Continuing to treat urinary tract infections as a short-term, routine ailment rather than a long-term food safety issue risks turning the responsible bacteria into a major health crisis.
Adrienne LeBeouf recognized the symptoms when they started. The burning and the urge to head to the bathroom signaled a urinary tract infection, a painful but everyday annoyance that afflicts up to 8 million U.S. women a year. LeBeouf, who is 29 and works as a medical assistant, headed to her doctor, assuming that a quick course of antibiotics would send the UTI on its way.
That was two years ago, and LeBeouf has suffered recurring bouts of cystitis ever since. She is one of a growing number of women, and some men, who have unknowingly become infected with antibiotic-resistant versions of E. coli, the ubiquitous intestinal bacterium that is the usual cause of UTIs.
There is no national registry for drug-resistant infections, and so no one can say for sure how many resistant UTIs there are. But they have become so common that last year the specialty society for infectious-disease physicians had to revise its recommendations for which drugs to prescribe for cystitis -- and many infectious-disease physicians and gynecologists say informally that they see such infections every week.
Dr. Jehan El-Bayoumi, LeBeouf's physician and an associate professor of medicine at George Washington University Medical Center, said she has seen "a really significant increase, especially within the past two to three years."
But the origin of these newly resistant E. coli has been a mystery -- except to a small group of researchers in several countries. They contend there is persuasive evidence that the bacteria are coming from poultry. More precisely, coming from poultry raised with the routine use of antibiotics, which takes in most of the 8.6 billion chickens raised for meat in the U.S. each year.
Their research in the United States, Canada, and Europe (published most recently this month, in June, and in March) has found close genetic matches between resistant E. coli collected from human patients and resistant strains found on chicken or turkey sold in supermarkets or collected from birds being slaughtered. The researchers contend that poultry -- especially chicken, the low-cost, low-fat protein that Americans eat more than any other meat -- is the bridge that allows resistant bacteria to move to humans, taking up residence in the body and sparking infections when conditions are right. Touching raw meat that contains the resistant bacteria, or coming into environmental contact with it -- say, by eating lettuce that was cross-contaminated -- are easy ways to become infected.
"The E. coli that is circulating at the same time, and in the same area -- from food animal sources, retail meat, and the E. coli that's causing women's infections -- is very closely related genetically," said Amee Manges, Ph.D., an associate professor of epidemiology at McGill University in Montreal who has been researching resistant UTIs for a decade. "And the E. coli that you recover from poultry meat tends to have the highest levels of resistance. Of all retail meats, it's the most problematic that way."
Policy concern over antibiotic-resistant bacteria -- where they come from and how they affect human health -- is at a peak right now.
About 80 percent of the antibiotics sold in the United States each year are given to livestock as "growth promoters" that allow animals to put on weight more quickly, or as prophylactic regimens that protect against the confined conditions in which they are raised. (That figure, taken from FDA documents, is not universally accepted; the Animal Health Institute, an industry group, puts non-human use closer to 28 percent.) For decades, public health and agriculture have been at loggerheads over the practice. Health officials argue that these uses create resistant bacteria that move off large-scale farms via wind, water, dust, and in the animals themselves and the meat they become -- and create difficult-to-treat human infections. Agricultural interests counter that human infections have far more to do with medical misuse of antibiotics than with farming, and that the cost of stopping the drugs would be too great for producers to bear.
The U.S. Food and Drug Administration which regulates agricultural use of antibiotics, has been aware for decades of evidence that farm overuse of antibiotics creates resistant human infections, but has done little to help. In 1977, the agency proposed withdrawing its own approvals for penicillin and tetracycline use as growth promoters, and the proposal remained on the books even though the FDA was repeatedly stymied by legislative opposition. Last December, the agency actually gave up, and announced that it was cancelling its then 34-year-old attempts, opting instead for a voluntary approach. But this March, and again in June, a district court judge in New York City ruled the FDA must go through with its original program for re-examining agricultural antibiotic use, including holding hearings to examine the drugs' off-farm effects.
The proposed link between resistant bacteria in chickens and those causing UTIs is not the first time researchers have traced connections between agricultural antibiotic use and human illness. But because the UTI epidemic is so large and costly, the assertion that it might be tied to chicken production has brought renewed attention to the issue.
