Update: The French case — not MRSA but so interesting

I'm flattered to have as a regular reader Dr. Peter Davies, a professor of swine health and production in the University of Minnesota's Department of Veterinary Population Medicine. (Disclosure: I worked part-time at U Minn from mid-2006 to mid-2010, but in a different school.) In a comment on my previous post, he points out — perils of reading on a smartphone — an important point where I erred: The staph strain involved in the death of the French 14-year-old was not MRSA, but MSSA, drug-sensitive staph, that had picked up a resistance factor.

Unpacking that a bit: At a minimum, MRSA is resistant to all beta-lactam antibiotics — penicillin, the semi-synthetic penicillins (including methicillin, what the M in MRSA stands for), several generations of cephalosporins, monobactams, and carbapenems. It is also separately, but variably, resistant to macrolides (such as erythromycin), lincosamides (clindamycin), aminoglycosides (gentamicin), fluoroquinolones (ciprofloxacin) and tetracycline.

Livestock-associated MRSA, known as ST398 for its performance on a particular test (multi-locus sequence typing) was first identified as having a tie to pig-farming because it was also resistant to tetracycline, which was being given to the pigs on the farms where the first human carriers worked. (Hence its jocular name, "pig MRSA," though it's since been found in other animals.)

The ST398 strain involved in the French girl's death does not have that broad array of resistance. Chiefly, it was not resistant to beta-lactams, and so can't be considered MRSA. On analysis, it was resistant to the macrolides, of which the best-known are erythromycin and azithromycin (Zithromax or Z-Pak). Here's something else intriguing: On another test (spa typing), the ST398 strain in the French girl was one known as t571; the ST398 that has spread from pigs to humans in the European Union, and subsequently to Canada and the United States, is usually t034.

Here's why this is all so interesting: MSSA ST398 t571 was reported just a few years ago in New York City, in a Bronx community that has close ties to the Dominican Republic, and also in the towns in the Dominican Republic where those Bronx residents come from and visit. (Here's my initial post on that finding from a medical meeting, and subsequent post when the paper was published.) In that case, the ST398 was fully drug-sensitive — and there was no visible link to pigs, though the authors speculated that livestock, perhaps poultry, might be playing a role on either side of the "air bridge" connecting the two communities.

In the paper (Bhat, Dumortier, Taylor et al., EID 2009, DOI: 10.3201/eid1502.080609), the authors expressed concern that, given staph's promiscuous ability to acquire resistance — and the fact that ST398 is not regularly surveilled for —  the ST398 in New York could become an undetected resistant strain:

Given ST398’s history of rapid dissemination in the Netherlands, its potential for the acquisition of methicillin resistance, and its ability to cause infections in both community and hospital settings, monitoring the prevalence of this strain in northern Manhattan and the Dominican Republic will be important to understand more about its virulence and its ability to spread in these communities.

 And now it appears it has become resistant — but in France, not New York City or the Dominican Republic, and to macrolides, not  beta-lactams. It's one more reminder of staph's genius at acquiring genetic defenses, and of how our lack of attention to its mutability and spread continues to allow it to take us by surprise.

One surgical infection with MRSA: $61,000

From a multi-state, public-private research team — Duke University, Wayne State University, and the Durham, NC VA — comes a precise and alarming calculation of MRSA's costs in hospitals: For one post-surgery infection, $61,681.

The group compared the course, costs and final outcome of three matched groups of patients from one tertiary-care center and six community hospitals in one infection-control network run by Duke. The three groups were: patients with a MRSA surgical-site infection; patients with a surgical-site infection (SSI) due to MSSA, drug-sensitive staph; and surgery patients who did not experience infections, matched to the other two groups by hospital, type of procedure, and year when the procedure took place. (This same cohort has been described in an earlier prospective study that looked at risks for MRSA SSIs.) Altogether, there were 150 patients with MRSA SSIs, 128 with MSSA SSIs, and 231 uninfected surgery patients to serve as controls.

Here's what they found. Patients with post-surgical MRSA infections:

• stayed in the hospital 23 days longer
• incurred an average extra cost of $61,681
• were more likely to be readmitted to the hospital within 90 days
• were more likely to die before 90 days had passed.

The authors write:

Our study represents the largest study to date of outcomes due to SSI due to MRSA. Our findings confirm that SSIs due to MRSA lead to significant patient suffering and provide quantitative estimates of the staggering costs of these infections. SSI due to MRSA led to a 7-fold increased risk of death, a 35-fold increased risk of hospital readmission, more than 3 weeks of additional hospitalization, and more than $60,000 of additional charges compared to uninfected controls.

