Abstract
Antimicrobial resistance, including multidrug resistance (MDR), is an increasing problem globally. MDR bacteria are frequently detected in humans and animals from both more- and less-developed countries and pose a serious concern for human health. Infections caused by MDR microbes may increase morbidity and mortality and require use of expensive drugs and prolonged hospitalization. Humans may be exposed to MDR pathogens through exposure to environments at health-care facilities and farms, livestock and companion animals, human food, and exposure to other individuals carrying MDR microbes. The Centers for Disease Control and Prevention classifies drug-resistant foodborne bacteria, including Campylobacter, Salmonella Typhi, nontyphoidal salmonellae, and Shigella, as serious threats. MDR bacteria have been detected in both meat and fresh produce. Salmonellae carrying genes coding for resistance to multiple antibiotics have caused numerous foodborne MDR outbreaks. While there is some level of resistance to antimicrobials in environmental bacteria, the widespread use of antibiotics in medicine and agriculture has driven the selection of a great variety of microbes with resistance to multiple antimicrobials. MDR bacteria on meat may have originated in veterinary health-care settings or on farms where animals are given antibiotics in feed or to treat infections. Fresh produce may be contaminated by irrigation or wash water containing MDR bacteria. Livestock, fruits, and vegetables may also be contaminated by food handlers, farmers, and animal caretakers who carry MDR bacteria. All potential sources of MDR bacteria should be considered and strategies devised to reduce their presence in foods. Surveillance studies have documented increasing trends in MDR in many pathogens, although there are a few reports of the decline of certain multidrug pathogens. Better coordination of surveillance programs and strategies for controlling use of antimicrobials need to be implemented in both human and animal medicine and agriculture and in countries around the world.
Introduction
A
Infections caused by MDR microbes may result in increased morbidity and mortality and are often more expensive to treat because more costly drugs and prolonged hospital stays may be required (Collignon, 2012). A recent analysis of the cost of treating tuberculosis (TB) in 2010 U.S. dollars found that non-MDR TB cost about $17,000 per patient, MDR TB cost about $134,000 per patient, and extensively-drug-resistant (XDR) TB cost $430,000 per patient in the United States (Marks et al., 2014).
A recent publication by the Centers for Disease Control (CDC) on antibiotic resistance threats in the U.S. listed drug-resistant foodborne bacteria, including Campylobacter, Salmonella Typhi, nontyphoidal salmonellae, and Shigella, as serious threats (CDC, 2013a). Salmonellae carrying genes coding for resistance to multiple antibiotics are the most frequent causes of foodborne MDR outbreaks.
Food processors and retailers may consider themselves removed from the issue of antibiotic resistance, but articles in the popular press warning of multiresistant “superbugs” in food raise questions about safety of the food supply overall. Consumer Reports recently reported that nearly 50% of retail packages of chicken breasts tested contained at least 1 strain of MDR bacteria (Consumer Reports, 2013). MDR has also been detected in Enterobacteriaceae on fresh produce in supermarkets (Blaak et al., 2014b).
Foodborne MDR bacteria on meat may have originated in veterinary health-care settings or on farms where animals are given antibiotics in feed or to treat infections. Fresh produce may be contaminated by irrigation or wash water containing MDR bacteria. Food handlers may cross contaminate foods during preparation and if they are carriers of MDR bacteria, they may contaminate foods themselves. All potential sources of MDR bacteria should be considered and strategies devised to reduce their presence in foods. Some review articles summarize information on antimicrobial resistance and the challenges it presents for the food industry (Hur et al., 2012; Capita and Alonso-Calleja, 2013; Doyle et al., 2013).
Multidrug-Resistant Bacteria in Foods
Surveillance reports
In the United States, the National Antimicrobial Resistance Monitoring System (NARMS) tracks antimicrobial resistance in bacteria isolated from humans, animals, and foods. Surveillance data from 2012 and 2011 on MDR in human and meat isolates of some enteric bacteria are listed in Table 1 (National Antimicrobial Resistance Monitoring System, 2014). European Food Safety Authority and European Centre for Disease Prevention and Control also publish a yearly report on antimicrobial resistance in bacterial isolates from humans, livestock, and food. Data on 2012 isolates are summarized in Table 2 (European Food Safety Authority, European Centre for Disease Prevention and Control, 2014). Multidrug resistance was found to be relatively high in Escherichia coli and Salmonella isolates from poultry in the United States and from poultry and swine in Europe.
