One safety concern that has been raised in connection with antibiotic use in agriculture is the transfer of antibiotic-resistant bacteria of animals into the human colon. At first, concern about transfers of this type was limited to foodborne pathogens such as Salmonella and Campylobacter. More recently, attention has focused on a different group of bacteria and their genes, bacteria such as antibiotic-resistant enterococci that might be able to colonize the human colon or spread their genes to bacteria that normally reside in the colon. Here, the concern is not about immediate disease but about increased risk of later developing a post-surgical infection that would be difficult to treat successfully with antibiotics. There is evidence that antibiotic use in agriculture selects of antibiotic-resistant bacteria in the intestines of animals fed antibiotics and that antibiotic-resistant bacteria can be isolated from meats or milk products sold to consumers. There is some evidence suggesting that animal enterococci would colonize only transiently because they are unable to compete with indigenous strains of enterococci, but this conclusion is still not well-documented. Even if colonization is transient, the animal bacteria could transfer their genes to colonic bacteria. There is some evidence that such transfers do occur in the colon, but the evidence is largely indirect. More work is needed to establish precisely the degree to which antibiotic use in agriculture affects human health.
The Debate Over Agricultural Use Of Antibiotics as Growth Promoters Intensifies
Is there a link between widespread use of antibiotics as growth promoters and the increased incidence of antibiotic-resistant bacterial pathogens? This issue has become a matter of serious public concern. Over the past year or two, news articles on agricultural use of antibiotics have appeared in U.S. newspapers and news magazines. Political action groups such as the Environmental Defense Fund, the Humane Society, and Citizens for Science in the Public Interest have taken public stands against the use of antibiotics as growth promoters. In January, 1999, the U.S. Food and Drug Administration proposed new regulatory guidelines that would probably have the effect of eliminating most of the antibiotics currently used as growth promoters. The FDA proposal has been controversial, but has had the effect of increasing interest in the growth promoter issue.
The recent decision of the European Union to ban a number of antibiotics for use as growth promoters has raised the political and economic stakes even further by raising the possibility that the EU might restrict imports of meats and produce from countries that do not have similar restrictions on antibiotic use. The new stance of the EU is an improvement over past actions. The EU had previously taken the very imprudent step of approving avoparcin (a vancomycin analog) for use as a growth promoter, a decision that has led to the introduction of vancomycin-resistant enterococci into the European food supply. Thus, although European hospitals had succeeded in keeping vancomycin-resistant enterococci from being the kind of hospital-associated problem it is in the U.S., the same bacteria are now moving through the food supply. No one is sure how this release will ultimately affect the number of vancomycin-resistant enterococcal infections in humans - infections that will be difficult or impossible to control with antibiotics - but it is an experiment most of us would have preferred not to do. The decision of the EU to ban not only avoparcin but other antibiotics that cross-select for resistance to important human use antibiotics was made largely on the basis of political considerations rather than scientific ones. Clearly, a dispassionate and logical analysis of the pros and cons of the use of antibiotics as growth promoters is needed to temper the debate that is currently underway.
The purpose of this article is to take a critical look at the question of whether agricultural use of antibiotics is likely to have an important impact on human health. As will become evident, there are still a number of missing information links in our understanding of how actions taken on farms affects the population at large. Yet, some of the evidence that has come to light in recent years is troubling enough to justify a sober, rational reassessment of agricultural use of antibiotics and perhaps some further studies to answer pressing practical questions for which we do not at present have answers.
The potential problems associated with agricultural use of antibiotics fall into two separate categories. The most direct effects of antibiotic use in agriculture will be on the humans and animals exposed daily to the antibiotics. This aspect of agricultural antibiotic use is seldom discussed but could well be having an important impact on farmers themselves. The second category of potential problems, the ones involving the general public, arises from antibiotic resistant bacteria moving through the food supply into the intestinal tracts of consumers. The latter connection assumes a series of steps that are summarized in Table 1. In this paper, we will examine the evidence for each of these steps.
Table 1. Hypothetical steps involved in movement of antibiotic-resistant bacteria from farm to the consumer. Asterisks indicate the steps supported by the most scientific evidence.
|
*Step 1. Antibiotic use on the farm selects for antibiotic-resistant bacteria. *Step 2. Antibiotic-resistant bacteria from the intestines of animals contaminates meet and produce that are purchased by consumers. *Step 3. Antibiotic-resistant bacteria colonize the intestines of consumers or transfer their resistance genes to bacteria normally found in the human intestinal tract. *Step 4. Person colonized by antibiotic-resistant bacteria is at higher risk for later development of an untreatable post-surgical infection. |
Does Antibiotic Use In Agriculture Select For Resistant Bacteria?