For just the patients in this study, the excess costs (across 7 hospitals) totalled $19 million.

This is a highly useful study on several axes. First, remarkably, there has not been agreement over whether and how much of a problem MRSA poses in post-surgical settings, particularly when compared to drug-sensitive staph. This study provides careful, thoughtful, well-documented proof that combating MRSA infection is worthwhile. (NB, MRSA infections did not increase the risk of death relative to MSSA infections, which should remind us both of the often-forgotten virulence of MSSA, and also that MRSA's perils can lie in extended illness and disability as much or more as in early death.) Second, by putting a very specific number on the cost of a post-surgical MRSA infection, it gives healthcare administrators a benchmark against which they can judge the cost of a prevention program. We've all heard complaints that prevention programs can be costly and their benefit is hard to measure in a bottom-line way. With this very specific number, that complaint should no longer be valid.

There's a final point that is implied in the paper but not called out, so let me call it out on the authors' behalf. These results are very likely an under-estimate of MRSA's costs. That's because, first, the specific procedures the patients underwent were cardiothoracic and orthopedic; those are not the surgical procedures most likely to be followed by a MRSA infection. And second, data collection for this study ceased in 2003, about a year after the first emergence of USA300 and several years before that very successful community strain began its current move into hospitals. However much MRSA was extant in 2003, there is more now.

The cite is: Anderson DJ, Kaye KS, Chen LF, Schmader KE, Choi Y, et al. 2009 Clinical and Financial Outcomes Due to Methicillin Resistant Staphylococcus aureus Surgical Site Infection: A Multi-Center Matched Outcomes Study. PLoS ONE 4(12): e8305. doi:10.1371/journal.pone.0008305

More MRSA in pigs, in Portugal

A brand-new report, in a letter to the International Journal of Antimicrobial Agents, indicates that ST398 "pig MRSA" has been found in Portugal for the first time.

Constanca Pomba and colleagues from the Technical University of Lisbon swabbed and cultured the noses of pigs and veterinarians on two pig farms in different regions of Portugal, and also checked the air at both farms.

What they found:

• On Farm A: All pigs and the veterinarian positive for ST398, the pig-origin strain that has been found so far in Iowa, Ontario, the Netherlands, France, Denmark, Germany and Austria and has, depending on the country, caused human disease and/been found on retail meat. The veterinarian was transiently colonized, which is to say that he was not carrying the bug long-term.
• On Farm B: All pigs — but neither of two veterinarians — positive for a different MRSA strain, CC (or ST) 30. This is very interesting, because CC30 is usually a drug-sensitive strain (MSSA, methicillin-sensitive S. aureus), and has been found in pigs primarily in Denmark and France. In Portugal, it is a human MSSA hospital-infection strain.

Strains from both farms were resistant to tetracycline; this is turning out to be a great marker for these strains having emerged due to antibiotic pressure in animals, because tetracycline is very commonly used in pigs. but not much used for MRSA in humans. The strains have the genes tetK and tetM, so they are resistant not just to tetracycline itself, but to the whole class of tetracyclines including doxycycline and minocycline. The Farm B strains also carried the gene ermC, which encodes resistance to erythromycin.

So what does this tell us?

• First, that (once again), every time people look for ST398, they find it; it is now a very widely distributed colonizing bug in pigs, and is repeatedly spreading to humans. What we don't know, because all these studies are so new, is whether ST398 is actively expanding its range, or has been present in all these countries for a while. We have been anticipating its presence or spread (take your pick at this point) through the European Union because of open cross-border movement of food animals, meat, and agriculture and health care workers.
• And second, it should tell us that it is really past time to start looking for this more systematically. Every finding of ST398 that we have (long archive of posts here) is due to an academic research team who decided to look for the bug. None of the findings, to date, have come from any national surveillance system. (NB: Except for the first human colonizations in the Netherlands, which were found as a result of the national "search and destroy" rules in hospitals.)

Of note, the European Union is running a study now that is supposed to report ST398 prevalence at any moment (as they have been saying since 2007). It is not expected to be comprehensive, since it was piggy-backed onto another study, but it is something. The US government has not been so enterprising.

The cite is: Pomba, C. et al. First description of meticillin-resistant Staphylococcus aureus (MRSA) CC30 and CC398 from swine in Portugal. Intl J Antimicrob Agents (2009), doi: 10.1016/j.ijantimicag.2009.02.019

Child deaths from flu + MRSA, again

Folks, I am close to manuscript deadline and so keep disappearing down the rabbit hole; forgive me if I don't post as regularly as usual, I'll be back as soon as I can.

