The MRSA Challenge
Which is Better: Targeted or Universal Surveillance

By Genna Rollins

After years of discouraging news about both the prevalence and virulence of methicillin-resistant Staphylococcus aureus (MRSA), recently there has been cause for optimism. Of note, the Centers for Disease Control and Prevention (CDC) reported in 2010 across-the-board decreases in six categories of MRSA infections in nine metropolitan areas that are part of the agency’s long-term, population-based surveillance program. Hospital-onset invasive infections declined by 28% during the 3-year study period, while MRSA bloodstream infections dropped by one-third (JAMA 2010;304:641–48). These findings complement data from the National Healthcare Safety Network showing an almost 50% decline in MRSA bloodstream infections between 1997 and 2007.Promising though these figures may be, experts warn that it’s way too early to declare victory over the pathogen, and that labs and clinicians will need to continue the close collaboration that’s been a hallmark of successful MRSA control efforts. “Though we’ve been making headway, it’s mostly attributable to increased attention to healthcare acquired infections in general. Even with reductions like those cited by CDC, that still leaves a sizable portion of infections still occurring,” said Daniel Diekema, MD, professor and director of the division of infectious diseases at the University of Iowa Carver College of Medicine in Iowa City. “Labs are key to all of this. Their major role first and foremost is to promptly and accurately identify MRSA. They also need a seamless notification program so clinicians know where infections are occurring and can intervene quickly, and they need to provide support during outbreak investigations.”

A Mixed Picture

Despite the reported progress, mixed outcomes reported in the literature exemplify how elusive inroads can be and how complex the science around MRSA is. As an example, two reports published earlier this year in the same issue of the New England Journal of Medicine had strikingly different results.

In one paper, Veterans Affairs (VA) researchers reported that a MRSA care bundle implemented throughout the VA Healthcare System led to a 62% decline in healthcare-associated MRSA infections in ICUs, and a 45% decline in non-ICUs (N Engl J Med 2011;364:1419–30). The MRSA care bundle consisted of universal surveillance using either culture or rapid PCR testing, contact precautions, hand hygiene, and institutional culture change. Nearly 2 million patients had nasal swabs taken at the time of admission, at the time of any room transfers, and at discharge. This quality improvement initiative was not designed as a clinical trial, so the investigators were unable to determine how much each bundle component contributed to the outcomes. However, in the eyes of co-author Martin Evans, MD, the foundation of the initiative’s success had to do with active surveillance testing of all patients.

“That was important for continually reminding our staff about infection control because whenever a patient was admitted, staff members had to sit down and explain why they were doing the swab and get verbal consent to do the swab. I think that really drove our culture transformation because people were continually reminded that infection control was important and that they personally were an important part of making infection control happen in the institution,” he explained. Evans is director of the VA’s multidrug-resistant organism (MDRO) program and an infectious diseases physician at the Lexington VA Medical Center in Kentucky.

While the VA initiative was successful, the Strategies to Reduce Transmission of Antimicrobial Resistant Bacteria in Intensive Care Units (STAR*ICU) trial unexpectedly showed no benefit from a targeted MRSA screening program in 18 participating ICUs. The STAR*ICU investigators found that an intervention involving culture-based surveillance for MRSA colonization and use of enhanced barrier precautions was not effective in reducing transmission of MRSA or vancomycin-resistant enterococcus (VRE) in comparison to usual practice (N Engl J Med 2011; 364:1407–18).

Although the STAR*ICU investigators explained that their study looked at MRSA transmission rather than infection outcomes, the findings were still a surprise since active surveillance identified patients not previously known to be colonized and all patients found to have MRSA or VRE were cared for with barrier precautions during their ICU stays. The authors cited a number of factors that may have caused this unexpected finding, including a prolonged turnaround time for reporting positive results, less-than-ideal compliance with contact precautions, and an insufficiently long intervention period, among others.

Universal or Targeted Surveillance

The VA and STAR*ICU initiatives illustrate the choices healthcare organizations have in setting up MRSA surveillance programs. The VA opted for universal active surveillance testing of all patients, while the STAR*ICU trial targeted high-risk patients. CDC’s 2006 guidance on the management of MDROs in healthcare settings contemplates both approaches. The Agency’s two-tiered recommendations suggest that organizations implement second-tier, more resource intensive interventions when the incidence or prevalence of MDROs does not decrease despite routine control measures, or when the first case or an outbreak of an epidemiologically important MDRO is detected (See Table, below).

