Is a new toxic C. diff strain emerging in the food supply?

Is a new toxic C. diff strain emerging in the food supply?

An unsettling theory about rise in community cases

The continuing emergence of a highly toxigenic strain of Clostridium difficile in the community has raised a troubling question for public health investigators: Is C. diff becoming a foodborne pathogen?

Though more investigation remains to establish a clear link, one disturbing possibility is that the emerging new epidemic strain of C. diff (ribotype 027) — which can have 20 times the toxins as typical strains and has caused deadly hospital outbreaks — is spreading in the community through meat sold at grocery stores.

"There is concern about the food supply," says Clifford McDonald, MD, medical epidemiologist in the division of health care quality promotion at the Centers for Disease Control and Prevention. "There is no evidence that I am aware of that C. diff is a foodborne illness, but the reason for concern is this burden of disease in the community. Recent reports [show] that C. diff can be found in retail meats, including strains that have caused human disease. A public health priority right now to determine whether it is or can be a foodborne illness."

The foodborne theory is one of several under discussion as epidemiologists track increasing infections in the community by a pathogen normally acquired in the hospital. Noscomial C. diff typically spreads via the hands of health care workers from the contaminated environments of patients whose primary symptom is often diarrhea. However, the traditional picture is changing as C. diff threatens to arise independently in the community such as community-associated methicillin-resistant Staphylococcus aureus (CA-MRSA). The implications are serious because the disease burden of C. diff already is significant, causing more than 400,000 cases of severe diarrhea and colitis and more than 5,000 patient deaths in the United States annually.

Establishment of the epidemic strain in the community would certainly threaten hospital patients via incoming admissions because hospitals who have battled outbreaks of the bug report strikingly high mortality rates for C. diff infections. Fatal infections also have been reported in people in the community, most recently in an otherwise healthy woman in Pennsylvania. Even in milder manifestations the sheer prevalence of community cases is unusual. For example, investigators looking at a group of hospitals in North Carolina found that nearly 20% of C. diff cases appeared to have been acquired in the community.

Epidemic strain detected in meat

The CDC first reported the phenomenon of community-associated C. diff in 2005, citing reports in four states of infections in patients previously thought to be at low risk for C. diff. Considered in the context of recent high-morbidity, hospital-associated outbreaks in North America, Great Britain, and the Netherlands, the cases of severe C. diff appear to reflect a "changing epidemiology," the CDC reported.¹ More evidence is being accumulated in the form of continuing cases in otherwise healthy people in the community, many of whom have no history of recent hospitalization or prolonged antibiotic use that typically precedes C. diff infection.

"There is concern about the source of transmission," McDonald says. "Is there person-to-person transmission? We had a little bit of evidence of that from [previous reports] but it was anecdotal." Indeed, some family clusters have been noted, but suspicions also have turned to a possible common source such as food.

"We found that some of our isolates from animals and some isolates from human disease had identical pulse field gel electrophoresis patterns," says Glenn Songer, PhD, a veterinary microbiologist at the University of Arizona in Tucson. "That suggests a connection between food animals and humans as far as disease goes. And since humans eat domestic animals fairly commonly and the reverse happens only under rare and unusual circumstances, we thought it was probably worth a try to look for it within food."

Somewhat surprised that no one had done it before, Songer tested for C. diff in ground beef, turkey, and pork in Tucson markets. "Frankly, I didn't enter into it with much hope of success, mainly because you figure that is something somebody would have done long ago," he tells Hospital Infection Control. "But it isn't in the literature. To our great surprise, there were not only positives, but a substantial rate of positives."

Indeed, C. diff was detected in 28% of beef, 35% of the pork, and 10% of the turkey tested. Research and analysis is continuing, but the most disconcerting finding is that the C. diff strains detected included the highly toxigenic strain that has been linked to infections in hospitals and the community.

"We sampled ground or otherwise prepared meats — some ready to eat — from three local grocery stores three times at monthly intervals in 2006," he says. "We fount that overall 28% of the food samples contained fully toxigenic C. diff. Among the strains isolated were the ribotype 078 strains that are commonly found in food animals like calves. And the ribotype 027 strain, which is the epidemic strain."

The CDC confirmed the results, warning there was no still no clear connection between C. diff in food and human disease. Still, a disturbing corollary to the finding is that cooking meat may not necessarily kill the anaerobic spore-forming bacterium. "I don't have any data to suggest that these bugs would be killed by cooking," Songer says. "They are in the form of spores, which are highly resistant. In the laboratory, getting rid of spores requires autoclaving or long-term exposure to some fairly harsh disinfectants."

No proven connection to human disease

Another group of researchers in Canada also is finding C. diff in livestock animals, concluding in one paper that "cattle may be reservoirs of C. difficile for humans."² Still, investigators in both countries emphasize that the research is preliminary and the link to human disease — specifically the increasing reports of C. diff in the community — has not been established.

"It is very important to emphasize that," Songer says. "We know that C. diff causes disease in humans, we know it causes disease in food animals, [and now] we find it in foods. But transfer from the food animals to the foods is not proven. Nor is transfer from the foods to the humans proven. There is a lot yet to be done. There are many control points where C. diff could theoretically get into the food supply."

