Antimicrobial Resistance in Food

1Department of Population Medicine, Ontario Veterinary College, University of Guelph; 2Laboratory for Foodborne Zoonoses, Population and Public Health Branch, Health Canada, Guelph, Ontario Correspondence and reprints: Dr Scott McEwen, Department of Population Medicine, Ontario Veterinary College, University of Guelph, Guelph, Ontario N1G 2W1. Telephone 519-824-4120 ext 54751, fax 519-763-3117, e-mail smcewen@uoguelph.ca Newspapers and other media provide almost daily reminders to Canadians of the potential risks of foodborne infection. The recent discovery of a case of bovine spongiform encephalopathy in this country and its dramatic impact on farming and international trade show some of the indirect effects of foodborne contaminants on society. But it’s more than media hype; our provincial and national notifiable disease surveillance programs provide abundant evidence of the direct effects on morbidity and mortality from a wide range of food and waterborne contaminants, such as Salmonella, Campylobacter, Escherichia coli O157:H7 and other bacterial (eg, Listeria, Yersinia), parasitic (eg, Cyclospora, Toxoplasma) and viral (eg, norovirus) infections. What about antimicrobial resistance as a food safety issue? Understandably, methicillin-resistant Staphylococcus aureus, vancomycin-resistant enterococci, and resistance among respiratory, genitourinary and some other human-derived bacterial infections receive top billing on lists of resistance problems facing the medical community (1). For these infections, antimicrobial prescribing practices in hospitals and the community can often be directly linked to emergence of resistance. Not to be forgotten, however, is resistance among foodand waterborne bacteria, where resistance also increases the burden of illness, but where animals and other environmental sources are major reservoirs of infection. Therefore, in contrast to the major nosocomial and community-acquired bacterial infections that have predominantly human reservoirs, we look to nonhuman use of antimicrobials as potential drivers of resistance in many of the foodand waterborne bacteria of animal origin (eg, the zoonoses). Most of the nonhuman use of antimicrobials is in food animals, although they are also used in pets, horses and in some types of plant production (eg, prophylaxis of some bacterial infections of fruit), but these are thought to be of lesser importance, at least to the extent that they have been studied (2). The article by Forward and colleagues (pages 226-230) in this issue of the Journal provides an excellent example of the potential for food to be a vehicle of infection because of antimicrobial-resistant pathogens and commensals. It also raises a number of issues that are central to the ongoing and often highly contentious discussion of nonhuman use of antimicrobials and whether changes in current practice are needed. These issues include the use in animals of antimicrobial drugs of critical importance in human medicine, the challenges to understanding and controlling resistance that are posed by the complexity of the food chain and the general ecology of resistance, the need for surveillance of both resistance and antimicrobial use, and the importance of improved food safety activities throughout the farm-to-fork continuum, including prudent use of antimicrobials in animals. Many foodborne bacteria are quite capable of causing illness in humans in the absence of antimicrobial resistance (eg, E coli O157:H7 and many Salmonella and Campylobacter infections), however, there are a number of mechanisms by which antimicrobial resistance in foodborne bacteria can increase the burden of illness. These mechanisms include: rendering infections more difficult or expensive to treat; enhancing virulence or pathogenicity, resulting in more severe or longer-lasting disease; increasing risk of infection (in particular among resistant Salmonella) in people taking antimicrobials for other reasons through reduction of colonization resistance; contributing to the pool of resistance determinants available for uptake by other human pathogens; and enhancing the spread of zoonotic infections in animals undergoing antimicrobial therapy, making these infections more available for human infection by direct or indirect means (3). The prospect of treatment failure, especially in lifethreatening situations, is perhaps the most intuitively obvious and serious of these impacts. However, an expert panel assessing salmonellosis risks from the subtherapeutic use of penicillin and tetracycline in animal feed considered that the greatest public health impact from resistance in Salmonella was probably the so-called ‘etiologic fraction’, those cases of salmonellosis that occurred because the infections were resistant (eg, associated with reduced colonization resistance from prior antimicrobial therapy), and the resistance was attributable to the use of antimicrobials in animals (4). Not surprisingly, the importance of antimicrobials to human health is one of the criteria being used to assess risks of nonhuman uses of antimicrobials, along with considerations of the organisms involved, methods of antimicrobial treatment in animals and other factors (5,6). Classification of a drug or class as critically important for human health may include consideration of its importance in treatment of enteric infections in humans, whether it is the only available therapy or one of few alternatives for treating serious human disease, and whether there is cross-resistance with other highly important drugs or known linked resistance to other important classes. Examples include third-generation cephalosporins, fluoroquinolones and glycopeptides, among others. Physicians and veterinarians are accustomed to assessing risks to health using direct evidence that may be acquired from