In some cases, surveillance data were reported by only one or a few countries in the EU. Prevalence of MDR also varies significantly among reporting countries.
European Food Safety Authority and European Centre for Disease Prevention and Control, 2014.
MRSA, methicillin-resistant Staphylococcus aureus.
In addition to these large annual surveillance reports, there have been numerous journal articles documenting the presence of MDR bacteria in different foods. See Table 3 for some recent examples. MDR bacteria have also been detected in imported horse meat in France (Salmonella Newport) (Espie et al., 2005), quail in Italy (Salmonella spp.) (Bacci et al., 2012), fish in India (Salmonella Oslo) (Kakatkar et al., 2011), shrimp in Bangladesh (Hossain et al., 2012), and eggs in Grenada (E. coli) (Arathy et al., 2011).
Year food samples were collected.
RTE, ready to eat.
Outbreaks
Although many genera of bacteria contain MDR strains, Salmonella spp. are the most common type of MDR bacteria associated with outbreaks of foodborne illness. Table 4 lists outbreaks attributed to MDR salmonellae. Widespread reports, in the mid-1990s, of Salmonella Typhimurium DT104 in meats and livestock were the first indication of this emerging problem. DT104 is resistant to ampicillin, chloramphenicol, streptomycin, sulfonamides, and tetracycline—a resistance pattern designated as ACSSuT (Hosek et al., 1997; Cody et al., 1999; Stokes and Gillings, 2011; Van Boxstael et al., 2012). This pattern of resistance has spread and was identified in 17% of Salmonella Typhimurium, 4% of Salmonella Newport, and 88% of Salmonella Dublin isolates tested by NARMS in 2012 (National Antimicrobial Resistance Monitoring System, 2014).
RTE, ready to eat.
An MDR, extended-spectrum β-lactamase (ESBL) producing strain of Klebsiella pneumoniae was identified as the cause of one foodborne outbreak affecting 156 hospital patients in Spain. The outbreak strain was isolated from some kitchen surfaces and food handlers (Calbo et al., 2011). MDR Enterobacteriaceae may pose a broader threat of foodborne disease, particularly in health-care facilities. They were detected in 92% of raw chicken samples from supermarkets and from hospital kitchens in Switzerland (Stewardson et al., 2014). Examination of vinyl gloves, after use by hospital kitchen staff for handling raw chicken, and of cutting boards, used in hospital and community kitchens for cutting raw chicken, in Switzerland revealed that 12% of boards and 50% of gloves contained MDR E. coli (Tschudin-Sutter et al., 2014).
Mechanisms and Evolution of Drug Resistance
Origin and selection for resistance
Many antibiotics used in human and veterinary medicine are derived from chemicals produced by microbes for the purpose of inhibiting or killing competitors or predators. Since these antibiotics were manufactured long before humans discovered and used them, there are bacterial genomes containing antimicrobial resistance genes in environments long isolated from any human contact (Finley et al., 2013; Walsh and Duffy, 2013).
Recently, more rapid selection for and sharing of resistance genes have occurred. Some bacterial strains are now resistant to three or more classes of antimicrobial substances—the current definition of MDR (Table 5). Other bacteria are XDR, susceptible to drugs in only one or two antimicrobial categories. Pandrug resistance is defined as resistance to all agents in all available antimicrobial categories (Magiorakos et al., 2012). Not all resistant bacteria are human pathogens. However, they may act as environmental reservoirs that can transfer antimicrobial resistance to pathogens.