I wanted to point out the announcement by the Centers for Disease Control late Friday that we are starting to see children dying from MRSA this flu season. (The architecture of the linked page is unfortunately way clumsy; at the link, scroll down to the subhead "Influenza-Associated Pediatric Mortality.")

Since September 28, 2008, CDC has received nine reports of influenza-associated pediatric deaths that occurred during the current season.
Bacterial coinfections were confirmed in six (66.7%) of the nine children; Staphylococcus aureus was identified in four (66.7%) of the six children. Two of the S. aureus isolates were sensitive to methicillin and two were methicillin resistant. All six children with bacterial coinfections were five years of age or older.

We've talked before (here, here and here, among other posts) among the emerging understanding of the particular danger that MRSA poses during flu season, when (it is hypothesized) inflammation from flu infection makes the lungs more vulnerable to secondary bacterial infection.

(For those paying attention to the hospital v. community MRSA debate, this is a community-associated infection, not a hospital one.)

This current CDC bulletin underlines, just in case we have forgotten, that drug-sensitive S. aureus (MSSA) can be a serious foe as well. Let's remember, resistance makes MRSA less treatable than MSSA, but it does not change its virulence; MSSA by itself can be a very serious foe. Yes, there are other changes in some strains, especially the community ones, that do appear to increase virulence, but the original MSSA strain is nothing to trifle with.

Also, here's an important addition to this unfolding story: My colleagues at the Center for Infectious Disease Research and Policy are keeping track of kid deaths around the country. According to them, these CDC numbers are already out of date; they have uncovered more that the CDC has not yet posted, but may take note of in future weekly updates.

"Pig MRSA" in New York City - via the Dominican Republic?

Folks: Back in October, I broke the news for you of an intriguing poster presentation at the ICAAC meeting. It revealed the discovery of ST 398, the anomalous staph strain found in pigs, pig farmers and health care workers in Europe, in residents of a Dominican-immigrant neighborhood in northern Manhattan, and also in the Dominican Republic.

Because there is so much traffic back and forth between those neighborhoods, the authors theorized that people are providing an "air bridge" for the bacterium — though they were unable to say whether the bug is moving from the Dominican Republic to the United States, or vice versa.

I was unable to link to that presentation at the time, because it was a meeting poster - yes, literally a poster, the authors stand by it to discuss it with anyone who wanders by. However, now it has been published as a paper, in the CDC journal Emerging Infectious Diseases; and because it is a CDC journal, the full text is available free online here.

Just to underline, despite my headline above, the strain found in NYC was not MRSA: It actually is MSSA, drug-sensitive staph. The ST 398 found in Europe, Canada and the American Midwest is MRSA. The authors hypothesize that the NYC strain is at risk of becoming MRSA also.

To see the multiple posts in this blog about MRSA ST 398 and other strains in the food chain, food animals, and pets, go to the labels under the time stamp on this post, and click "animals" or "food."

The cite for the paper is: Bhat M, Dumortier C, Taylor B, Miller M, Vasquez G, Yunen J, et al. Staphylococcus aureus ST398, New York City and Dominican Republic. Emerg Infect Dis. 2009 Feb; [Epub ahead of print]

MRSA in meat in Louisiana: pig meat, human strain

On Nov. 3, I posted on an enterprising group of TV stations in the Pacific Northwest who had retail meat in four states tested for MRSA. I said at the time that it was the first finding of MRSA in meat in the US that I knew of.

Turns out that I was wrong by three days. On Oct. 31, the journal Applied and Environmental Microbiology published an electronic version of a study that they will be printing in the paper journal on some future date. Journals do this when a finding is so important or timely that it should see the light immediately, rather than wait through the additional weeks or months of print production.

And this finding is certainly timely. Shuaihua Pu, Feifei Han, and Beilei Ge of the Louisiana State University Agricultural Center have made what appears to be the first scientifically valid identification of MRSA in retail meat in the United States. But — and this is an important point — it is not the swine strain, ST 398, that has been found in meat in Canada and Europe, and in hospital patients in Scotland and the Netherlands, and in pigs in Iowa; and in humans in New York, though that strain was drug-sensitive.

Instead, what the researchers found (in 5 pork and 1 beef samples, out of 120 bought in 30 grocery stores in Baton Rouge, La. over 6 weeks in February-March 2008) was USA300, the dominant community MRSA strain, and USA100, the main hospital-infection strain. In other words, they found meat that had been contaminated during production by an infected or colonized human, not by a pig. As they say:

...the presence of MRSA in meats may pose a potential threat of infection to individuals who handle the food. ... (G)reat attention needs to be taken to prevent the introduction of MRSA from human carriers onto the meats they handle and thereby spreading the pathogen.

