Human Salmonella infections are common; most infections are self-limiting but severe disease may occur. Fluoroquinolones and third-generation cephalosporins are the drugs-of-choice for invasive Salmonella infections in humans. Little correlation is noted between the antimicrobial resistance patterns of isolates collected from persons with Salmonella infections and antimicrobial agents used for the treatment of Salmonella infections in humans. Direct evidence is available which demonstrates that antimicrobial resistant Salmonella results from the use of antimicrobial agents in food animals and these antimicrobial resistant Salmonella are subsequently transmitted to humans. Because of the adverse health consequences in humans and animals associated with the increasing prevalence of antimicrobial resistant Salmonella, there is an urgent need to emphasize non-antimicrobial strategies, such as improved sanitation and hygiene, to develop guidelines for the prudent usage of antimicrobial agents, and to restrict the use of fluoroquinolones in food animals.
Much of the discussion about the adverse human health effects associated with the veterinary use of antimicrobial agents has been clouded by confusion surrounding the clinical significance and sources of antimicrobial-resistant nontyphoidal Salmonella infections in humans (Institute of Medicine, 1989; U.S. Congress, Office of Technology and Assessment, 1995). This issue needs to be revisited in the light of the recent discussions concerning the public health implications of veterinary use of fluoroquinolones (Anonymous, 1994; Beam, 1994), a class of antimicrobials essential for the treatment of several life-threatening infections in humans (Wilcox and Spencer, 1992; Conte, 1995). Because of these public health concerns, the Food and Drug Administration prohibited the extra-label use of fluoroquinolones in food animals in the United States in August, 1997. Two fluoroquinolones, enrofloxacin and sarafloxacin, are approved, however, for use in poultry in the United States. To address the human health implications of the veterinary use of fluoroquinolones, we reviewed the medical literature and data available at the Centers for Disease Control and Prevention (CDC) about the epidemiology of human i infections. In this report, we discuss the clinical significance of antimicrobial-resistant Salmonella infections in humans, the epidemiology of antimicrobial-resistant pathogens including Salmonella, the sources of Salmonella (including antimicrobial-resistant Salmonella) infections in humans, the causes and consequences of development of antimicrobial-resistant Salmonella, and suggest necessary actions to protect the public's health.
Clinical Significance of Antimicrobial-resistant Salmonellae
Salmonellosis results in considerable
human morbidity in the United States. Although most human Salmonella
infections result in a mild, self-limiting gastrointestinal illness
characterized by diarrhea, fever and abdominal cramps, the infection
can spread to the bloodstream, meningeal linings of the brain,
or other deep tissue sites, leading to a severe and occasionally
fatal illness. Each year, there are an estimated two to four
million human Salmonella infections in the United States (Chalker
and Blaser, 1988; Council for Agricultural Science and Technology,
1994), causing an estimated 80,000 to 160,000 persons to seek
medical attention. Clinical specimens collected from some of
the persons who have sought medical attention result in approximately
40,000 culture-confirmed cases a year reported to CDC (Hargrett-Bean
et al., 1988; Centers for Disease Control and Prevention:
Salmonella Surveillance, 1996). Each year in the United
States, an estimated 8,000 to 18,000 persons are hospitalized
and 500 persons die of Salmonella infections (Cohen and
Tauxe, 1986).
Antimicrobial agents are not essential for the treatment of most
Salmonella infections which manifest as uncomplicated
gastroenteritis because such infections usually are self-limiting,
and treatment may prolong the carrier state, and may result in
the emergence of a resistant infection in the treated person
(Wilcox and Spencer, 1992; Conte, 1995). Antimicrobial agents
are, however, commonly prescribed for persons with Salmonella
infections who seek medical attention. In surveys conducted by
CDC in 1990 (Lee et al., 1994) and 1995, 40% of persons
with Salmonella infections who sought medical attention
were treated with antimicrobial agents. Ciprofloxacin, a fluoroquinolone
antimicrobial agent, was the most commonly prescribed antimicrobial
agent for Salmonella infections. Ciprofloxacin, which
became available for oral use in humans in the United States
in 1988, was used by approximately 25% of persons who received
antimicrobial agents in the 1990 survey and 33% in the 1995 survey,
suggesting that >100,000 persons with Salmonella infections
have been treated with ciprofloxacin in the past 10 years in
the United States (between 2-3 million prescriptions of ciprofloxacin
are dispensed annually, for a variety of conditions, in the United
States).