The answer to this question might seem to obvious, but the connection between antibiotic use and the emergence of antibiotic-resistant bacteria can be complex. The first issue is the species of bacterium involved. There are some bacteria such as staphylococci, enterococci and pseudomonads that seem to become resistant very quickly after an antibiotic is in use. In fact, some of these bacteria - for reasons we still do not understand completely - may already be resistant to an antibiotic before it is introduced. Sometimes this is due to characteristics of the bacteria that make them naturally resistant. For example, most Gram-negative bacteria are naturally resistant to vancomycin because the porins in their outer membranes prevent the antibiotic from gaining access to peptidoglycan. In other cases, actual resistance genes are present in advance of the use of an antibiotic. For example, we found strains of Bacteroides isolated prior to the use of tetracycline that carried the tetracycline resistance gene tetQ (unpublished data). Recently, in a letter to Lancet, Schwalbe et al. (1999) reported finding vancomycin-resistant enterococci in animal feed in the U.S. Avoparcin has never been approved for use in the U.S., and carriage of vancomycin-resistant enterococci by people outside hospitals is rare in the U.S (Coque et al., 1996). Where did these resistant bacteria come from?
Just as there are bacterial groups that seem to become resistant very rapidly, there are bacterial groups that develop resistance very slowly or not at all despite exposure to antibiotics. Streptococcus pneumoniae took decades to develop resistance to penicillin. Streptococcus pygenes is still susceptible. Both of these bacteria colonize the human body regularly and are exposed constantly to antibiotics. In these cases, the answer may be that to become resistant to penicillin, these Gram-positive bacteria had to mutate important housekeeping genes involved in peptidoglycan biosynthesis and for some reason this was harder for them to do this than it was for staphylococci and enterococci. Whatever the explanation, the relationship between antibiotic use and development of resistance clearly differs from one species to another.
In agriculture, the few careful studies that have been done show that there is also variation in the levels of resistance from one type of animal to another. In the DANMAP reports, bacteria from cattle are generally less resistant than those in pigs and chickens (Aaerstrop et al., 1998; DANMAP, 1997). Why is this? One obvious possibility is that the way antibiotics are administered makes a difference.
Long term, low dose use of antibiotics is the pattern most likely to select for antibiotic-resistant bacteria because the antibiotic concentration is low enough to allow the bacteria to grow at some level but high enough to exert a selective pressure in an already competitive environment. Moreover, long-term exposure of bacteria to antibiotics is the type of exposure that creates bacteria whose resistance is so stable that the resistant strains can compete with susceptible strains after the antibiotic is no longer present (Salyers and Amabile-Cuevas, 1997; Scrag et al., 1997). Long term exposure allows bacteria not only to become resistant but to make compensatory mutations that reduce the fitness toll of the resistance proteins on the bacteria in the absence of antibiotic. Everything we know about the emergence of resistance and the relative stability of the resistance phenotype suggests that we should divide resistant bacteria into two categories - those that cannot compete with sensitive strains if antibiotic selection is removed and those that can. Conditions that select for very stable carriage of resistance genes should be more tightly monitored and controlled than those that do not.
Although we can explain many patterns in the rise or fall of resistance on the basis of antibiotic use, it is troubling that there are still cases in which we cannot explain the patterns of resistance we see. Certainly, antibiotic use is associated with an increase in incidence of antibiotic-resistant bacteria in a general way, but there are many variations on this general theme. Are there thresholds of antibiotic exposure, both in terms of length of administration or in amount of antibiotic encountered? We do not have enough information to answer this question, and the answer is almost certainly different for different groups of bacteria.
Another twist to this issue is the likelihood that there are non-antibiotic selections for antibiotic-resistant bacteria. Bacteria often form resistance gene clusters and any drug that selects for one resistance in the cluster selects for maintenance of the entire cluster (Hall and Collis, 1995). For example, if a bacterium is resistant to penicillin and mercury, mercury pollution will select for maintenance of this strain in the environment and thus select indirectly for penicillin resistance. Are there other farm practices outside of antibiotic use, or industrial practices far from the farm, that contribute to the antibiotic resistance problem other than feeding antibiotics? Also, are there general environmental factors not taken into account? Recently, scientists discovered that some antibiotics survive the process for treatment of human sewage and that antibiotics such as fluoroquinolones can be detected in water effluents from sewage treatment plants (Raloff, 1998). Scientists doing large farm surveys like the very ambitious DANMAP surveys rarely have the time or the money to sample the environment outside the farm being studied to see whether there are forces other than farm use of antibiotics at work. The relationship between antibiotic use in agriculture and the emergence or non-emergence of resistant strains is a long way from being understood in its completeness.