As we've discussed before, the primary danger from MRSA in meat is not that people will take the bug in by mouth (though that is a danger, since S. aureus because of its toxin production can cause severe foodborne illness — and these researchers found, overall, an S. aureus contamination rate of 46% of their pork samples and 20% of their beef samples). Rather, the danger is that people handling the raw meat will be careless in preparing it, and will colonize themselves by touching the meat and then touching their own noses or mucous membranes, leading to a possible future infection. As reader Rhoda pointed out in a comment last week, people could also infect themselves directly, by getting MRSA-laden juice or blood into an abrasion or cut.

So: Be careful in the kitchen, keep meat separate from other foods, wash cutting boards and knives, and (say it with me, now) wash your hands, wash your hands, wash your hands.

The cite for the new paper: Pu, S. et al. Isolation and Characterization of Methicillin-Resistant Staphylococcus aureus from Louisiana Retail Meats. Appl. Environ. Microbiol. doi:10.1128/AEM.01110-08. Epub ahead of print 31 Oct 08.

Housekeeping note: This is the 16th post I've written on MRSA in food animals and/or meat. Providing all the links to the previous posts is starting to obstruct the new news. So if you are looking for all those past posts, go to the labels at the end of this post, below the time-stamp, and click on "food." You should get something that looks like this.

TV stations find MRSA in retail pork in Pacific Northwest

In the comments, Coilin Nunan of the UK's Soil Association (which published the wonderful 2007 report MRSA in Farm Animals and Meat report) calls attention to a report that I also spotted over the weekend.

A network of TV stations in Washington, Idaho, Oregon and California did a joint report in which they bought 97 packages of ground pork or pork cutlets and sent them to a laboratory for testing. The lab found that three of the packages, all ground pork, contained MRSA.

I believe this is the first time anyone has found (or, perhaps, looked for) MRSA in retail pork in the US. You'll remember that MRSA ST 398 has been found in meat in Canada and Europe, and in hospital patients in Scotland and the Netherlands, and in pigs in Iowa; and MSSA ST 398 in humans in New York City.

There are some important unanswered questions about this report:

• We aren't told the strain. If it's ST 398, that would be information on the spread of ST 398 in the US. If it's USA300, on the other hand, it could be contamination from an infected or colonized human, perhaps someone in the preparation chain.
• We aren't told the provenance of the pork. Was it bought from a variety of markets, or one chain of supermarkets that might have one regional supplier? Was it organic v. conventional? Small-farm versus feedlot?
• We can't draw any broad conclusions from this. I am a poor biostatistician, but to me, this is purely a convenience sample. (If anyone disagrees with me, please weigh in.) In other words, it's one data point. It says: There is MRSA in these packages of pork — which is an important piece of information — but it doesn't say: 3% of all US pork contains MRSA.

Also, while the written version of the report that I linked above isn't bad, overall, it contains one significant error. It says:

This drug-resistant bacteria is already responsible for more deaths in the US than AIDS. What makes MRSA so potentially dangerous is the bacteria can cause sickness just by touching it.

Well, not exactly. The concern with MRSA in meat is that, if you handle it without strict cleanliness, you might become colonized with the bacteria. That is not at all the same as developing a MRSA infection, much less the invasive MRSA the first sentence of that quote refers to. And yes, colonization can lead to infection. But to say that touching MRSA-contaminated meat will inevitably cause an invasive MRSA infection is alarmist.

I'm assuming the stations undertook this because it is sweeps month. (For those who have so far been spared the internals of TV news, "sweeps" are months — usually February, May, July and November — when stations' audiences are measured to determine market rank and advertising rates. Because it is in the stations' interest to attract as much audience as possible during those months, sweeps is usually when news stations run big investigative projects.) Interesting that they chose this topic. I think we can take this as an indicator — again, just one data point, but an interesting one — of emerging US concern over MRSA in meat.

Microbes in US meat, but no MRSA

The ICAAC-IDSA meeting has ended, but there are still many abstracts that I have not been through. While I pore over them, though, an interesting paper has just been published that somewhat contradicts earlier research on the presence of MRSA in meat. (Earlier posts are here, here, here, here, here, here, here and here.)