In contrast to patients with uncomplicated gastroenteritis, effective
antimicrobial agents are essential for the treatment of patients
with bacteremia, meningitis, or other extra intestinal Salmonella
infections (Wilcox and Spencer, 1992; Conte, 1995). In approximately
six percent of the culture-confirmed cases reported to CDC, Salmonella
is isolated from specimens collected from extra-intestinal sites
- usually from blood (Hargrett-Bean et al., 1988; Centers
for Disease Control and Prevention: Salmonella Surveillance,
1996). Since approximately 40,000 culture-confirmed cases are
reported to CDC each year, effective antimicrobial agents are
critical and may be life-saving for at least 2,400 persons a
year in the United States. Unfortunately, the selection of antimicrobial
agents for the treatment of invasive infections has become increasingly
restricted due to increasing antimicrobial resistance among Salmonella
isolates. In the past, chloramphenicol, ampicillin, and trimethoprim-sulfamethoxazole
were used to treat Salmonella infections (Riley et
al., 1984; McDonald et al., 1987; Lee et al., 1994).
However, among 1,272 randomly selected Salmonella isolates from
humans tested at CDC in the National Antimicrobial Resistance
Monitoring System in 1996, 21% were resistant to ampicillin,
10% to chloramphenicol, and 4% to trimethoprim-sulfamethoxazole.
(Centers for Disease Control and Prevention. CDC/FDA/USDA National
Antimicrobial Monitoring System 1996 Annual Report). In contrast,
almost all of the Salmonella isolates tested at CDC have
been susceptible to fluoroquinolones and third-generation cephalosporins
(Centers for Disease Control and Prevention. CDC/FDA/USDA National
Antimicrobial Monitoring System 1996 Annual Report; Herikstad
et al., 1997a; Herikstad et al., 1997b). For
this reason, and because of the favorable pharmacodynamic properties
of these antimicrobial agents, fluoroquinolones and third-generation
cephalosporins are the drugs-of-choice for the treatment of invasive
Salmonella infections in adults and children, respectively.
Should Salmonella develop antimicrobial resistance to
these two antimicrobial agents, suitable alternative antimicrobial
agents are not currently available and adverse human health consequences,
including prolonged hospitalizations and increased frequency
of treatment failures, which may result in deaths, are expected.
While reviewing the clinical significance of antimicrobial-resistant
Salmonella infections, it is useful to evaluate the correlation
between the antimicrobial agents used to treat persons with Salmonella
infections and antimicrobial-resistance among human Salmonella
isolates (Figure 1). Information provided from surveys conducted
by CDC within selected counties in the United States in 1985
(McDonald et al., 1987), 1990 (Lee et al., 1994)
and 1995 indicates that the proportion of persons with a Salmonella
infection receiving an antimicrobial agent who were treated with
ampicillin declined from 60% in 1985 to 5% in 1995, while the
proportion of isolates resistant to ampicillin steadily increased.
The proportion of persons treated with trimethoprim-sulfamethozone,
in contrast, remained constant while trimethoprim-sulfamethozone
resistance increased slightly. Most significantly, the proportion
of patients with salmonellosis treated with ciprofloxacin or
extended-spectrum cephalosporins markedly increased without an
emergence of resistance to either of these antimicrobial agents
among human Salmonella isolates. The continued susceptibility
of human Salmonella isolates to fluoroquinolones and extended-spectrum
cephalosporins was confirmed in over 4,000 isolates in 1995 (Herikstad
et al., 1997a; Herikstad et al., 1997b) and
1,200 isolates in 1996 (Centers for Disease Control and Prevention.
CDC/FDA/USDA National Antimicrobial Monitoring System 1996 Annual
Report). Taken together, these data suggest there is little correlation
between the antimicrobial agents used in persons with Salmonella
infections and development of antimicrobial resistance among
human Salmonella isolates. If human antimicrobial use
is not associated with the increasing antimicrobial resistance
seen among Salmonella isolates, what is causing the increasing
prevalence of antimicrobial-resistance observed among Salmonella
isolates? Prior to addressing that question, it is useful to
review the epidemiology of antimicrobial-resistant pathogens.
Epidemiology of Antimicrobial-resistant Pathogens, Including Salmonella
Antimicrobial agents are used to treat microbial infections in humans, plants and animals; they are given prophylactically to healthy humans, plants and animals to prevent infections, and they are given in low doses to food animals to improve their growth rate and feed conversion. There is a preponderance of evidence that the use of antimicrobial agents, at subtherapuetic or therapeutic concentrations, results in antimicrobial resistance. However, in order for an antimicrobial-resistant pathogen to have a public health consequence, there must be both usage of the antimicrobial agent and dissemination of the resistant pathogen. The role of dissemination can be illustrated by observing the consequences of using fluoroquinolones for the treatment of three different infections: methicillin-resistant Staphylococcus aureus (MRSA) in humans, Salmonella in humans, and Salmonella in food animals. Fluoroquinolones have been widely used in the United States for the treatment of human MRSA and Salmonella infections, but since they have not been widely used for the treatment of Salmonella infections in food animals in this country, this latter instance will be described using events which have occurred in the United Kingdom (it is reasonable to assume such events would occur in the United States under similar conditions of fluoroquinolone usage).