Antimicrobial resistance in E. coli isolates from humans and livestock during 1952 to 2002 in the United States showed an increasing trend of resistance. Only 7.2% of E. coli were MDR during the 1950s as compared to 63.6% in the 2000s (Tadesse et al., 2012). Another pathogen that has become progressively more resistant to antibiotics is Neisseria gonorrhoeae. The first antibiotics effective in treating gonorrhea were sulfonamides in the 1930s. However, by the 1940s, resistance to sulfa drugs became widespread and then increasing doses of penicillin, fluoroquinolones, and cephalosporins were used successively to treat this disease. Cephalosporin resistance was observed in Asia during the 2000s and in Europe in the past few years. It has become very difficult to stay ahead of the rapid acquisition of antimicrobial resistance by some important pathogens (Hook et al., 2013).
Numerous serovars of Salmonella contain MDR strains. CDC reported that while MDR declined in Salmonella in recent years, this was due to a reduction in numbers of Salmonella Typhimurium isolated. MDR has increased in other Salmonella serovars, including Salmonella Newport and Salmonella Heidelberg. Resistance to ceftriaxone and ciprofloxacin also increased during this time (Medalla et al., 2013). Several highly drug-resistant Salmonella Kentucky ST198-X1 strains have been recently detected in poultry flocks and turkey meat in Europe and Canada (Le Hello et al., 2013; Mulvey et al., 2013). In France during 2000–2008, about 40% of Salmonella Kentucky isolates were resistant to ciprofloxacin; in 2009–2011, 83% of Salmonella Kentucky were resistant to this drug. Some strains are also resistant to carbapenems, fluoroquinolones, trimethoprim-sulfamethoxazole, and azithromycin (Le Hello et al., 2013). Since Salmonella Kentucky is one of the most common Salmonella strains detected in chickens and ground chicken meat, there is concern that this strain may become a problem in the United States.
Increasing levels of antimicrobial resistance are believed to result from widespread use of antibiotics in human and veterinary medicine and as growth promoters for intensive livestock production (Capita and Alonso-Calleja, 2013). Human and veterinary health-care facilities commonly harbor drug-resistant bacteria, including MDR strains. Low concentrations of antibiotics added to animal feed may select for resistant bacteria in livestock and may be excreted in animal wastes. When manure is spread on agricultural fields and possibly washed into surface waters, antibiotic residues may select for resistance in environmental bacteria. Antibiotics used in the aquaculture industry have selected for antibiotic resistance among pond bacteria (Nonaka et al., 2012).
Pharmaceutical manufacturing wastes and urban wastewater from health-care facilities and homes may contain excess or outdated antimicrobial compounds and their derivatives that select for antimicrobial resistance in bacteria at water treatment plants (Marathe et al., 2013; Michael et al., 2013). These bacteria are later released into the environment.
In addition to direct selection for specific antimicrobial resistance by a particular antibiotic, selection may be indirect because genes encoding resistance may be located close to each other on genetic elements. In-feed antibiotics containing chlortetracycline, sulfamethazine, and penicillin that were fed to pigs also selected for resistance to unrelated aminoglycoside antibiotics (Looft et al., 2012). Genes encoding resistance to heavy metals (Kuenne et al., 2010; Del Castillo et al., 2013) and disinfectants (Hegstad et al., 2010; Buffet-Bataillon et al., 2012; Tandukar et al., 2013) may also be linked to antimicrobial resistance genes.
Mechanisms
Bacterial cells can survive exposure to antibiotics simply by drastically decreasing metabolic activity and entering a dormant state where antibiotics that interfere with cell wall synthesis, protein synthesis, or other processes have little effect. These “persister” cells comprise about 1% of cells in biofilms and stationary cultures and they can resume growth when drug levels fall (Wood et al., 2013). Other bacteria are intrinsically resistant to some antibiotics because of thick, lipid-rich cell walls that many antibiotics cannot penetrate (Mueller et al., 2013).
Bacteria also contain genes coding for altered proteins that do not bind to antibiotics and for enzymes that break down antibiotics, prevent their entry into cells, or actively expel antimicrobial compounds from cells. Multiple resistance is often associated with ATP-powered efflux systems. A recent review described genetic resistance mechanisms present in bacteria from U.S. livestock (Frye and Jackson, 2013).