The researchers, from the Warren Alpert Medical School of Brown University and Rhode Island Hospital, bought ground beef, boneless chicken breasts and pork chops from 10 stores in and around Providence. Two stores offered both conventional and "natural" choices, so they bought both, giving them 36 (=[10+2]x3) samples all told. They cultured for MRSA, vancomycin-resistant Enterococcus, extended-spectrum beta-lactamase producing Gram-negative bacteria and E. coli 0157:H7.

And they found... almost nothing. Only one samples grew a resistant microbe, the ESBL Gram-negative Serratia fonticola. A secnd level of testing, however, uncovered four samples carrying S. aureus — but all methicillin-sensitive, not MRSA.

So are we in the clear? Not necessarily. It is, as they say themselves, as small study, in which only a third of the samples were pork, though pigs are the animals most associated with MRSA via the strain ST398. And the presence of S. fonticola is troubling, because it not only causes disease directly (in animals and in humans), but also harbors a plasmid that can transfer resistance to other bacterial strains.

Nevertheless, it is a comforting reminder that, though MRSA has been found in meat, it has not been found everywhere. (Or at least, not in Providence.) Still, we shouldn't let our personal vigilance lapse. The hypothetical danger from MRSA in meat is not that we'll swallow it, but rather that we'll be colonized if we handle the raw meat without being careful enough about kitchen hygiene. So keep raw meat away from other food, wash your cutting boards and counters, and (say it with me, now), wash your hands, wash your hands, wash your hands.

The cite is: Philip A. Chan, Sarah E. Wakeman, Adele Angelone and Leonard A. Mermel, Investigation of multi-drug resistant microbes in retail meats. Journal of Food, Agriculture & Environment, Vol.6 (3&4), July-October 2008.

ST 398 in New York City - via the Dominican Republic?

Here's a piece of MRSA news from the ICAAC meeting (see the post just below) that is intriguing enough to deserve its own post.

US and Caribbean researchers have found preliminary evidence of the staph strain ST 398, the animal-origin strain that has caused human illness in the Netherlands and has recently been found in Ontario and Iowa, in Manhattan. How it may have arrived: Via the Dominican Republic.

Th researchers (from Columbia University and Montefiore Medical Center in New York, three institutions in the Dominican Republic and one in Martinique) examine the influence of an "air bridge" — very frequent household travel — that is bringing MRSA and methicillin-sensitive staph back and forth between the Dominican Republic and the immigrant Dominican community at the north end of Manhattan. They compared 81 staph isolates from Dominican Republic residents and 636 from Manhattan residents and, among other findings, say that 6 Dominican strains and 13 Manhattan strains were ST398.

It is the first time ST398 has been found in Manhattan or in the Dominican Republic. (Most likely also the first time anyone has looked.)

The authors observe with some understatement:

Given the history of ST398's rapid dissemination in the Netherlands, its history of methicillin-resistance and its ability to cause infections in both hospital and community, it will be important to monitor its prevalence in these new regions.

It is important to note that these ST398s were not MRSA — they were MSSA, methicillin-sensitive. However: Earlier this year, the Dutch researchers who have delineated the emergence of ST398 in Holland commented on the diversity of ST398 they have found on different pig farms and hypothesized that the resistance element has been acquired several different times by methicillin-sensitive staph. (van Duijkeren, E. et al. Vet Microbiol 2008 Jan 25; 126(4): 383-9.)

So it is possible to hypothesize that this strain arrived in Manhattan from the more rural Dominican Republic, though with the growth of hobby urban farming in NYC, one could also make the case that transmission went the other way. And it is also possible — I emphasize possible — that this could be a precursor to ST398 MRSA emerging in Manhattan. An interesting thought.

(This research is not online, because it is a poster presented at a medical meeting. For reference, the cite is: C. DuMortier, B. Taylor, J. E. Sanchez et al. "Evidence of S. aureus Transmission Between the USA and the Dominican Republic." Poster C2-224. 48th ICAAC-46th IDSA, Washington DC, 24-28 Oct 2008.)

Child deaths from flu + MRSA

Steve Smith of the Boston Globe (who is really good, and I say that as someone who used to compete against him) has a story up regarding state and national concern over children's deaths from MRSA pneumonia. There have been two such deaths in Massachusetts this year. These are the sort of deaths that make headlines, as they did last October with the death of 17-year-old Ashton Bonds in Virginia: The pneumonia is very fast-moving and very destructive of lung tissue, and young children have died from it in less than 24 hours.