Possible Impact Of Antibiotic-Resistant Bacteria On Farmers and Their Animals
Before going on to discuss the problem of resistant bacteria in the food supply and their potential effect on the general populace, let us first take a look at a problem that has been largely overlooked: the potential effect of antibiotic-resistant bacteria on farmers and their animals. There are anecdotal accounts from physicians encountering antibiotic failures when treating members of farm families, presumably due to long-term exposure to antibiotics while handling antibiotic-laced feed. There are also anecdotal accounts of infections that destroyed a farm because no antibiotics worked. On the other side, most farmers and their families certainly seem healthy for the most part. And antibiotics still cure animals of infections in most cases. Which of these pictures is the correct one or is the true picture some of both?
It is time to get past the "urban legend" level of anecdotal accounts and to do a thorough assessment of the effect of antibiotic use on the general health of farm workers and farm animals. Like it or not, farm workers and farm animals are the "canary in the mine" that should give early warning of potential problems. Yet, we have no concrete information on whether there have been a significantly higher incidence of treatment failures - in farm workers or farm animals - in places where antibiotic use is highest. There is evidence, however, that farmers can become colonized by bacteria resistant to the antibiotics being used on the farm (see, for example, Simonsen et al., 1998; Van den Bogaard, 1997).
Antibiotic-Resistant Bacteria In The Food Supply
In countries where antibiotic use in agriculture is heavy, it has been easy to find antibiotic-resistant bacteria in foods offered for sale to the consumer (see, for example, DANMAP, 1997; Jensen, 1998; Klein and Reuter, 1998). What is the evidence that these bacteria come from farms and not from contamination by food processors who have become colonized with resistant bacteria in their community due to human use of antibiotics? Perhaps the best evidence comes from studies tracking vancomycin resistant enterococci or streptogramin-resistant enterococci in foods. In the U.S., vancomycin and its analogs have never been approved for use outside human medicine and most of the medical use is in hospitals. In Europe, by contrast, avoparcin (a vancomycin analog) was in use for a few years before it was banned. Vancomycin resistant enterococci were found in European but not U.S. poultry during the avoparcin use period and now that avoparcin has been banned, the incidence of vancomycin-resistant enterococci has been declining (Klare et al., 1999; Pantoski et al, 1999). These two trends strongly suggest that avoparcin use in agriculture was responsible for the rise in vancomycin-resistant enterococci, especially in view of the fact that vancomycin resistant enterococci have not emerged as a problem in European hospitals they way they have in U.S. hospitals.
Similarly, streptogramins have been used widely in agriculture both in Europe and the US (e.g. virginiamycin) but not in human medicine until the release of Synercid the first human use streptogramin preparation. If early reports of Synercid resistant bacteria in the food supply (see, for example, Welton et al., 1998) - but not yet in hospitals - are born out, this would be another example of resistance that clearly emerges from agriculture and enters the human food supply. As mentioned earlier, however, it is important to understand that the answer for enterococci, vancomycin and Synercid may not be the answer for all antibiotics and all bacterial species.
In most reports of antibiotic-resistant bacteria in the food supply, meat - especially poultry - has been the focus. This focus is arbitrarily narrow. Milk products should also be checked. There have been a few reports of antibiotic-resistant bacteria in cheese (Teuber et al., 1996), but more thorough investigations are needed. Vegetables and fruits also need to be checked. Vegetables may have been irrigated or sprayed with water contaminated with manure runoff from animal production operations. Also, antibiotics sprayed on fruit trees may be contributing to the entry of resistant bacteria into the food supply. It is ironic that chicken, and pork have received so much attention in connection with contamination by antibiotic-resistant bacteria. These meats are usually well-cooked, so that resistant bacteria would be killed. By contrast, fruits and some vegetables are often consumed uncooked. It is time for a look for antibiotic-resistant bacteria on produce.
The existing scientific literature on whether enterococci of animal origin can colonize the human intestine is confusing. Some studies find evidence for colonization of humans by animal strains but others do not (see, for example, Woodford et al., 1998). Whether animal strains can colonize animals and vice versa is not nearly so important as the fate of the resistance genes they carry. If bacteria moving through the intestinal tract transmit resistance genes to bacteria that normally colonize the colon, these bacteria may in turn cause infections later or transfer their resistance genes on to pathogens such as Streptococcus pneumoniae or Staphylococcus aureus, which pass through the colon on a routine basis.