> The emergence of human MRSA infections in hospitals in the United States is a major public health concern (Panlilio et al., 1992). When the first fluoroquinolone (ciprofloxacin) was approved in the United States for human use there was optimism that ciprofloxacin, which was highly effective against MRSA, might lessen the public health impact of MRSA infections. Shortly after the approval of ciprofloxacin, it became widely used for the treatment of human MRSA infections. With the widespread usage, human infections with ciprofloxacin-resistant MRSA rapidly emerged in the United States; by 1991-1992, 85% of MRSA isolates from hospitals in the National Nosocomial Infection Surveillance System were resistant to ciprofloxacin (Coronado et al., 1995). The rapid emergence of ciprofloxacin-resistance MRSA is clearly related to both the increased use of ciprofloxacin in humans and the efficiency of dissemination, via person-to-person transmission, of ciprofloxacin-resistant MRSA in hospitals.As indicated earlier, ciprofloxacin has also been widely used for the treatment of Salmonella infections in humans. Because antimicrobial-resistance follows antimicrobial usage, Salmonella with decreased susceptibility to ciprofloxacin may have developed in persons with salmonellosis who were treated with ciprofloxacin, however, because person-to-person transmission of Salmonella is rare, Salmonella with decreased susceptibility to ciprofloxacin has not been disseminated in the United States. This is in sharp contrast to the situation in the United Kingdom, where a fluoroquinolone (enrofloxacin) has been widely used in food animals following its approval for veterinary use in 1993. Following the approval of enrofloxacin for veterinary use in the United Kingdom, decreased susceptibility to fluoroquinolones (Minimum Inhibitor Concentration [MIC] >0.25) rapidly emerged among human Salmonella isolates, particularly among isolates of multiply-resistant S. serotype Typhimurium DT104 which were resistant to ampicillin, chloramphenicol, streptomycin, sulfonamides, and tetracycline (R-type ACSSuT). DT014 R-type ACSSuT has emerged as the second most common strain of Salmonella isolated from humans. In 1993, none of the DT104 R-type ACSSuT isolates had a decreased susceptibility to fluoroquinolones; by 1996, 14% had a decreased susceptibility (Threfall et al., 1997). A parallel rapid emergence of decreased susceptibility to fluoroquinolones, as indicated by antimicrobial resistance to nalidixic acid (resistance nalidixic acid, a quinolone, indicates decreased susceptibility to fluoroquinolones) has been observed among animal S. Typhimurium DT104 R-type ACSSuT isolates at the central veterinary diagnositic laboratory in the United Kingdom.
Sources of Antimicrobial-resistant Salmonella Infections
Salmonella live in the intestines of mammals, birds and reptiles. Once shed into the environment in the feces of infected animals, Salmonella may survive for long periods in water, soil, and on or within foods. Although Salmonella infections occur commonly in humans, person-to-person transmission of Salmonella is uncommon in the United States. In less developed countries, in contrast, person-to-person transmission may be the source for a greater proportion of human Salmonella infections and nosocomial sources of antimicrobial-resistant Salmonella have been identified (Cherubin, 1981). Historically, before the development of hospital infection control procedures in the United States, nosocomial outbreaks of salmonellosis with person-to-person transmission, particularly in newborn nurseries, were not infrequent (Tauxe, 1986).
Most human Salmonella infections in the United States occur from the ingestion of contaminated food and many of these foods are foods of animal origin. Direct fecal-oral transmission following contact with animal feces is another, less common, source of human salmonellosis. At least six lines of evidence can be presented which, taken together, demonstrate that foods of animal origin are the dominant source of human salmonellosis, and suggest that person-to-person transmission is an uncommon source of human salmonellosis in the United States.
Although the prevalence varies, Salmonella is frequently isolated from the feces of food animals, companion animals, and wild animals. In longitudinal studies, some animals may excrete Salmonella for long periods of time. For example, many birds, including poultry, are infected with Salmonella and shed the organism in their feces (an estimated 20% of retail packages of poultry are contaminated with Salmonella). Fecal excretion of Salmonella by humans, in contrast, is relatively uncommon among apparently healthy individuals and is fairly short-lived among persons with salmonellosis. A review of several surveys of stool specimens from apparently healthy persons found a median carriage rate of Salmonella of 0.15% (Buchwald and Blaser, 1988). A review of several outbreak investigations determined the median duration of excretion by persons with salmonellosis to be about four weeks (Chalker and Blaser, 1988).
Volunteer studies among healthy adults suggest that a large oral infectious dose of Salmonella (>106 organisms) given in water usually is necessary to cause infection in a high proportion of recipients (McCullough and Eisele, 1951). These observations are supported by the relative low frequency of secondary illness within households in which a primary culture-confirmed case is identified and the rarity of day-care center outbreaks of salmonellosis; such settings commonly result in person-to-person transmission of enteric pathogens with low infectious doses (e.g., Shigella and Escherichia coli O157). The infrequent occurrence of secondary infections of salmonellosis and day-care center associated outbreaks suggest that person-to-person transmission of Salmonella occurs infrequently.