The CDC is concerned about this: Twice in two years (last May, blogged here, and last January), the agency pushed out an advisory to state health departments, asking them to report any children's deaths in which flu played a role. Surveillance for pediatric flu deaths is a relatively new thing for the CDC — the agency set up a system after the bad early flu season of 2003-04, in which more than 150 children died — so there is relatively little history to draw on. But MRSA has played a role in child deaths in each of the past three years, according to that January bulletin:

From October 1, 2006 through September 30, 2007, 73 deaths from influenza in children were reported to CDC from 39 state health departments and two city health departments. Data on the presence (or absence) of bacterial co-infections were recorded for 69 of these cases; 30 (44%) had a bacterial co-infection, and 22 (73%) of these 30 were infected with Staphylococcus aureus.

The number of pediatric influenza-associated deaths reported during 2006-07 was moderately higher than the number reported during the two previous surveillance years; the number of these deaths in which pneumonia or bacteremia due to S. aureus was noted represents a five-fold increase. Only one S. aureus co-infection among 47 influenza deaths was identified in 2004-2005, and 3 co-infections among 46 deaths were identified in 2005-2006. Of the 22 influenza deaths reported with S. aureus in 2006-2007, 15 children had infections with methicillin-resistant S. aureus (MRSA).

Among MRSA researchers, concern over these necrotizing pneumonia cases has been growing for a few years. Some surveillance suggests that such cases may be increasing, though that could be an issue of, "once you start looking for something, you find it." And the cases are undeniably severe: 56% of children with MRSA pneumonia die, according to a 2007 paper.

But at leaast some of those deaths may be avoisable — or would be if doctors in the community were more attuned to the possibility of MRSA. In a poster at last autumn's ICAAC meeting (first author AJ Kallen) CDC researchers reported thay reviewed charts of all the children admitted with a stah infection at Atlanta's three children's hospitals during the 2006-07 flu season. There were 53 cases of Staph aureus pneumonia; 22 of the children saw a physician an average of 3.5 days before being admitted to the hospital, and THREE of them got drugs that would work against staph.

And yes, you read that right: Active case-finding in Atlanta in 2006-07 found 53 cases of flu-related staph pneumonia; 22 of them, according to the paper, were MRSA. But from the entire country during 2006-07, according to the advisory quoted above, the CDC received reports of 22 flu/staph pneumonias, 15 of them MRSA. Which suggests that flu/MRSA pneumonias in children are more common than current surveillance reveals.

A staph vaccine: How much would it help?

One more post on research from the meeting of the Society for Healthcare Epidemiology of America: Many MRSA researchers believe that the only way to truly control the pathogen — especially out in the community — will be through a vaccine.

Lay aside for the moment how problematic introducing a new vaccine can be these days, since the cost issues, along with shifts in the public's willingness to accept new vaccines, are ferocious hurdles. And lay aside also the difficulties that pharma companies have already faced in attempting to develop a staph vaccine.

But if such a vaccine were achieved, how many people could it help? Researchers from the Centers for Disease Control and Prevention attempted to answer that question in research presented at SHEA.

Background assumptions, part 1: The number of invasive MRSA infections now tops an estimated 105,000/year (a recalculation of the 94,000/year estimate from last October); more than 40% of invasive infections occur in those over 65; more than 50% are associated with a recent hospitalization; and 15% of MRSA infections recur at least once. And background assumptions part 2: A vaccine would have an efficacy rate of 40-75%, and an acceptance rate similar to flu-vaccine uptake: 20-50% among those 15-44, 35-70% among those 45-64, and 50-70% among those 65 and older.

Given those assumptions, Cynthia Lucero, MD and colleague predicted:

• If given only to those 65 and older, a vaccine would prevent from 12,720 to 32,270 invasive MRSA infections;
• If given to those over 65 and also those 15 and older who have already had an invasive infection, a vaccine would prevent 14,130 to 38,310 invasive MRSA infections;
• And if given to those over 65 and also anyone over 15 who is being discharged from a hospital, a vaccine would prevent from 17,240 to 49,940 invasive MRSA infections.

The best bang for the buck, the agency said, would be the middle strategy: It would prevent from 660 to 1,170 cases for every million doses of vaccine used. And since the vaccine would also cover methicillin-sensitive staph, it would likely prevent an equal number of serious MSSA cases as well.

The CDC is not by law allowed to lobby — or even, for the most part, allowed to offer a professional opinion unless Congress has asked it to do so. So these numbers are purely a thought experiment. But they're also a strong argument for the broad usefulness of a staph vaccine if one could be achieved.