Structural changes
Antibiotics bind to macromolecules in order to gain entry into cells. Methicillin-resistant Staphylococcus aureus (MRSA) synthesizes altered cell wall proteins with low affinity for methicillin, thereby diminishing its uptake (Robinson and Enright, 2004). Fluoroquinolones target bacterial gyrases and topoisomerases that maintain supercoiling of bacterial chromosomes. Mutations in these enzymes prevent fluoroquinolones from attaching to them (Webber et al., 2013).
Inactivation enzymes
Bacterial enzymes can break down or alter antibiotic molecules to render them less effective or ineffective. Aminoglycoside resistance is often mediated by enzymes that modify the antibiotic (Hur et al., 2012).
Resistance to penicillin-type drugs is generally mediated by β-lactamase enzymes. Early resistance genes encoded enzymes capable of degrading only a few penicillin variants. As more effective, modified penicillin drugs were produced, some bacteria synthesized extended-spectrum β-lactamases (ESBL), which can attack a wide variety of third-generation β-lactams, including cephalosporins. About 12% of E. coli, Klebsiella spp., and Proteus mirabilis collected at U.S. hospitals in 2012 carried ESBL genes (Castanheira et al., 2014).
A zinc-containing enzyme, New Delhi metallo-β-lactamase (NDM), was first described in 2008. NDM genes encode enzymes that can inactivate all penicillins, cephalosporins, and carbapenems. These genes have spread to several genera of Enterobacteriaceae and other bacteria and to countries all over the world (Kumarasamy et al., 2010; Borgia et al., 2012; Denisuik et al., 2013; Goettig et al., 2013; Shoma et al., 2014). The gene encoding NDM-1 has also been detected in E. coli isolates from companion animals in the United States and in bacteria from livestock (Shaheen et al., 2013; Woodford et al., 2014).
Efflux systems
Numerous efflux systems have been characterized in bacteria (Fernández and Hancock, 2012). Some are specific for certain molecules, such as the tetracycline pump. However, most efflux systems interact with many molecules including dyes and detergents as well as antibiotics. Salmonella Typhimurium cells may have as many as 10 types of efflux pumps (Yamasaki et al., 2013). A multidrug efflux system, CmeABC, was detected first in Campylobacter jejuni and then found to be present in four other species of Campylobacter (Guo et al., 2010).
Efflux systems also aid cells in adapting to acidic conditions (Deininger et al., 2011) and oxidative stress (Ramón-Garcia et al., 2009; Bogomolnaya et al., 2013). Some data indicate that efflux pumps contribute to biofilm production and to the enhanced antimicrobial resistance of biofilms (Matsumura et al., 2011; Soto, 2013).
Efflux proteins known as ABC (ATP-binding cassette) proteins are found in fungi, including some opportunistic pathogens, and confer resistance to a variety of antimicrobials (Morschhaeuser, 2010; Prasad and Goffeau, 2012). Similar ABC efflux systems and MDR have also been identified in Toxoplasma and Cryptosporidium (Sauvage et al., 2009) and in human cancer cells (Kunjachan et al., 2013).
Transfer to other genera
Genes encoding antimicrobial resistance can be located on chromosomes or on mobile genetic elements including plasmid DNA, transposons, integrons, and genomic islands. MDR in salmonellae is often encoded by Salmonella genomic island 1, which confers resistance to ampicillin, chloramphenicol, streptomycin, sulfamethoxazole, and tetracycline (Beutlich et al., 2011). Bacteria can transfer genes from one cell to another by conjugation, transduction, and transformation (Kelly et al., 2009; Stokes and Gillings, 2011; Domingues et al., 2012).