“We think as complex as the problem of antimicrobial resistance—including MRSA—is, there’s probably not a one-size-fits-all approach that can be applied equally to all healthcare settings,” said John Jernigan, MD, director of CDC’s Office of HAI Prevention Research and Evaluation in the Division of Healthcare Quality Promotion. “The key thing about our guidance is that it’s a results-directed document. If you’re having adequate success controlling MRSA without active detection, then that’s good, and you don’t need to pursue it. On the other hand, if a facility is not seeing control of MRSA without the use of active surveillance, then that may be something it considers as an additional measure until they do gain control.” Jernigan also is a clinical associate professor of medicine at Emory University School of Medicine in Atlanta.

Despite its success, the VA approach is not without controversy. Technically, the program could be considered a vertical infection prevention strategy focused on controlling one particular pathogen. Some experts, like Diekema, have argued that horizontal infection prevention efforts, with interventions aimed at detecting and controlling a range of organisms, are more beneficial in the long run. “I think it’s a mistake to have a silo-type approach for particular organisms. If you feel, for example, that in order to prevent infection with a particular bacterium you have to know every person who carries that pathogen, you’ll always be behind the eight-ball,” he contended.

The VA’s Evans argued that what at first blush appeared to be a vertical infection prevention strategy against MRSA turned out not to be. “You might categorize our effort as more vertical than horizontal, but I dispute how vertical it really is. We did focus on MRSA initially and active surveillance for that single organism was a big part of the program. However, remember, there were three other components to the bundle. So I would argue that what we achieved overall was more of a horizontal effect. If you look carefully at our paper you’ll see that rates for C. difficile and VRE also went down during our surveillance period,” he said.

A Question of Resources

For many organizations, the method of MRSA surveillance boils down to the availability of resources. “If everyone had the resources and financial support, the ultimate surveillance would be universal surveillance where you’re screening all patients as they come in in the door and taking appropriate precautions,” contended Michael Rybak, PharmD, MPH, professor of pharmacy at the Eugene Applebaum College of Pharmacy and Health Sciences at Wayne State University in Detroit. “Those kinds of maneuvers are going to make the greatest impact in terms of controlling the rate of MRSA colonization and infection, but every institution has to make that determination based on their finances and ability to manage a total surveillance system.” Rybak also is director of the Anti-Infective Research Laboratory and adjunct professor of medicine at Wayne State University.

Others have urged a more pragmatic surveillance approach based on the rising prevalence of community acquired MRSA. “Cutting down MRSA in the hospital is a good goal, but the problem is, an increasing number of people in the community have it whether they’re in or out of the healthcare system,” observed Susan Boyle Vavra, PhD, associate professor and lab director at the University of Chicago Medical Center’s MRSA Research Center. “We’ve seen an uptick in community acquired MRSA genotypes—especially the USA 300 strain. It’s circulating pretty nicely in the community, so screening people at admission is only going to deal with part of the MRSA issue.”

Beyond questions of resources and infection patterns at individual institutions and in the community, at least 15 states have enacted legislation around MRSA (See Map, below). For instance, in 2007, the Illinois legislature passed a mandate requiring active surveillance screening of all hospital ICU patients throughout the state.

Rapid Testing Versus Culture

Rapid molecular testing for MRSA costs approximately $25–75 per test, compared with about $5 per culture, but delivers results within a few hours as opposed to 24–72 hours depending on the culture medium used. These tests also have reported sensitivity of 100% and specificity >98% in comparison to culture. With hospital stays for MRSA costing about twice that of others and the nationwide excess cost associated with MRSA infections estimated to be $3.2–4.2 billion annually, many argue that the trade-off between increased costs and speedy results makes rapid testing as part of an active surveillance program well worth the investment.

“Results like we had are staggering. The impact on patient safety, drop in the rate of infections, and impact to the bottom line with a cost savings of almost $2 million over a three-year period will get the attention of anyone in the C-suite,” said Denise Uettwiller-Geiger, PhD, director of clinical trials and laboratories at John T. Mather Memorial Hospital in Port Jefferson, N.Y. Mather Memorial implemented a rapid MRSA screening method in 2008 as part of the hospital’s The Bug Stops Here campaign to decrease MRSA and other healthcare-associated infections.