Gene Transfer In The Human Colon - How Likely and How Often?
There are two ways to follow gene transfer in the intestinal tract. One is to introduce an antibiotic-resistant strain of bacteria and watch what happens to its antibiotic resistance genes. There are very few examples of experiments of this type and in those few cases mice were used as the experimental animal. Such experiments involving real food animals such as chickens, pigs or cattle are expensive and time-consuming and thus are unattractive to funding agencies. A second approach is to take an archeological approach and ask whether the same resistance gene is found in unrelated bacteria. Finding virtually identical copies of resistance genes in genera and species that are only distantly related is good evidence that some sort of gene transfer has occurred (Salyers and Shoemaker, 1996; Teuber et al., 1996). Since this approach is cheap and relatively simple experimentally, it is not surprising that most of the evidence for gene transfers in the human intestinal tract is of this type.
It has not been difficult to find examples of virtually identical resistance genes in bacteria isolated from foods, bacteria found in the human intestine and bacteria isolated from infected humans (see, for example, Jensen, 1998; Teuber et al., 1996; van den Braak et al., 1998). The ease with which such examples have been found supports the hypothesis that resistance gene transfers between foodborne, intestinal and disease-causing bacteria are rather common. A limitation of this type of analysis is that we do not know how a gene came to be in species A and species B. Was it direct transfer from A to B, from B to A or between them via a series of intermediate species C, D, E and F? Even if we cannot - on the basis of the archeological evidence - answer these questions, the fact that there are obviously genetic conduits open between bacteria in food, bacteria in the human intestine and bacteria that cause disease in other parts of the body should be a cause of concern.
What Do We Do With This Information?
Lobbyists employed by organizations that represent farmers have tended, understandably, to use the gaps in the scientific literature to argue that there is no cause for concern. I can sympathize with this argument and also with the enthusiastic approval by farmers of arguments that show they are not the "bad guys" in antibiotic resistance. Finally, I agree completely that the real "bad guys" are physicians who have over-prescribed and generally abused antibiotics and that the contribution of agricultural use of antibiotics to resistance problems in human medicine has not been conclusively proven. Yet, I am concerned that this temporary sense of rectitude on the part of farmers and their representatives could prove illusory. I have had the misfortune to serve as an expert witness in lawsuits over human disease issues not connected to agricultural use of antibiotics, and my experience has caused me to conclude that - flawed as it is - the chain of evidence in the current literature, viewed by an ordinary juror and influenced by an "expert" less conscientious than myself could lead to a successful outcome of a lawsuit brought by relatives of someone who died of an antibiotic resistant bacterial infection against agricultural organizations.
There is also the problem of public perception of an issue. I am not a pollster, but I am confident that if the general public were polled on whether they think use of antibiotics to treat human disease is optional or essential, they would answer "essential". And if they were asked whether agricultural use of antibiotics as growth promoters is optional or essential, they would answer "optional". Farmers can make good arguments against this assessment, but they have not so far made these good arguments to the public in an effective way.
If I were a farmer, and in a
sense I am - a farmer of bacteria - I would be looking for the
positive side of the current picture. The organizations that
represent farmers have been uniformly the prophets of gloom and
doom. In the short term, this works. As we can see from the European
example, rational arguments do not convince frightened politicians.
Is there a silver lining to this very dark cloud? As a group,
farmers have been the most resilient and creative members of
the economy when it comes to responding to natural disasters.
In recent years, farmers have been bedeviled by overproduction.
This may be the opportunity to cut back on production, a move
that might benefit everyone in the farm community if it causes
an increase in prices and a more stable and predictable market.
Finally, I see this debate as an opportunity for farmers to tell
their story to the public and to help the public understand the
realities and strains farmers currently face. In the end, this
story could help to generate sympathy for farmers. Given that
public pressure for lower food prices has had the unintended
effect of addicting farmers to antibiotics, is it fair to withdraw
abruptly and without aid the drugs that have made food prices
lower as a percentage of the average person's earnings than anytime
in history?
[Key references are provided here, but a more complete set of references can be obtained by visiting the ROAR (reservoirs of antibiotic resistance) web site: www.roar.antibiotic.org.]
Aaerstrup, F. M., F, Bager, N. E. Jensen, M. Madsen, A. Meyling and H. C. Wegener. (1998). Surveillance of antimicrobial resistance in bacteria isolated from food animals to antimicrobial growth promoters and related therapeutic agents in Denmark. APMIS 106: 606.