Numerous examples of transfer of antimicrobial resistance have been reported: • Vibrio cholerae often forms biofilms on zooplankton, and this attachment increases transfer of genetic information by transformation and transduction (Blokesch, 2014). • Genome sequencing of some MDR bacteria from wastewater treatment plants shows that they acquired resistance genes from several different bacteria (Johnning et al., 2013). • Sublethal doses of antibiotics mixed in animal feed also appear to encourage transfer of resistance genes among bacteria by inducing replication of prophages (Allen et al., 2011; Looft et al., 2012). • Genes conferring methicillin resistance in Staphylococcus aureus reside on a large mobile genetic element that has been transferred to multiple methicillin-susceptible S. aureus strains (Robinson and Enright, 2004). • Prior to 1995, Salmonella Typhi strains contained a variety of plasmids carrying antibiotic resistance genes. Since that time 98% of isolates worldwide were found to contain one version of the IncHI1 plasmid, indicating that one MDR plasmid has spread among Salmonella Typhi globally (Holt et al., 2011). • IncA/C plasmids mobilize MDR genes for transfer in Salmonella (Douard et al., 2010) and in E. coli and other enteric bacteria (Fricke et al., 2009; Fernandez-Alarcon et al., 2011). • Experiments with animals including meal worms (Poole and Crippen, 2009) and poultry (Van Essen-Zandbergen et al., 2009) have demonstrated that horizontal gene transfer occurs in vivo.
Reservoirs of MDR in Animals and the Environment
Some MDR bacteria apparently originated in human health-care facilities while others developed in animals either in veterinary facilities or on farms. Research has documented their presence in livestock, companion animals, wild animals, insects, and the natural environment, including soil and surface waters. Even resistant nonpathogens may constitute a health risk because of their ability to transfer genetic information to pathogenic microbes.
MDR bacteria can be transmitted from person to person (Vanhoof et al., 2012), among animals (Schaer et al., 2010), and between animals and humans (Ajiboye et al., 2009). Insects may also be vectors, dispersing antibiotic-resistant bacteria to animals and humans (Ahmad et al., 2011).
Livestock
Cattle
Prevalences of MRSA and MDR Salmonella in cattle in Europe in 2012 were reported to be 33.4% and 34.2%, respectively, and MDR Salmonella in U.S. cattle was reported to be 28.9% in 2010 (Table 2). Salmonella serotypes, isolated from humans and cattle, are similar but generally a greater proportion of isolates from cattle are MDR (Oloya et al., 2009; Hoelzer et al., 2010). A median prevalence of 11.8% was observed for Salmonella contamination of lymph nodes from feedlot cattle, and 8.3% of those were determined to be MDR (Gragg et al., 2013).
A longitudinal study of the acquisition of new MDR Salmonella strains by dairy herds in the United States found that this was a fairly common event. Herd size and off-farm heifer raising were significantly correlated with the introduction of new MDR salmonellae (Adhikari et al., 2009a, b). In 1 U.S. study, >81% of E. coli isolated from calves with diarrhea were MDR (Barigye et al., 2012). In Australia, by contrast, 72.4% of Salmonella isolates associated with diarrhea in calves were susceptible to all drugs tested (Izzo et al., 2011).
MDR Salmonella Stanley and E. coli have been isolated from Japanese cattle (Dahshan et al., 2011; Ohnishi et al., 2013). Very high levels of antibiotic resistance were detected in E. coli from cattle in Nigeria (Amosun et al., 2012; Ogunleye et al., 2013) and in
Swine
Prevalences of MRSA, MDR Campylobacter coli, MDR E. coli, and MDR Salmonella in swine in Europe in 2012 were reported to be 27.5%, 34.6%, 30.9%, and 73.5%, respectively (Table 2). MDR Salmonella in U.S. swine was reported to be 27.9% in 2010 (Table 1).
MDR salmonellae are present in lymph nodes, feces, housing, and environments where swine are raised, lairage pens, and on carcasses in slaughterhouses. Salmonella Typhimurium is often identified as the dominant MDR serovar, but other serovars are also MDR (Miller et al., 2011; Haley et al., 2012; Schmidt et al., 2012; Bolton et al., 2013; Keelara et al., 2013).