Investigators have been examining a possible link between growth promoters, chickens, and human infections since at least 2001, when Manges and others published in the New England Journal of Medicine an analysis of clusters of UTIs in California, Michigan, and Minnesota. The striking thing at the time was that the clusters appeared to be outbreaks caused by very similar E. coli strains that were resistant to the common drug Bactrim. In the United States, one out of every nine women has a UTI every year. If a single small group of E. coli was causing some proportion of the infections, that would be alarming -- but it might also offer a clue to defusing the overall epidemic. Initially, though, the researchers had no idea where the strains were coming from.
As a follow-up, Manges and other investigators looked for vehicles that might be transporting particular E. coli strains. That was an unusual challenge, because E. coli is one of the most common organisms on the planet, with a huge variety residing in the guts of humans and every warm-blooded animal, and in reptiles and fish as well. The particular subset of strains they examined are called "ExPEC," for "extra-intestinal pathogenic E. coli" -- that is, E. coli that escapes the gut to cause illness elsewhere in the body, including in the urinary tract.
ExPECs were already a medical-research concern, because E. coli that moves from the gut into the bladder may not stay there. Infections that are not treated can climb up to the kidneys and enter the bloodstream. ExPEC E. coli cause up to 40,000 deaths from sepsis -- the most serious form of bloodborne bacterial infection -- in the United States each year, and since about 2000, antibiotic resistance in ExPEC strains has been climbing.
In 2005, University of Minnesota professor of medicine Dr. James R. Johnson published results of two projects in which he analyzed meat bought in local supermarkets during 1999-2000 and 2001-2003. In both cases, he found resistant ExPEC E. coli strains that matched ones from human E. coli infections. Other researchers soon found similar matches in meat--particularly poultry--from across Europe, in Canada, and in additional studies from Minnesota and Wisconsin.
In that research, investigators began to sort out two things. They became convinced that the resistance pattern could be traced back to animal antibiotic use, because resistance genes in the bacteria causing human infections matched genes found in bacteria on conventionally raised meat. And they began to understand that E. coli's complexity would make this new resistance problem a difficult one to solve. The strains that cross to humans via poultry meat "don't establish themselves as big, successful lineages" of bacteria that would be easy to target, Johnson said. "But collectively they can cause a lot of infections, because there are just so many of them and they're so diverse."
There has been no way, to this point, to prove that a single specific UTI arose from a portion of meat that in turn came from a single animal given antibiotics. The investigators tracing the connection acknowledge this is a weakness in their case, but point out that modern medical ethics do not permit experimenters to deliberately cause infections in healthy humans as a way to prove a disease risk. What researchers do, in cases like this, is to gather evidence from big groups of people that shows a disease emerging on a population level -- and based on the molecular evidence from animals, meat, and humans, they believe they have done so with ExPEC E.coli from chicken and UTIs.
Not everyone agrees, of course. Dr. Charles Hofacre, professor at the University of Georgia's Center for Food Safety and an officer of the American Association of Avian Pathologists, points out that while the resistance factors in chicken- and human-associated bacteria resemble each other, no study has yet proven that a transfer occurs. Antibiotic resistance is so common, Hofacre said, that "it isn't surprising that genes carried by human E. coli are going to be similar to resistance genes in chicken E. coli -- or pig E. coli, or salamander E. coli." He adds: "That doesn't necessarily mean the antibiotic resistance genes in the human came from the salamander, or the chicken or the pig."
Dr. Randall Singer of the University of Minnesota's College of Veterinary Medicine points out that some recent research suggests that antibiotic resistance genes in E. coli may actually originate from humans, spreading through sewage into ground and surface waters, and from there into the environment and livestock. The resistance found in human and poultry E. coli "is a typical multi-drug resistant pattern that you find all over the world, including in wild animal populations that have had no exposure to" humans, he said. "To say these genes exist in a person because of an antibiotic that was given to a chicken is too narrow an interpretation."