CA-MRSA — not just a US problem

One of the mysteries of MRSA research has been the apparent difference in MRSA prevalence in different countries. Hospital-associated MRSA is a well-understood foe in Europe, but identification of CA-MRSA is infrequent. So is CA-MRSA actually less common in other countries, or is that an artifact produced by less surveillance there or more awareness here?

A new journal sheds some light on the question, and also sounds a warning. A team from University College-London's Centre for Infectious Disease Epidemiology analyzed data that are easily available in the UK because the country has a national healthcare system: "hospital episode statistics" (hospital admissions by primary diagnostic code) from 1989 to 2004. The team chose the codes that would indicate admissions for a variety of community-onset staph-related diseases: septicemia, pneumonia, bone and joint infections, and an array of skin and soft-tissue infections.

What they found: Over those 15 years, admits for staph septicemia, pneumonia and scalded-skin syndrome rose 5-fold; abscesses and cellulitis, 3-fold, and bone and joint infections, 1.5-fold. These were much larger increases than in the only national surveillance system which collects voluntary reports just of Staph bacteremia and showed a 2.5-fold increase from 1990 to 2004.

Because the study is based on ICD-9/ICD-10 codes, it contains no microbiological information. However, the authors point out that the invasive diseases captured by the codes are associated with PVL toxin, which in turn is associated with CA-MRSA more than HA-MRSA or MSSA. They say:

We identified a previously undescribed but major increase in pathogenic community-onset staphylococcal disease over the past 15 years. These trends are of concern given the international emergence of invasive community-onset staphylococcal infections.

The paper has been published ahead of print by Emerging Infectious Diseases. The cite is: Hayward A. et al. Increasing hospitalizations and general practice prescriptions for community-onset staphylococcal disease, England. Emerg Infect Dis. 2008 May; [Epub ahead of print]

Random MRSA research, ICEID

Came back from the conference and got slammed with work. (Also, snow, three days in succession. What is it about "Spring" that Minnesota does not understand?) So especial thanks to the 15 readers — you know who you are and Google Analytics does too — who have been diligently checking regardless.

I'm going to do a quick wrap on some of the remaining MRSA papers presented at the Emerging Infections conference. (For a wrap-up of flu and foodborne-disease research, see my final conference story at CIDRAP.)

A quick explainer for those who don't make a habit of going to science conferences, you lucky souls you: For many researchers, this is the first presentation of new or incremental findings. Thus, there's no publication to link to — that's why these ICEID posts aren't so content-rich. Many of these papers may end up in a medical journal in the next year, but for now, not even the abstracts are online.

So:

• To generate hypotheses about what leads to CA-MRSA infection, a team from the Minnesota Department of Health and the CDC analyzed the life circumstances of 150 people diagnosed with CA-MRSA or MSSA, and found the strongest correlation was between CA-MRSA and a prior history of boils or prior use of antibiotics. (Lead author: K. Como-Sabetti.)
• In another Minnesota-based report, researchers from the VA Medical Center found that patients who developed MRSA in the hospital had a risk of dying within 6 months that was three times higher than for patients with non-resistant staph. (Lead author: C. Lexau.)
• And, OK, Minnesota trifecta: A team from the MN DoH and the CDC checked the colonization levels of MRSA patients and their household members over a year and found that, even a year after the first diagnosis. one out of every five patients' households had at least one household member who was still colonized — and that use of Bactroban had made no difference. (Lead author: J. Buck.)
• Confirming the hypothesis that MRSA can persist on surfaces and contribute to colonization, Texas researchers swabbed bathroom and common-room surfaces at a university and a jail and found the presence of MRSA was 5 times higher on the jail surfaces (6.1% of samples v. 1.2%). Jails are already known to be hotbeds of CA-MRSA — some of the earliest recognized outbreaks were recorded in jails — and this suggests that in a setting where there is a large amount of MRSA, environmental contamination may keep the bug circulating. (Lead author M. Felkner.)
• An analysis by the Connecticut Department of Public Health of lab-confirmed diagnoses of invasive MRSA between 2001 and 2006 confirms how tricky sorting out HA- and CA-MRSA can be. The incidence of invasive MRSA stayed stable over those five years, but the proportion of community-associated MRSA rose while the proportion of "hospital-onset" (developed no sooner than 48 hours after admission to the hospital) decreased. That makes it sound as though CA-MRSA is increasing overall. But: When the Connecticut state laboratory finegrprinted the strains, they found that 2% of the hospital-onset and 4% of the "hospital-acquired/community-onset" cases were actually caused by community strains, and 76% of the community cases were actually caused by hospital strains. (Lead author: S. Petit.)