MDR Campylobacter have been reported from swine in Brazil (Biasi et al., 2011), China (Qin et al., 2011), and Australia (Obeng et al., 2012). Over 90% of Enterococcus isolates from pigs in China were MDR (Liu et al., 2013b). MDR E. coli were isolated from pigs in Thailand (Lay et al., 2012), Australia (Smith et al., 2010), and Canada (Varga et al., 2008).
Several studies provided evidence for the spread of drug-resistant bacteria from swine and their environments to humans working in those areas (Alali et al., 2010a; Oppliger et al., 2012; Novais et al., 2013).
Chickens
Prevalences of MDR Campylobacter coli, MDR E. coli, and MDR Salmonella in broilers in Europe in 2012 were reported to be 22.5%, 31.1%, and 46.4%, respectively (Table 2). MDR Campylobacter, Enterococcus, E. coli, and Salmonella in U.S. chickens were reported to be 1.3%, 61.6%, 38.3%, and 15.2%, respectively, in 2010 (Table 1).
Analyses of antimicrobial resistance in Salmonella isolates from chicken fecal pellets in the United States revealed that nearly 40% of those from conventional farms were resistant to 6 antibiotics while none of the Salmonella from organic farms had this type of MDR (Alali et al., 2010b; Thakur et al., 2013). Of salmonellae isolated from broilers in Japan, 90% were MDR (Sasaki et al., 2012). Important MDR Salmonella serovars include Salmonella Infantis (Shahada et al., 2010; Nogrady et al., 2012) and Salmonella Paratyphi B (Doublet et al., 2014).
All Campylobacter coli and C. jejuni isolates from Italian broiler farms were MDR in a recent survey (Giacomelli et al., 2014). High levels of MDR were also detected in Campylobacter spp. from broilers in Turkey (Cokal et al., 2009) and Malaysia (Mansouri-Najand et al., 2012).
Of 600 E. coli isolates from poultry flocks in Alberta, Canada in 2005, 54.3% were MDR (Mainali et al., 2013). E. coli from poultry in the United States are more likely than human isolates to be MDR (Johnson et al., 2012).
MDR in Listeria monocytogenes in poultry in Spain increased dramatically between 1993 and 2006: from 18.6% to 84% (Alonso-Hernando et al., 2012). MDR has also been detected in Enterococcus spp. in Canada (Diarra et al., 2010) and China (Liu et al., 2013b), and in Clostridium perfringens in Egypt (Osman and Elhariri, 2013).
Turkeys
Prevalence of MDR Salmonella in U.S. turkeys was reported to be 37.1% in 2010 (Table 1). MRSA, and MDR Salmonella in turkeys in Europe were reported to be 12%, and 69%, respectively, in 2012 (Table 2). A clonal MDR strain of Salmonella Saintpaul is widespread in German turkeys and has been isolated from Dutch poultry. This strain has also been detected in some turkey products and in people (Beutlich et al., 2010).
A clonal multidrug strain of C. coli was identified in turkeys in the United States (D'Lima et al., 2007). C. coli and C. jejuni from turkeys raised on commercial farms in Italy had a very high rate of MDR (98%) in one recent study (Giacomelli et al., 2014). Other reports of MDR bacteria in turkeys include enterococci in Canada (Tremblay et al., 2011), C. jejuni in Germany (El-Adawy et al., 2012), Salmonella and E. coli in Greece (Iossifidou et al., 2012), and Salmonella and E. coli in the United Kingdom (Mueller-Doblies et al., 2013).
Other farm animals
MDR bacteria have also been isolated from farmed rabbits in Italy (Salmonella Typhimurium DT104) (Borrelli et al., 2011), feedlot lambs in the United States (E. coli O157:H7) (Edrington et al., 2009), and farmed ducks in Tanzania (Campylobacter spp.) (Nonga and Muhairwa, 2010).
Other animals
Companion animals
MDR bacteria have been isolated from pets and companion animals, such as horses, particularly after treatment in a veterinary hospital (Schaer et al., 2010; Gibson et al., 2011; Maddox et al., 2011; Hamilton et al., 2013; Williams et al., 2013a). These MDR bacteria may increase health risks for pet owners and those caring for sick animals (Wieler et al., 2011). Outbreaks of MDR Salmonella Typhimurium occurred among employees and clients of veterinary facilities caring for cats with diarrhea (CDC, 2001) and among persons exposed to pet hamsters, mice, or rats (Swanson et al., 2007). The European Medicines Agency published a paper discussing the risk of transfer of antimicrobial resistance from companion animals to humans (European Medicines Agency, 2013).
