As Maryn McKenna and others have reported, a paper was released on Friday showing a high percentage of drug-resistant Staphylococcus aureus contaminating raw, retail-available meat products. There has been a lot of media coverage of this finding–so what does the study say, and what are its implications? More after the jump.
First, a bit about S. aureus itself, and why this study was carried out. Historically, Staph has been a relatively common cause of food poisoning. The bacterium produces toxins that can collect to a high level when prepared foods are left at room temperature, such as potato salad at picnics (a common vehicle for staph food poisoning). For much staph food poisoning, the ultimate source is generally thought to be a human–Aunt Bea who perhaps had S. aureus on her hands while cutting the potatoes, and then the bacterium grew and generated bacterial toxins while it was sitting out at the family gathering.
However, a number of recent studies have shown that S. aureus doesn’t always have to come from people. Studies in the Netherlands and here in the US have shown the presence of S. aureus (including methicillin-resistant strains, or MRSA) on raw meat samples. That makes the epidemiology of S. aureus food poisoning much more difficult. Did Aunt Bea really contaminate the potato salad via her own S. aureus (about a third of us carry the organism, usually in our noses), or did she use the same cutting board for the potatoes that she’d just used to shape the turkey burgers, thereby cross-contaminating the potatoes that way? To try and figure out which had occurred, you could take a swab from Aunt Bea’s nose and see if her staph matched what you isolated from the potato salad–but how can we be sure that Aunt Bea didn’t also rub her nose while she was working on the turkey burgers, and spread that staph to herself as well? As you can see, it can be potentially messy.
That also introduces a second issue with staph in meat–it’s not only the “classic” problem of toxin-mediated staph food poisoning, but we also have to consider the potential for food-borne S. aureus to be transferred from the meat to the food handler. If they’re not careful with their procedures, they could end up introducing staph into any wounds on their hands (thereby possibly causing an infection), or rubbing their nose/eye/skin and introducing the bacterium that way, potentially leading to long-term carriage. Really, in that manner it’s no different from picking up S. aureus at your local gym, or school, or hospital–you touch something that’s contaminated with the bacterium, and then unknowingly colonize yourself with the germ.
So–for those reasons, there are two main concerns with the finding of staph in meat: classic food poisoning, and self-inoculation with subsequent potential spread to family and community. How often does either really occur? We have no idea at this time, and that’s what makes studies like the current Price paper in Clinical Infectious Diseases critical.
For this study, they collected 136 meat samples (chicken, turkey, beef, and pork) from 5 large US cities: Flagstaff, Washington DC, Los Angeles, Chicago, and Fort Lauderdale. They tested all of the meat samples for S. aureus, and then did molecular typing on the isolates, and also determined what antibiotic resistances they demonstrated.
What did they find? A lot of staph, and a lot of drug resistance. From the paper, S. aureus contamination was most common among turkey samples (77%; 20/26) followed by pork (42%; 11/26), chicken (41%; 19/46), and beef (37%; 14/38). Almost all of the isolates were resistant to at least one antibiotic, and multidrug resistance (being resistant to at least 3 different classes of antibiotics) occurred among more than half of the isolates.
What about MRSA? Most of the stories don’t specifically mention it, but Figure 1 in the paper shows that 3 of the tested isolates were MRSA (oxacillin resistant), for a prevalence of 2.2%. One was from a turkey sample originating in Washington, DC; another from a pork sample from Fort Lauderdale; and a third from a beef sample in LA. All three were also multi-drug resistant, meaning that the options would be limited if someone ended up infected with those and needed antibiotic treatment.
The other important finding was that most of the strains were not common “human” types of staph, and that each type of meat had its own unique “fingerprint” of staph types. Oddly enough, what we usually think of as “pig staph” (ST398, see also here) was identified more commonly in turkey samples than in pork, where the most common type was ST1. In chicken meat, ST5 was most commonly identified (to be expected; see this post by Ed Yong covering a study about how people gave staph to chickens decades ago). In beef, a number of different types were found.
Not surprisingly, the meat industry has criticized the study and the press accounts covering it. They note–rightly–that it’s a small study, and there are a lot of unanswered questions as I describe above. A quote from the article (registration required):
“Authors of the new study, which involved a small number of samples from retail stores, claim that their findings suggest that a significant public health risk exists,” AMI [American Meat Institute] said in a statement.
“However, federal data from the Centers for Disease Control and Prevention (CDC) show steady declines in foodborne illnesses linked to consumption of meat and poultry overall and indicate that human infections with Staphylococcus aureus comprise less than one percent of total foodborne illnesses, ” AMI said.
When the U.S. Department of Agriculture studies the prevalence of bacteria, their work involves thousands of samples collected over long periods of time to ensure accuracy, AMI said.
The bacteria are destroyed through normal cooking procedures, AMI noted.
“While the study claims that the many of the bacteria found were antibiotic resistant, it does note that they are not heat resistant,” AMI said.
They are correct that foodborne illness overall has decreased, and that cooking likely kills the organisms, but they completely gloss over the fact that 1) not all foods that will be in contact with Staph contamination from the meat products will be cooked (e.g., vegetables or anything that may be cross-contaminated during prep procedures), and 2) that there is the potential colonization aspect that I raised above. So some media outlets have oversold the risk, acting as if we have good evidence that Staph from meat is killing people (right now, we don’t, though there are many cases of human infections with “animal” staph in people who lack contact with livestock). Others, such as AMI, are minimizing the risk by only focusing on one line of exposure (ingestion of cooked food that was contaminated with staph).
The bottom line is that we simply don’t have the studies yet to tell us if Staph on food is riskier than Staph at your gym, or Staph at your school, or Staph directly from a pig, etc. However, there’s nothing good or beneficial about having yet another source of S. aureus that’s multi-drug resistant, and as such, it’s yet another reminder to do what you can regarding safe food handling practices.
Waters, A., Contente-Cuomo, T., Buchhagen, J., Liu, C., Watson, L., Pearce, K., Foster, J., Bowers, J., Driebe, E., Engelthaler, D., Keim, P., & Price, L. (2011). Multidrug-Resistant Staphylococcus aureus in US Meat and Poultry Clinical Infectious Diseases DOI: 10.1093/cid/cir181