That's all on ICEID; tune in two years from now when the conference returns, or stay tuned for more on hospital-acquired infections, mandatory reporting, and disagreements over "search and destroy."

MRSA and animals - not pets, meat

The International Conference on Emerging Infectious Diseases ended today with more news about a range of infections including MRSA. Most eyebrow-raising: Dr. J. Scott Weese of the Ontario Veterinary College in Guelph, an author of the original MRSA-in-cats paper I described a few posts ago.

Weese's news, tucked in at the end of a comprehensive presentation on MRSA and animals: Analysis of 212 raw pork products sold in four Canadian provinces reveals an average rate of MRSA contamination of more than 9%.

This shouldn't be surprising: Research by Weese and others has been revealing a complex and not well-understood interplay of infection between pigs and nearby humans.

• A 2005 French paper reported a rate of resistant-staph nasal colonization in pig farmers that was twice as high as among human controls; 57% of the isolates from the farmers were identical to nasal isolates in pigs.
• Last June, a Dutch study found 39% of pigs in nine major slaughterhouses in the Netherlands carried an identical novel MRSA strain.
• In December, two studies filled out the picture. A Dutch study reported the prevalence of that novel MRSA strain (dubbed ST 398 and first found in a human in 2003) has risen to more than 21% of all MRSA isolates in the country. A Canadian study (with Weese as senior author), published online ahead-of-print, found MRSA colonization rates of 25% among Ontario pigs and 20% among pig farmers, with most of them sharing the ST 398 strain but some possessing the CA-MRSA strain USA100.
• A Dutch study in January (first author Engeline van Duijkeren, who did the S. intermedius study from a few posts ago) found that pigs were colonized with MRSA on a variety of types of farms, such as ones that birth pigs and ones that raise them to slaughtering weight.
• And just to erase any doubts, several Dutch studies have established that the ST 398 strain causes human disease: endocarditis, mastitis, severe hospital-acquired pneumonia and bloodstream infection.

Weese stressed today that his team's finding of MRSA in raw retail meat is very preliminary (though it is backed by a Dutch study from last November that found 34 MSSA strains and two MRSA strains — ST 398 and USA300 — in 79 pork samples). He added, though:

"People don’t tend to handle pork like it is biohazardous, unlike chicken. So there may be a theoretical concern that pork could be a vehicle of methicillin resistance colonization — but it is way too early to say anything about that."

But if you're not already handling raw meat in a careful manner (sterilizing cutting boards, avoiding cross-contamination), it might not be too early to start.

7-fold increase CA-MRSA in parts of Chicago

Via the Archives of Internal Medicine, a new study from Cook County Hospital and Rush University Medical Center in Chicago. Betwen 2000 and 2005, the incidence of CA-MRSA at the hospital and its neighborhood clinics increased 6.84 times. Notably, meth-sensitive staph (MSSA) did not decrease - this was a true addition, not a substitution of one strain for another. And 79% of the strains identified were USA300. Important clues to the rapid spread of the bug: Those infected were more likely to have been incarcerated more than once (carrying the bug from the known epicenters of jails and prisons back out into the community) or to live in public housing (possibly because of overcrowding as Chicago demolishes its old-style projects and moves the people who live there into its remaining public housing). The paper raises but can't answer the question of a synergy between those factors: People coming out of jail may be returning to households in public housing creating a bridge between a known epicenter and a population whose living conditions put them at greater risk.

Find the paper here.

MRSA pneumonia deaths in children - on the rise?

On Wednesday afternoon, the Centers for Disease Control and Prevention pushed out an official Health Advisory (an electronic communication to state and local health departments) asking them to report any children in their area who have died from MRSA pneumonia. Deaths seem to be on the rise. Money quote:

Only one influenza and S. aureus co-infection was identified in 2004-2005, and 3 were identified in 2005-2006. Of the 16 children reported with S. aureus so far in 2006-2007, 11children had methicillin-resistant (MRSA) isolated from a sterile site (9) or sputum (2), and 5 had methicillin-susceptible S.aureus isolated from a sterile site (3) or sputum (2). ... Children with influenza and S. aureus co-infections were reported to be in good health before illness onset but progressed rapidly to severe illness. Influenza strains isolated from these children have not been different from common strains circulating in the community and the MRSA strains have been typical of those associated with MRSA skin infection outbreaks in the United States.

The possibility always exists that surveillance has changed - in other words, that more cases are being found because doctors are authorities are looking harder for them. Still, the rise is troubling.

Here's the full text: Influenza-Associated Pediatric Mortality and the Increase of Staphylococcus aureus co-infection