Wild animals
MDR E. coli and Salmonella spp. have been detected in wild mammals, including wolves (Simões et al., 2012), wild boar (Literak et al., 2010), and mongoose (Pesapane et al., 2013). These bacteria were very similar to MDR strains from humans or domestic animals. MDR salmonellae were isolated from reptiles and amphibians living in a produce-growing region of California (Gorski et al., 2013). MDR bacteria are also present in birds that are very mobile and may disseminate MDR bacteria widely (Molina-Lopez et al., 2011; Da Silva et al., 2012; Kitadai et al., 2012; Poirel et al., 2012; Veldman et al., 2013).
Insects
Most insects are also mobile and may transport bacteria to new locations. MDR bacteria have been detected in flies at Dutch poultry farms (Blaak et al., 2014a), house flies and cockroaches at U.S. swine farms (Ahmad et al., 2011), flies at a Chinese airport (Liu et al., 2013c), and cockroaches at an Ethiopian hospital (Tilahun et al., 2012).
Environment
MDR bacteria have been isolated from a variety of natural environmental sources and from indoor environments where humans are active. The importance of the environment in maintaining antibiotic resistance levels was discussed in a recent review (Finley et al., 2013).
Agricultural environments where antibiotics are used may contain many species of resistant bacteria, including MDR strains (Walsh and Duffy, 2013). Antibiotic-resistant Salmonella were detected in soil and irrigation water on tomato farms and may contaminate tomatoes (Micallef et al., 2012). MDR bacteria can survive in manure and waste lagoons and may spread from these sources to areas outside of farms (Antunes et al., 2011). Compared to reference sites distant from intense agricultural activity, river water downstream from concentrated animal feeding operations in the United States contained much higher levels of MDR bacteria (West et al., 2011). MDR bacteria are also present in some aquaculture systems in Japan (Nonaka et al., 2012).
Examination of 93 bacterial strains from a wastewater treatment plant serving an antibiotic-producing plant in India found that 86% were resistant to 20 or more antibiotics (Marathe et al., 2013). Treated water from wastewater plants is discharged into surface waters, and there are numerous reports of MDR bacteria in surface waters including river water in Mexico (E. coli) (Ramirez et al., 2013), Lake Erie water (Aeromonas spp.) (Skwor et al., 2014), coastal marine sediments near Italy (Vignaroli et al., 2012), Salmon River in British Columbia (Xu et al., 2011), and surface waters in Turkey (Yilmaz et al., 2013).
Environmental surfaces in locations where antibiotics are frequently used, such as medical and veterinary facilities, may also harbor MDR bacteria. A chronic outbreak of MDR Klebsiella oxytoca in a Spanish hospital was finally traced to a safe niche where bacteria survived in sink drains and traps (Vergara-Lopez et al., 2013). MDR bacteria have been isolated from other environmental surfaces (for example, high-touch surfaces on a university campus (MRSA) [Roberts et al., 2013] and domestic kitchen surfaces in Tennessee [Cronobacter sakazakii] [Kilonzo-Nthenge et al., 2012]).
Control Strategies
Preventing the emergence of MDR bacteria and protecting food from contamination with these resistant strains are complex problems. Overuse and misuse of antimicrobials occurs in human medicine when antimicrobials are prescribed for viral diseases (not susceptible to antibiotics) or for bacterial infections without testing the sensitivity of the pathogens. Some patients receiving appropriate antibiotic prescriptions do not finish the full course of medicine. This may allow survival of resistant strains. In some developing countries, antibiotics can be purchased over the counter and may be used to treat conditions where they are ineffective.