Although outbreaks only represent a fraction of the cases of Salmonella infections which occur, much insight into the epidemiology of salmonellosis has been provided through investigations of outbreaks. Most outbreak investigations are conducted by state or local health departments who report foodborne disease outbreaks to CDC as part of the Foodborne Disease Outbreak Surveillance System (Bean et al., 1996). A small number of investigations are conducted each year by CDC in collaboration with state and local health departments.
Between 1988 and 1992, an average of 110 outbreaks of Salmonella were reported each year to CDC (Bean et al., 1996). Sixty percent of these outbreaks were caused by Salmonella serotype Enteritidis and most of these were attributed to eating undercooked eggs. Many of these egg-associated outbreaks were traced back to their farm of origin and it was demonstrated that infected hens were the source of the outbreak. Among outbreaks caused by Salmonella serotypes other than Enteritidis, a variety of food items were implicated, particularly other foods of animal origin. A small number of fresh fruits and vegetables were also implicated. A few of these outbreaks were traced back to their farm of origin. Although some outbreaks involved infected food handlers, rather than being evidence of person-to-person transmission of Salmonella, in most cases, the food handlers probably became infected because they also ate the contaminated foods. Taken together, outbreak investigations demonstrate that foods, particularly foods of animal origin, are an important source of Salmonella infections in humans.
4. Sporadic Case-Control Studies
Further evidence that foods of animal origin are associated with
many human Salmonella infections is provided from investigations
of persons with sporadic Salmonella infections (infections
that were not recognized to be associated with an outbreak).
Few such investigations have been reported, however, perhaps
because investigations of sporadic infections are less likely
to implicate a common source because several sources may be involved.
Case-control studies of sporadic cases of Salmonella Enteritidis
were conducted in New York in 1989 (Morse et al., 1994),
California in 1994 (Passaro et al., 1996), and Utah in
1995 (Centers for Disease Control and Prevention. EPI-AID 96-16,
July 1996). These studies each demonstrated that eating raw or
undercooked eggs was the most important risk factor for acquiring
infections. Case-control studies of sporadic Salmonella
cases involving serotypes other than Enteritidis include an investigation
in Switzerland in 1996 of infections caused by a variety of serotypes
which implicated eggs as the most important sources of infections
(Schmind et al., 1996), California in 1984 of infections
again caused by a variety of serotypes which implicated poultry
(Kass et al., 1992), California in 1985 of Salmonella
Dublin which implicated raw milk (Richwald et al., 1988),
and a study of Salmonella Typhimurium and Enteritidis
infections in Minnesota in 1989 and 1990 which found eggs as
the most important source for both infections (Hedberg et
al., 1993).
After identification of the particular Salmonella serotype, several procedures (e.g., phage typing, pulsed-field gel electrophoresis, plasmid profiling, ribotyping) may be used to further differentiate Salmonella isolates (Threfall et al., 1994). Such subtyping techniques may be useful in epidemiological investigations to support or refute the postulated source of the outbreak. For example, in the 18 outbreaks of Salmonella Enteritidis in 1990 and 1991 in which eggs were implicated as the source, trace backs and environmental investigations on the implicated farms led to the detection of the human outbreak strain of S. Enteritidis, as determined by phage type, from the environment (100%) and from internal organs (88%) of implicated flocks strongly suggesting that the implicated farms were the sources of the outbreaks (Altekruse et al., 1993). Another example is provided from Denmark where phage-typing, and plasmid profiling of Salmonella Typhimurium isolates from human and animal sources showed some animal strains and human strains to be indistinguishable and concluded that spread of the strains from animals to humans was the most probable explanation (Seyfarth et al., 1997).
6. Emergence of Usual Strains in Humans
Monitoring of human Salmonella surveillance data, when supported
by serotyping and perhaps additional subtyping techniques, can
enable the detection of the emergence of a unusual strains of
Salmonella. Possible sources for an increased number of
an unusual strain of Salmonella among human isolates may
sometimes be indicated by the emergence of the same unusual strain
among isolates from animals, foods, and other sources. When such
investigations have been conducted, often the source of the increase
has been traced to foods of animal origin. For example, beginning
in 1969 there was a marked increase in human isolates of Salmonella
Agona detected in the United States and several other
countries. Salmonella Agona had not been isolated
in the United States before 1969, but by 1972 it was the eighth
most common serotype isolated from humans in the United States
(Glynn et al., in press). Field investigations and surveillance
data determined Peruvian fish meal fed to chicken to be the source
of the infections. Critical to the investigation was the identification
of Salmonella Agona from Peruvian fish meal in
routine surveillance sampling of fish meal in 1970 (Clark et
al, 1973).
