman, Jordan occurred despite routine
surveillance of kitchen employees for Salmonella carriage [75]. The authors concluded that screening was not cost-effective and
should not be used as a substitute for health education and proper hygienic
practices. A Thai study of asymptomatic food handlers shedding Salmonella
randomized people to five days of norfloxacin,azithromycin, or placebo. The study drugs were no better than placebo in eradicating
shedding in this area endemic for nontyphoidal salmonellosis. Also, in this
endemic setting, selection of drug resistant Salmonella was demonstrated [76].
Nosocomial
fecal-oral transmission can occur, and is related to poor handwashing practices
[77,78]. The importance of handwashing was illustrated in a foodborne outbreak
of salmonellosis among nurses; there was no transmission to patients,
presumably because of adherence to proper handwashing procedures [79].
These studies
highlight the primacy of good hygiene and handwashing in control of
salmonellosis from infected health care workers or food handlers. It is
reasonable and customary to exclude such individuals from work while
gastroenteritis symptoms persist [80], since organism excretion is more common and at higher levels during
active disease. Some states require one or more negative stool cultures (more
than 48 hours after discontinuation of antibiotics, if given) before employees
can return to work. A similar approach may be locally mandated, or instituted
during outbreaks in day care centers, but scrupulous attention to proper
infection control procedures are most important.
While
reassuring, a single negative stool culture does not ensure that the individual
is no longer shedding [62]. Obtaining multiple negative stool cultures from employees may be
warranted occasionally in individuals whose work requires direct handling of
food consumed without further cooking, or direct contact with infants or
immunosuppressed patients. However, this approach does not seem reasonable
routinely for employees with lower risk occupations. As national policy for
these settings is not mandated, judgments based upon individual employees and
local health administration standards are appropriate.
INFORMATION FOR PATIENTS — UpToDate offers two types of patient
education materials, “The Basics” and “Beyond the Basics.” The Basics patient
education pieces are written in plain language, at the 5th to 6th grade
reading level, and they answer the four or five key questions a patient might
have about a given condition. These articles are best for patients who want a
general overview and who prefer short, easy-to-read materials. Beyond the
Basics patient education pieces are longer, more sophisticated, and more
detailed. These articles are written at the 10th to 12th grade
reading level and are best for patients who want in-depth information and are
comfortable with some medical jargon.
Here are the
patient education articles that are relevant to this topic. We encourage you to
print or e-mail these topics to your patients. (You can also locate patient
education articles on a variety of subjects by searching on “patient info” and
the keyword(s) of interest.)
REFERENCES
No hay comentarios:
Publicar un comentario