Antimicrobial use in agriculture also drives evolution of resistant bacteria. According to a recent U.S. Food and Drug Administration (FDA) report, a greater quantity of antibiotics (in kilograms) are sold for use in animals than in humans (Center for Veterinary Medicine, 2013). Although some drugs are used to treat sick animals, large amounts are used in feed to promote growth and feed efficiency. These are often available over the counter.
More judicious use of antibiotics in both veterinary and human medicine has been shown to reduce numbers of resistant bacteria. The FDA has recently issued new guidance for drug companies to voluntarily discontinue the use of antimicrobials for production purposes and to change marketing status from over the counter to prescription by a veterinarian (FDA, 2013a). A new proposed rule, Veterinary Feed Directive, encourages judicious use of antibiotics in animal agriculture, particularly for drugs that are important in human medicine (FDA, 2013b). However, the prevalence of resistant bacteria does not revert to zero when the use of antibiotics is ended. Reservoirs of resistance genes may persist for long periods of time.
Several approaches may be utilized for reducing antibiotic usage in livestock. The U.S. Department of Agriculture discusses these on its web page (USDA, 2014). These include proper use of immunomodulators that increase immune function and disease resistance of animals; timely inspections to identify and treat sick animals before disease spreads; maintenance of a hygienic and healthy living environment; and use of laboratory tests to detect animals at risk of developing disease. Components for such a system were reviewed for dairy cattle (Trevisi et al., 2014).
National surveillance in Denmark and programs to combat the increasing problem of resistance to antimicrobials, particularly with reference to animal agriculture, have effectively reduced prevalence of antibiotic-resistant bacteria in livestock. However, implementation of programs and policies for more prudent use of antimicrobials and better animal husbandry methods was not always easy and various stakeholders were involved (Wielinga et al., 2014). Strategies to reduce antibiotic use in livestock have also been implemented in the Netherlands and other European countries (Lam et al., 2012).
Australia has maintained fluoroquinolone resistance at low levels by legislation controlling the use of these drugs in humans and animals (Cheng et al., 2012).
Efforts to combat MDR TB and MRSA in health-care facilities, including enhanced hygiene measures and testing and isolation of infected patients, have aided in containing these infections (Fournier et al., 2012; Huang et al., 2013; Suzuki et al., 2013). The European Society of Clinical Microbiology and Infectious Diseases has published guidelines for reducing transmission of MDR Gram-negative bacteria in hospitals (Tacconelli et al., 2014). Some of these practices may also be useful in veterinary clinics, animal husbandry, and food processing (Doyle et al., 2013).
Conclusions
MDR in microbes has become widespread geographically, with new MDR strains spreading around the world in a fairly short time. This is not just a problem of developed countries where large quantities of antibiotics are used for human medicine and in animal agriculture. Less-developed countries often do not have adequate controls on use, sale, and distribution of antibiotics. While there is some level of resistance to antimicrobials in environmental bacteria that have not been exposed to human and veterinary drugs, the widespread use of antibiotics in medicine and agriculture has driven the selection of microbes with greater levels of resistance, thereby significantly increasing costs for treatment of some common infections as well as their morbidity and mortality.
Humans may be potentially exposed to MDR pathogens through a variety of routes including environments at health-care facilities, farm environments and animals, companion animals and their food, foods from animals carrying MDR bacteria, fresh produce carrying MDR pathogens acquired from contaminated soil or water, and exposure to other individuals carrying MDR microbes.
Several reviews have documented the increasing trends of resistance to multiple drugs in diseases such as TB, gonorrhea, and typhoid fever. Surveillance studies have also documented increasing trends in MDR in many other pathogens, although there are a few reports of the decline of certain multidrug pathogens. For example, MRSA is still a major health concern, but numbers have declined in some health-care facilities that have adopted stringent controls on antibiotic use and screening and treatment of incoming patients to prevent importation of new sources of MRSA into hospitals. Strategies for controlling use of antimicrobials need to be implemented in both human and animal medicine and agriculture and in countries around the world.
Footnotes
Acknowledgments
This publication was supported in part by a grant from the American Meat Institute Foundation.
Disclosure Statement
No competing financial interests exist.