The widespread geographic distribution of unusual strains also
supports a limited role for person-to-person transmission of
Salmonella in the developed world. For example, the almost
simultaneous emergence in the United States and Europe of Salmonella
Agona, and more recently an indistinguishable clone of
multi-drug-resistant Salmonella Typhimurium DT104
R-type ACSSuT (Glynn et al., in press), suggests transmission
via the contamination of a widely distributed vehicle, such as
food, rather than infected persons.
Although comparisons between human and animal Salmonella
surveillance data are useful in investigating the epidemiology
of salmonellosis, such comparison should consider the specimen
collection practices inherent in the submission of specimens
to clinical laboratories within each surveillance system. For
example, for both human and animal isolates, the specimens submitted
to the clinical laboratories usually are collected from ill individuals.
Since the serotypes of Salmonella in ill animals and in
foods of animal origin, which come from apparently healthy animals,
are likely to be different, crude comparisons of the "top
ten" Salmonella serotypes in humans and in animals
can lead to the erroneous conclusion that Salmonella serotypes
which are common in certain animals (e.g., Salmonella
serotype Cholerasuis in swine) and rare in humans are
nonpathogenic to humans. Thus, comparisons of the most common
serotypes in certain animals and humans can not be used to conclude
that food animals are not the most common sources of certain
serotypes.
Sources of Antimicrobial-resistant Salmonella Infections
Since most human Salmonella infections in the United States are acquired from ingestion of contaminated foods, and because most stool specimens which yield Salmonella (including antimicrobial-resistant Salmonella) are obtained from patients before the patient takes antimicrobial agents (if the patient takes antimicrobial agents), it follows that most antimicrobial-resistant Salmonella infections are acquired from ingestion of foods contaminated with antimicrobial-resistant Salmonella. Another, less common, source of antimicrobial-resistant Salmonella is direct fecal-oral transmission following contact with animal feces. Strong supporting evidence that, in the United States, persons infected with antimicrobial-resistant Salmonella rarely obtain their infection from other infected persons is provided by the inability to identify fluoroquinolone-resistant among 4,000 Salmonella isolates in 1995 - since ciprofloxacin is widely used for the treatment of patients with salmonellosis, it is reasonable to assume the fluoroquinolone-resistant Salmonella emerged in the intestinal tract of some of the individuals infected with Salmonella who received ciprofloxacin, however, transmission to other persons must be rare because no domestically acquired resistant infections were detected.
If antimicrobial-resistant Salmonella infections are acquired from the ingestion of foods contaminated with antimicrobial-resistant Salmonella, what causes the emergence and increasing prevalence of antimicrobial-resistant Salmonella? The emergence and increasing prevalence of antimicrobial resistant Salmonella is the direct result of antimicrobial agents usage. In the United States, antimicrobial agents are mostly used in humans, animals and on plants. Since human usage in the United States has little impact on resistance among Salmonella, and because most persons infected with antimicrobial-resistant Salmonella do not have a history of recent international travel (Riley et al., 1984; McDonald et al., 1987; Lee et al., 1994) and few antimicrobial agents are used on plants, the only likely cause for the emergence and increasing prevalence of antimicrobial-resistant Salmonella in the United States is the use of antimicrobial agents in animals, predominately food animals. Four lines of evidence support the conclusion that most antimicrobial-resistance among Salmonella isolates in humans results from the use of antimicrobial agents in food animals.
1. Trace Backs of Selected Foodborne Disease Outbreaks
Several outbreak investigations of antimicrobial-resistant Salmonella infections in humans have combined epidemiologic fieldwork and laboratory subtyping techniques to trace back antimicrobial-resistant Salmonella through the food distribution system to farms, and antimicrobial use on the farms was found to be associated with the antimicrobial resistance (Holmberg et al., 1984; Lyons et al., 1985; Tack et al., 1985; Spika et al., 1987). In one investigation, hamburgers contaminated with antimicrobial-resistant Salmonella were traced, using a unique plasmid profile, from supermarkets, through meat processing, to well beef cattle which had been feed subtherapuetic antimicrobial agents (Holmberg et al., 1984). In another investigation, approximately 1,000 persons were infected by hamburger contaminated with antimicrobial-resistant Salmonella serotype Newport with an unusual marker - chloramphenicol resistance. Chloramphenicol-resistant S. Newport was traced from ill persons, through processing, to dairy cattle on farms where chloramphenicol had been used (Spika et al., 1987). Although such investigations provide considerable insight into the complexity of Salmonella transmission, they suffer from the limitations of epidemiology studies. However, when combined with other lines of evidence, such investigations illustrate the potential human health consequences of the use of antimicrobial agents on farms.