Songer is continuing the research, trying initially to confirm that the findings are not limited to meats sold in Tucson. The working hypothesis is that the findings will be replicated elsewhere, but C. diff in general is marked by a lack of surveillance and documentation because it is difficult to work up in the lab.

"I think that one of the key factors here is that for a variety of reasons labs don't culture, they don't isolate C. diff and characterize the isolates," Songer says. "They haven't for a long time. It's more expensive than doing a diagnosis by testing for the presence of toxin. Of course, difficile gets its name from being 'difficult' to grow, but it isn't impossible to grow from stool specimens. It just requires proper anaerobic conditions and selected media. It isn't very much different than isolating most bacteria. You just have to put them in the right environment with the right nutrients. It'll grow."

Indeed, in many cases under investigation the lab work has not been completed to confirm that the ribotype 27 strain is the cause of a given infection in the community. "We don't have that many isolates yet from these community-associated cases even with [the recent data from North Carolina] is all retrospective," McDonald says. "Folks there hadn't collected isolates yet and we are trying to get those."

However, the strain was detected in the community in a recent Canadian study, he notes.³ "A prominent player in their community was this epidemic strain," he says. "They didn't do careful definitions of community associated C. diff. They were just isolates from their community cases. They were diagnosed in the community, but some of them may have been recently hospitalized. Nevertheless, they found quite a few."

No lack of questions, theories

As more community infections are investigated both the primary etiologic agent and transmission factors should become clearer. An obvious question is whether the increase in community cases — as has been noted with CA-MRSA — is a direct result of increased antibiotic use in nonhospital settings. "That's one possibility," McDonald says. "We don't have a [comparative] baseline, but maybe it is a real increase [related] to more antibiotic use in the community — patient care being shifted to outpatient settings."

Indeed, prior administration of fluoroquinolones in particular seems to trigger the appearance of cases. Given that, McDonald and fellow researchers recently questioned "whether a large-scale restriction of the use of these antimicrobial agents will be necessary to slow the geographic spread of the strain. . . . If this epidemic strain continues to spread and to contribute to increased morbidity and mortality, it will be important either to reconsider the use of fluoroquinolones or to develop other innovative measures for controlling C. difficile-associated disease."⁴

However, such a restriction could prove difficult and potentially counterproductive because fluoroquinolones are used for the treatment of many common infections. "We know that certain strains are more resistant to antibiotics and that is why they are successful," McDonald says. "One issue is clindamycin, a drug we would like to be able to use for community MRSA. But clindamycin has a fairly deserved reputation of causing C. difficile. We are between a rock and hard place there."

Another unknown is whether the ribotype 027 strain has qualities that enhance its transmission, perhaps an ability to persist longer in the environment or to cause symptomatic infection at lower bacterial levels than typical strains. "There are some data that suggest that this organism is potentially more transmissible than the nontoxigenic strains," says Judith O'Donnell, MD, an infectious disease physician at Hahnemann University Hospital in Philadelphia. "It may be more efficient at being transmitted within the environment from person to person. That may be why there has been some community-associated disease because it is a better transmitter. It can be more easily and efficiently transmitted from person to person through fecal-oral contact."

A related factor is that the epidemic C. diff strain now in circulation appears to be more virulent, causing infections in the community that would have previously gone undetected.

"I think that's what is happening because this epidemic strain makes about 20 times more toxin," says Fred A. Zar, MD, professor of medicine at the University of Illinois in Chicago. "These mild cases in the community used to be there. You would give somebody an oral antibiotic for sinus infection and they would get a little diarrhea, but because they were otherwise healthy people and it was a short course of antibiotics, it would go away and it would never come to medical attention. But now you are getting [infections with] this epidemic strain, which is making so much more toxin they are coming to medical attention. I have no proof, but I suspect the incidence — the number of cases — of C. diff is probably the same. It is not like this bug is being found in more people's stool. It is because of much more severe disease that people are finding [cases]. They are coming to medical attention and being reported."

Another question is whether some of the cases are the result of C. diff colonization — whether from a prior hospital exposure, food or other source — that subsequently develops into symptomatic infection. "That's an interesting point because the first thing you would think was that even if it was coming in through the food supply, it would just be colonization," McDonald says. "In general though, the epidemiology of C. diff is different from MRSA or VRE. Those are organisms where we know that longstanding colonization increases your risk of getting disease at any particular time [later]. That is not true with C. diff. There is evidence that people who are long-term colonized are actually protected from getting C. diff during their hospital stays. It is probably because they are somewhat immunized to it. It appears that C. diff-associated disease is usually associated with the new acquisition of a strain."

Reference

1. Centers for Disease Control and Prevention. Severe Clostridium difficile-associated disease in populations previously at low risk — Four states, 2005. MMWR 2005; 54(47); 1,201-1,205.
2. Palacios A, Stämpfli HR, Duffield T, et al. Clostridium difficile PCR ribotypes in calves, Canada. Emerg Infect Dis [serial on the Internet] 2006 Nov; available from www.cdc.gov.
3. Duncan R. MacCannell DR, Louie TJ, et al. Molecular Analysis of Clostridium difficile PCR Ribotype 027 Isolates from Eastern and Western Canada. J Clin Microbiol 2006; 44:2,147-2,152.
4. McDonald LC, Killgore GE, Thompson A, et al. An epidemic, toxin gene-variant strain of Clostridium difficile. N Eng J Med2005; 353:2,433-2,441.