On the front lines of medicine, physicians report that they regularly see rising amounts of resistant infections in patients for whom the resistance has no obvious explanation -- for example, in patients who have not been treated in a hospital or other health-care facility where antibiotics might have been overused or misused. Because they are front-line physicians, and not microbiologists, these doctors do not analyze their patients' diets and match their infections to any animal strains. But when they do perform enough genetic analysis of their patients' infections to be able to tell which drugs will work, they see the same resistance factors in their patients' E. coli that Johnson, Manges, and others have spotted in their research. And for many of them, the proposed connection between agricultural antibiotic use, resistant animal infections, and resistant human infections makes intuitive sense. And particularly in the case of the new outbreaks of UTIs.
"Medicine certainly does contribute to [antibiotic-resistant bacteria], but there have been studies of other infectious diseases that have implicated animals and antibiotics in propagating certain types of infections," said Dr. Connie Price, chief of infectious diseases at Denver Health & Hospital in Colorado. "It makes sense to me that resistant urinary tract infections could absolutely be one of those."
In Washington, El-Bayoumi said resistant UTIs are common among her patients, describing one woman whose infection did not respond to the first drug she tried but did to the second, and another whose infection recurred despite rounds of three different antibiotics before finally responding to a fourth drug. She has treated LeBeouf for nine recurrences so far without ever being able to eradicate her multi-drug resistant infection. "It stops for a while, and then it eases back in," said LeBeouf, who describes losing work hours and sleep time to the nagging pressure and pain. "We do a urine culture to see what medications will work. Dr. El-Bayoumi's at the point where she is saying, 'I don't know what else we can do.'"
People unlucky enough to contract these infections describe a consistent pattern. They assume they have an ordinary UTI, go to their doctors for treatment, get a prescription, and feel better for a few days -- and then are puzzled to find that the same painful symptoms are recurring, and they have to return to the doctor again.
Because UTIs are such an everyday occurrence, the problem of rising resistance -- along with the question of where the resistance comes from -- has not been a major priority for medicine. Nor has tracing the possible cause back to chicken: by the time women realize they need treatment, they usually have long forgotten when and how they might have been in contact with raw meat, and their doctors are seldom epidemiologists.
"We tend to dismiss bladder infections as trivial," said Dr. Richard Colgan, an associate professor at the University of Maryland School of Medicine. "But a woman who gets one -- and they mostly occur in women -- usually endures symptoms for an average of a week until she can get treated. She usually has to miss school or work on average of one week. A woman on average will postpone sexual relations for a week."
The victims are not always women. And the infections are not always uncomplicated. The cost in the United States of treating UTIs runs more than $1 billion per year, including hospitalizations for the most serious complications and intermediate care for patients whose infections are resistant to the easy-to-administer drugs.
There have been a few times, in the past few decades, where disease-causing E. coli crossing to humans from meat became a national priority. The poster-child case is E. coli O157:H7, which became notorious after the 1993 Jack-in-the-Box hamburger outbreak in which hundreds were sickened and three children died; in response, the U.S. Department of Agriculture declared the O157 strain an adulterant, making it illegal to distribute. But in contrast, it took almost two more decades -- until September last year -- for other similar strains to be declared adulterants as well.
Researchers who have been tracking the highly resistant E. coli wonder what it will take for these strains to have their Jack-in-the-Box moment. They cause more illness than O157 -- but in a diffuse, slow-moving epidemic that even the victims may not know they are part of, like the current outbreaks of antibiotic-resistant UTIs. And defusing this one will be far more politically complex, because it will require addressing the economic imperatives that drive farmers to use antibiotics -- and consumers' role in supporting large-scale agriculture as well.
"I see people voting with their feet, buying cheap produce, meat that is less expensive, eggs that are less expensive," said Dr. Jorge Parada, professor of medicine and infectious disease at Loyola University's Stritch School of Medicine in Chicago. "My personal point of view is, this is unsustainable in the long run. It has a whole series of side effects that are not negligible, and antibiotic resistance is important among them."
This article was produced in collaboration with the Food and Environment Reporting Network, an independent, nonprofit news organization producing investigative reporting on food, agriculture and environmental health.
We want to hear what you think. Submit a letter to the editor or write to letters@theatlantic.com.