2. Emergence of Salmonella Typhimurium DT104 R-type ACSSuT with Decreased Susceptibility to Fluoroquinolones in the United Kingdom
The continued emergence of S. Typhimurium DT104 with decreased susceptibility to fluoroquinolones in humans in the United Kingdom provides increasingly strong evidence that antimicrobial-resistance among Salmonella isolates in humans results from the use of antimicrobial agents in food animals. Decreased susceptibility to fluoroquinolones among human Salmonella isolates was rare in the United Kingdom prior to 1993, despite the widespread use of ciprofloxacin in humans since 1987. As previously mentioned, following the 1993 approval and widespread use of enrofloxacin in veterinary medicine, human Salmonella isolates (and animal isolates) with decreased susceptibility to ciprofloxacin rapidly emerged beginning in 1994 (Threfall et al., 1997).
3. Comparison of Patterns of Antimicrobial Resistance Patterns of Salmonella Isolates from Humans and Animals
If veterinary use of antimicrobial agents is responsible for the development of antimicrobial resistant Salmonella in animals which may be transmitted to humans, then the patterns of antimicrobial resistance observed among Salmonella isolates collected from healthy animals and humans should be similar. These similarities become most evident when focusing on a serotype of Salmonella in humans which are predominately derived from a single animal source. For example, human infections with S. serotype Heidelberg are often associated with eating undercooked chicken. Resistance patterns of S. Heidelberg isolates from humans and healthy chickens are similar (Figure 2).
4. Comparison Patterns of Antimicrobial Usage in Humans and Animals with Antimicrobial Resistance Patterns Among Humans and Animals
Although limited data is available on antimicrobial agent usage (subtherapuetic and therapeutic) in food animals, the available data suggests that the patterns of antimicrobial agent usage in food animals are similar to the spectrum of antimicrobial resistance observed among Salmonella isolates from food animals and humans. In contrast, the patterns of antimicrobial agent usage in humans are dissimilar to the spectrum of antimicrobial resistance observed in humans (Figure 3).
Human Health Risks of Fluoroquinolone Use in Food Animals
With the recognition that foods of animal origin are the source of most human Salmonella infections (and most antimicrobial-resistant Salmonella infections), and that most antimicrobial resistance among Salmonella isolates in the United States is caused by the use of antimicrobial agents in food animals, it is possible to evaluate the potential human health consequences of unrestricted veterinary use of fluoroquinolones in the United States using the human health risks model developed by the Institute of Medicine in 1988 (Institute of Medicine, 1989). Each year in the United States, most of the approximately 2,400 persons with life-threatening invasive Salmonella infections are treated with fluoroquinolones, of whom (although the isolates are susceptible to fluoroquinolones) approximately 500 die. Fluoroquinolones may therefore be life-saving for approximately 1,900 persons each year in the United States. If widespread fluoroquinolone resistance were to emerge among Salmonella isolates in the United States, evidence presented in this review demonstrates that this resistance would be the direct result of fluoroquinolone use in food animals. Because few persons have had fluoroquinolone-resistant Salmonella infections, the clinical significance of fluoroquinolone-resistant is not precisely known, however, because other antimicrobial treatment options are limited, treatment failures and serious outcomes, including deaths, would be expected. If 10% of Salmonella isolates in the United States were fluoroquinolone-resistant, and 10% of persons with invasive fluoroquinolone -resistant infections were to die, fluoroquinolone usage in food animals, which is currently limited to use in poultry, under such a scenario, would result in 19 deaths each year.
Need for Prudent use of Antimicrobial Agents in Food Animals
The emergence and increasing prevalence of antimicrobial-resistant Salmonella complicates the treatment of Salmonella infections in humans and animals. For example, few antimicrobial agents are available for the treatment of Salmonella Typhimurium DT104 R-type ACSSuT which becomes resistant to trimethoprim and fluoroquinolones. The increasing prevalence of antimicrobial resistance among Salmonella isolates, and the potential emergence of fluoroquinolone-resistant infections with adverse human health consequences, demonstrates the urgent need to develop strategies to reduce antimicrobial agent usage in food animals. Since antimicrobial agent usage can be reduced through the implementation of non-antimicrobial means of controlling infectious diseases, such as improved hygiene and sanitation, such efforts, which will minimize development of antimicrobial resistance and dissemination of antimicrobial-resistant pathogens, should be emphasized (Helmuth and Protz, 1997). Efforts should also be taken to ensure that antimicrobial agents are used prudently in food animals - prudent usage of antimicrobial agents maximizes the therapeutic effect of the antimicrobial agent and minimizes the development of antimicrobial resistance. Since subtherapuetic (growth promoter) uses of antimicrobial agents do not exert a therapeutic effect, such uses are non-prudent and should be replaced by non-antimicrobial methods of growth promotion. Because of the particular contribution of subtherapeutic use of penicillin and tetracycline in the development and dissemination of antimicrobial resistant Salmonella, the subtherapuetic use of penicillin and tetracycline should be terminated (World Health Organization, Division of Emerging and Other Communicable Diseases Surveillance and Control, 1997). Because fluoroquinolones are a vital class of antimicrobial agents for the treatment of potentially life-threatening Salmonella infections in humans, and widespread usage of fluoroquinolones in food animals will lead to rapid emergence and dissemination of resistance to humans with adverse health consequences, the use of fluoroquinolones in food animals should be restricted, particularly until validated guidelines for the prudent use of antimicrobial agents in food animals have been implemented and certified.
