News

The case for animal antibiotics

Banning routine antibiotics will not combat “superbugs” and may increase rates of food poisoning, experts say

Source: Flickr/carlos_lorenzo

Since the 1950s farmers have fed antibiotics to animals at sub-therapeutic doses to promote growth and reduce the incidence of disease. But concerns that routine antibiotic use in animals could lead to the emergence of drug-resistant strains of human pathogens led the EU to ban the practice in 2006 and scientists, consumer groups and politicians have called on the US government to follow suit.

Questioning the evidence behind these concerns, an article published this week in Risk Analysis¹ argues that banning sub-therapeutic antibiotic use would have no benefit for human health — and might even cause harm by increasing the risk of disease from food-borne bacteria.

“The very real and serious threats of MRSA [methicillin-resistant Staphylococcus aureus] and MDR [multi-drug resistant] “super-bugs” in human patients can be addressed effectively only by changing the use of antibiotics in human medicine,” argue the authors, Tony Cox and Doug Popken, independent risk analysis consultants in Colorado, USA.

Opponents of sub-therapeutic antibiotic use argue that sub-therapeutic antibiotic use increases the risks of drug-resistant pathogens, by selecting for resistant strains of bacteria in animals and the environment which go on to infect humans or pass resistance genes on to human pathogens. The authors say that this argument, while emotive and even scientifically plausible, has little or no basis in evidence.

Although antibiotic use on farms does increase the number of resistant bacteria in the soil and groundwater nearby, there is no evidence demonstrating that agricultural use of antibiotics causes treatment failure due to antibiotic resistance in humans, claim the authors. Studies that have declared a link have generally misinterpreted the statistics, or are based on prior assumptions, such as a subjective estimate of the proportion of treatment failures attributable to agricultural use of antibiotics, rather than experimental evidence, they say.

On the other hand, meat produced without the use of antibiotics has a higher microbial load than that produced from animals fed with the drugs. Cox and Popken say this may increase the chances of food-borne illness in people. In an earlier analysis, using a model based on data on resistance rates in human cases of Enterococcus faecium, they estimated² that the cost to human health of this increase outweighs any possible increases in resistance resulting from the use of antibiotics in animals by up to 10,000 times.

The ban on the use of antibiotics at sub-therapeutic levels in Europe has not reduced antibiotic resistance rates, the authors point out. In fact, following the ban in Denmark, the rate of food poisoning increased, leading to greater prescription of human antibiotics and in increased rates of resistance to some antibiotics, according to a briefing paper by researchers at the University of Iowa, which was part-funded by the US National Pork Board and based on Danish government data.³

Morgan Scott, professor of epidemiology at Kansas State University, shares the authors’ concern that banning sub-therapeutic antibiotics will not have the intended effect. “Wholesale bans on classes or uses of antimicrobials may paradoxically hinder efforts to mitigate resistance in those classes of antimicrobials for which we are critically concerned,” he says. “If farmers use less sub-therapeutic antibiotics, will they use more therapeutics instead?” Even restricting just those antibiotics designated as critically important for health care could be counter-productive, Cox and Popken argue, “if such restrictions result in less healthy animals, more unkilled bacteria in meats, or greater use of antibiotics that pose a larger resistance threat.”

In the paper, Cox and Popken use sub-therapeutic use of tetracyclines as a case study to examine the likely effect of a ban. Tetracyclines are heavily used in agriculture and medicine, and high levels of resistance to tetracyclines have been found in many species of food-borne bacteria. However, tetracyclines are not typically used to treat food-borne illnesses and so this resistance does not pose a risk to human health, the authors point out.

They found that of the infectious diseases that are treated with tetracyclines, only three are potential hazards for resistance: gingivitis, acne, and illnesses caused by Mycobacterium fortuitum, a nontuberculous mycobacterium that is relatively rare in the USA. They do not consider the others a hazard for resistance in the USA, either because they are very rare in that country, such as malaria and anthrax, or because resistance has never been observed, despite decades of tetracycline use.

Of the three conditions identified as potential hazards for resistance, none pose a risk of treatment failure due to antibiotic use in animals, according to the authors. Drug-resistant gingivitis can still be successfully treated with the recommended dose of tetracycline, and M. fortuitum is only treated with tetracycline following susceptibility testing, so resistance doesn't lead to treatment failure. Resistance in acne is more of a problem, occuring in up to 30% of cases, but the authors point out that rates of resistance are similar between the USA, where tetracyclines are heavily used in agriculture, and the UK, where they have not been used as growth promoters for nearly 40 years.

For the diseases in which resistance has not been observed, such as stomach ulcers caused by Helicobacter pylori, it is possible that resistance has emerged but not yet been detected, according to the authors. However, any such resistance would have to be very rare to have not been detected, they argue. They estimate the maximum likely incidence of resistance to treatment that could exist for these conditions in the USA, and conclude that “the net effect [of a tetracycline ban] would be to prevent only perhaps one treatment failure every two centuries.”

Scott agrees that a ban on tetracycline use would have little or no effect on resistance in the short term, but contends that the analysis ignores the time period over which resistance can emerge. “The effects [of antibiotic use] are cumulative over decades and reflect temporal scales and associated evolutionary changes to bacterial populations that we cannot readily fathom in our risk management time frames,” he says.

Reference and links

1.	Cox LA and Popken DA. Assessing Potential Human Health Hazards and Beneﬁts from Subtherapeutic Antibiotics in the United States: Tetracyclines as a Case Study. Risk Analysis 2010; 30:432–457. doi: 10.1111/j.1539-6924.2009.01340.x
2.	Cox LA and Popken DA. Quantifying human health risks from virginiamycin used in chickens. Risk Analysis 2004; 24:271–288. doi: 10.1111/j.0272-4332.2004.00428.x
3.	Hayes DJ and Jensen HH. Lessons from the Danish Ban on Feed-Grade Antibiotics. Iowa State University Center for Agricultural and Rural Development Briefing Paper 03-BP 41. Article