Coque, T. M., J. F. Tomayko, S. C., Ricke, P.C. Okhyusen, B.E. Murray. (1996) Vancomycin-resistant enterococci from nosocomial, community and animal sources in the United States. Antimicrob. Agents Chemother 40, 2605.
DANMAP. 1997. Consumption of antimicrobial agents and occurrence of antimicrobial resistance in bacteria from food animals, food and humans in Denmark. No. 1, 1997, Danish Zoonosis Centre, Danish Veterinary Laboratory, Bulowsvej 27. D-1790 Copenhagen V.
Hall, R. M., C. M. Collis. (1995). Mobile gene cassettes and integrons: capture and spread of genes by site-specific recombination. Molecular. Microbiology 15: 593.
Jensen, L. B. (1998) Differences in the occurrence of two base pair variants of Tn1546 from vancomycin-resistant enterococci from humans, pigs and poultry. Antimicrobial Agents and Chemotherapy 42:2463.
Klare, I., D. Badstubner, C. Konstabel, G. Bohme, H. Clause, W. Wittee. (1999) Decreased incidence of VanA-type vancomycin-resistant enterococci isolated from poultry meat and from fecal samples of humans in the community after discontinuation of avoparcin usage in animal husbandry. Microbial Drug Resistance 5: 45.
Klein, G., Pack, A. and Reuter, G. (1998) Antibiotic resistance patterns of enterococci and occurrence of vancomycin-resistant enterococci in raw minced beef and pork in Germany. Applied and Environmental. Microbiology 64, 1825.
McDonald, L. C., M. J. Kuehnert, F. C.., Tenover, W. R. Jarvis. (1997) Vancomycin-resistant enterococci outside the health care setting: Prevalence, sources and public health implications. Emerging Infectious. Diseases 3, 311.
Pantosti, A., M. Del Grosso, S. Tagliabue, A. Marci, A Caprioli (1999) Decrease of vancomycin-resistant enterococci in poultry meat after avoparcin ban. The Lancet 354: 741.
Raloff, J. (1998). Drugged waters. Science News 153: 187.
Salyers, A. A. and C. F. Amabile-Cuevas.
(1997). Why are antibiotic resistance gene so resistant to elimination?
Antimicrobial Agents and Chemotherapy 41: 2321
Salyers, A. A. and N. B. Shoemaker. (1996). Resistance gene transfer
in anaerobes: New insights, new problems. Clinical Infectious
Diseases. 23 (Suppl.): S36.
Schrag, S. J., V. Perrot, and B. R. Levin. (1997). Adaptation to the fitness costs of antibiotic resistance in Escherichia coli. Proceedings. Royal Society London. B. Biological. Sciences. 264: 1287.
Schwalbe, R. S., A. C. McIntosh, S. Qauyumi, J.A. Johnson, J. G. Morris. (1999) Isolation of vancomycin-resistant enterococci from animal feed in USA. Lancet 353: 722.
Simonsen, G. S., H. Haaheim, K.H. Dahl, H. Druse, A. Lovseth, O. Olsvik, A. Sundfjord. (1998) Transmission of VanA-type vancomycin-resistant enterococci and van A resistance elements between chicken and humans at avoparcin-exposed farms. Microbial Drug Resistance 4: 313.
Teuber, M., V. Perreten, F. Wirsching, F. (1996) Antibiotikumresistente bakterien: eine neue dimension in der lebensmittel-mikrobiologie. Lebensmittle-technologie 29: 182.
Van den Bogaard, A.E, L. B. van den Jensen, E. E. Stobberingh (1997). Vancomycin-resistant enterococci in turkeys and farmers. New England Journal of Medicine 337: 1558.
Van den Braak, N., A. Van Belkum, M. vanKeulen, J. Vliegenthart, H. Verbrugh, H. P. Endtz. (1998). Molecular characterization of vancomycin-resistant enterococci from hospitalized patients and poultry products of the Netherlands. Journal of Clinical Microbiology 36: 1927.
Welton, L. A., L.A. Thal, M.B. Perri, S. Donabedian, J. McMahon, J. W. Chow, M. J. Zervos. (1998). Antimicrobial resistance in enterococci isolated from turkey flocks fed virginiamycin. Antimicrobial Agents Chemotherapy. 42: 705.
Woodford, N., A-M. Adebiyi, M-F. Palepou and B. D. Cookson. (1998). Diversity of VanA glycopeptide resistance elements in enterococci from humans and nonhuman sources. Antimicrobial Agents Chemotherapy. 42: 502.
© Queen's Printer for Ontario, 2001
[Source: www.gov.on.ca/OMAFRA/english/livestock/animalcare/amr/facts/sayler.htm]