Figures
Figure one: Comparison of antimicrobial agents received by persons
with Salmonella infections (left) and antimicrobial resistance
among Salmonella isolates (right), CDC selected county
studies, 1985-1995.
Figure two: Comparison of antimicrobial resistance patterns,
by serotype, of human isolates collected in 1990 and Salmonella
Heidelberg isolates collected from chickens in 1988.
Figure three: Comparison of antimicrobial agents used in humans
in 1984 and on farms in 1979-1981, and antimicrobial resistance
among human Salmonella isolates collected in 1984.
Altekruse, S., J. Koehler, F. Hickman-Brenner, R.V. Tauxe, and K. A Ferris(1993).Comparison of Salmonella enteritidis phage types from egg-associated outbreaks and implicated laying flocks. Epidemiology and Infection; 110:17.
Anonymous (1994). FDA panel urges restricted use of fluoroquinolones in animals. ASM News; 60(7):350.
Beam Jr., TR (1994). Fluoroquinolones in animal feeds. ASM News; 60(7):348.
Bean, N.H., JS Goulding, C Lao, and FJ Angulo (1996). Surveillance for Foodborne Disease Outbreaks - United States, 1988-1992. Morbidity and Mortality Weekly Report Surveillance Summary 1996; 45(SS-5):1.
Buchwald DS and MJ Blaser (1984). A review of human salmonellosis: II. Duration of excretion following infection with nontyphi Salmonella. Reviews of Infectious Diseases; 6:345.
Centers for Disease Control and Prevention. Salmonella Surveillance, 1996. Atlanta, GA
Centers for Disease Control and Prevention. CDC/FDA/USDA National Antimicrobial Monitoring System 1996 Annual Report. Atlanta, GA.
Centers for Disease Control and Prevention. EPI-AID 96-16 (July, 1996.), Trip report: investigation of increased incidence of Salmonella Enteritidis isolates in Utah
Chalker, R.B. and M.J Blaser (1988). A review of human salmonellosis: III. Magnitude of salmonella infection in the United States. Review of Infectious Diseases; 9:111.
Chalker RB and MJ Blaser (1988). A review of human salmonellosis: III. Magnitude of Salmonella infection in the United States. Reviews of Infectious Diseases; 10:111.
Cherubin CE (1981). Antibiotic resistance of Salmonella in Europe and the United States. Reviews of Infectious Diseases; 3:1105.
Clark GM, AF Kaufmann, and EJ Gangrosa (1973). Epidemiology of an international outbreak of Salmonella agona. Lancet; 490.
Cohen, M.L. and R.V. Tauxe (1986). Drug-resistant Salmonella in the United States: An epidemiologic perspective. Science; 234:964.
Conte JE (1995). Manuel of antibiotics
and infectious diseases. Williams & Wilkins;
Baltimore.
Coronado VG, JR Edwards, DH Culver, RP Gaynes, and the National Nosocomial Infections Surveillance (NNIS) System (1995). Infection Control Hospital Epidemiology;16:71.
Council for Agricultural Science and Technology (1994). Foodborne pathogens: Risks and Consequences. Task Force Report No. 122, Ames, IA.
Glynn MK, C Bopp, W Dewitt, P Dabney, M Mokhtar, and FJ Angulo. Recent emergence of distinct pent-resistant Salmonella serotype Typhimurium DT104 infections in the United States. New England Journal of Medicine (in press).
Hargrett-Bean, N.T., A.T. Pavia, and R.V. Tauxe (1988). Salmonella isolates from humans in the United States, 1984-1986. Morbidity and Mortality Weekly Report Surveillance Summary; 37(SS-2): 25.
Hedberg, C.W., MJ David, KE White, KL MacDonald, and MT Osterholm (1993). Role of egg consumption in sporadic Salmonella enteritidis and Salmonella typhimurium infections in Minnesota. Journal of Infectious Diseases; 167:107.
Helmuth R and D Protz (1997). How to modify conditions limiting resistance in bacteria in animals and other reservoirs. Clinical Infectious Diseases; 24(Suppl 1):136.
Herikstad H, P Hayes, M Mokhtar, ML Fracaro, EJ Threlfall, and FJ. Angulo (1997) a. Emerging quinolone-resistant-Salmonella in the United States. Emerging Infectious Diseases; 3:371.
Herikstad H, PS Hayes, J Hogan, P Floyd, L Snyder, and FJ Angulo (1997) b. Ceftriaxone-resistant Salmonella in the United States. Pediatric Infectious Diseases Journal; 16:904.
Holmberg SD, MT Osterholm, KA Senger, and ML Cohen (1984). Drug-resistant Salmonella from animals fed antimicrobials. New England Journal of Medicine; 311:617.
Institute of Medicine (1989). Human health risks with subtherapuetic use of penicillin or tetracycline in animal feed: National Academy Press, Washington, DC.
Kass PH, TB Farver, JJ Beaumont, C Genigeorgis, and F Stevens (1992). Disease determinants of sporadic salmonellosis in four northern California counties: a case-control study of older children and adults. Annuals of Epidemiology; 2:683.
Lee LA, Puhr ND, K Maloney, NH Bean, and RV Tauxe (1994). Increase in antimicrobial-resistant Salmonella infections in the United States, 1989-1990. Journal of Infectious Diseases; 170:128.
Lyons RW, CL Samples, HN DeSilva, and KA Rose (1980). JAMA; 243.
McCullough NB and CW Eisele (1951). Experimental human salmonellosis: I. Pathogenicity of strains of Salmonella meleagridis and Salmonella anatum obtained from spray-dried whole egg. Journal of Infectious Diseases; 88:278.
McDonald KL, ML Cohen, N Hargrett-Bean, JG Wells, ND Puhr, SF Collin, and PA Blake (1987). Changes in antimicrobial resistance of Salmonella isolated from humans in the United States. JAMA; 258:1496.
Morse, D.L., G.S. Birkhead, J. Guardino, S.F. Kondracki, and JJ Guzewich (1994). Outbreak and sporadic egg-associated cases of Salmonella enteritidis: New York experience. American Journal of Public Health; 84: 859.
Panlilio AL, DH Culver, R Gaynes, JS Tolson, and WJ Martone (1992). Methicillin-resistant Staphylococcus aureus in U.S. hospitals, 1975-1991. Infection Control Hospital Epidemiology; 13:582.
Passaro, D.J., R. Reporter, L. Mascola, L Kilman., G.B Malcolm, J. Rolka, S.B. Werner and D.J. Vugia (1996). Epidemic Salmonella enteritidis infection in Los Angeles county, California - the predominance of phage type 4. Western Journal of Medicine; 165:126.
Richwald GA, S Greenland, BJ Johnson, JM Friedland, EJC Goldstein, and DT Plichta (1988). Assessment of the excess risk of Salmonella dublin infection associated with the use of certified raw milk. Public Health Reports; 103:489.
Riley LW, ML Cohen, JE Seals, MJ Blaser, KA, Birkness NT Hargrett, SM Martin, and RA Feldman (1984). Importance of host factors in human salmonellosis caused by multiresistant strains of Salmonella. Journal Infectious Diseases; 149:878.
Schmind H, AP Burnens, A Baumgartner, and J Oberreich (1996). Risk factors for sporadic salmonellosis in Switzerland. European Journal of Clinical Microbiology and Infectious Diseases; 15:725.
Seyfarth AM, HC Wegner, and N Frimodt-Moeller (1997). Antimicrobial resistance in Salmonella enterica subsp. enterica serovar typhimurium from humans and production animals. Journal of Antimicrobial Chemotherapy; 40:67.
Spika JS, SH Waterman, GW Soo Hoo, ME St. Louis, RE Pacer, SM, James ML Bissett, LW Mayer, JY Chiu, B Hall, K Greene, ME Potter, ML Cohen, and PA Blake (1987). Chloramphenicol-resistant Salmonella newport traced through hamburger to dairy farms. New England of Journal of Medicine; 316:565.
Tacket CO, LB Dominguez, HJ Fisher, and ML Cohen (1985). An outbreak of multiple-resistant Salmonella enteritis from raw milk. JAMA; 253:2058.
Tauxe RV (1986). Antimicrobial resistance in human salmonellosis in the United States. Journal Animal Science; 62(Suppl 3):65.
Threlfall EJ, NG Powell, and B. Rowe (1994) . Differentiation of salmonellas by molecular methods. PHLS Microbiology Digest; 11:199.
Threlfall EJ, LR Ward, and B Rowe(1997). Increasing incidence of resistance to trimethoprim and ciprofloxacin in epidemic Salmonella Typhimurium DT104 in England and Wales. Eurosurveillance; 2:81.
U.S. Congress, Office of Technology
Assessment (1995 [OTA-H-629]). Impacts of antibiotic-resistant
bacteria: U.S. Government Printing Office, Washington, DC
Wilcox MH, RC Spencer (1992). Quinolones and salmonella gastroenteritis.
Journal Antimicrobial Chemotherapy; 30:221.
World Health Organization, Division
of Emerging and other Communicable Diseases Surveillance and
Control (1997). The medical impact of the use of antimicrobials
in food animals. Geneva, World Health Organization;WHO/EMC/Z00/97.4:1-12.
© Queen's Printer for Ontario, 2001
[Source: www.gov.on.ca/OMAFRA/english/livestock/animalcare/amr/facts/angulo.htm]