This 248-bed community hospital does not have a traditional microbiology service, so rapid molecular testing was an excellent option, according to Uettwiller-Geiger. The lab ultimately went with a method that was slightly more expensive but requires one less step, an important consideration for a start-up service. “We incorporated the instrument in our core lab with no additional staff, and testing is available 24/7 on demand. We also felt from a workflow standpoint it involved fewer steps and was slightly easier to use.” Prior to implementing surveillance in high risk ICU, cardiac care, orthopedic surgery, and telemetry patients, the hospital’s infection rate was 0.90 per 1000 bed days, but by 2010 had dropped to 0.25 per 1000 bed days, corresponding to an estimated $1.9 million in avoided costs.

Some experts have argued that rapid testing is a game-changer because of its critical role in enabling caregivers to implement isolation and contact precautions earlier in a patient’s hospitalization. One such analysis proposed that capturing at least 80% of MRSA isolation days is a critical threshold that’s basically not attainable with culture-based surveillance (J Clin Microbiol 2010;48:683–89). However, Jernigan and others pointed out that some hospitals have successfully reduced MRSA rates using only culture in their surveillance efforts.

The Role of Culture

Regardless of how it’s used in day-to-day surveillance, culture will continue to have a place in MRSA control efforts, according to Brandi Limbago, PhD, deputy chief of the Clinical and Environmental Microbiology Branch in CDC’s Division of Healthcare Quality Promotion. “You can’t know what a bacterium can be treated with successfully without doing susceptibility testing, and that’s phenotypic testing,” she explained. “Also, if you’re doing any epidemiologic strain typing studies you need an isolate for that as well. Those are two practical reasons why culture is important.”

Limbago also emphasized that rapid methods would not immediately detect genetic changes in the pathogen. “There’s a new MRSA strain just recently described in Europe. It doesn’t appear to be terribly common yet, but what’s important about it is that it wouldn’t be detected with the current molecular methods, but it would be with all the routine phenotypic culture-based methods,” she said. “So if you were going with an active surveillance approach with a rapid assay, this is one type of MRSA that would be missed, were it to be circulating in the U.S.”

What’s the MIC

Another area where the evidence base is changing involves the use of vancomycin susceptibility testing to guide therapy. Guidelines on the management of MRSA issued earlier this year by the Infectious Diseases Society of America (IDSA) call for clinicians to rely on patients’ clinical response to vancomycin rather than a minimum inhibitory concentration (MIC), when MIC is ≤2 µg/mL (Clin Infect Dis 2011;52:1–38). “Over time, there’s almost a folklore that’s developed about vancomycin MIC and treatment response,” explained Limbago. “The idea was that isolates with a MIC of two—which is still susceptible—don’t respond as well as those with lower MICs. The problem is, for most MIC assays, the value you get is the correct value plus or minus one dilution. So if you get a one, it could be a two, and if you get a 0.5, it could be a one. That’s a known limitation of the testing.”

Limbago added that the automated testing most labs use for this purpose provides a breakpoint of at, above, or below a value, like a qualitative score. “Most of the automated testing doesn’t give a range of MICs, so determining the actual MIC number can be difficult,” she said. “The IDSA guidelines are saying the responsible thing to do is rely on patient response, especially in people who’ve already been treated with vancomycin. If they’re not responding well or continue to have positive cultures, and the MICs are increasing, that’s more likely to represent failure or resistance to vancomycin.”

Rybak concurred that there has been confusion around the role of MIC in monitoring response to vancomycin, but added that the evidence base is changing around this issue. “There are quite a number of papers that have examined the association of vancomycin MIC and outcome in patients with S. aureus infection,” he said. “I think most agree that a MIC to vancomycin of between 1 and 2 µg/L is not that different and within the dilution error of the broth microtiter method of determining the MIC. However, it does make sense that there would be a relationship between increasing MIC and response to vancomycin. We just have to determine at what level of the MIC is it no longer warranted to treat with vancomycin.”

A Team Approach

As the science around MRSA evolves, experts promoted laboratorians as crucial members of the infection control team. Diekema and Evans both noted the benefits their respective organizations have realized by having laboratorians actively involved in infection control rounds and other team meetings. Rybak recommended that labs develop an alert system like that employed at Detroit Medical Center. “We have an electronic data capture system, and if there’s a positive result for MRSA, a team of clinicians is paged, emailed, or texted so they can see if the patient is receiving appropriate therapy according to the most recent IDSA guidelines for MRSA,” he explained.

Rybak stressed, however, that technologies alone are not the answer to MRSA control. “If you don’t put into place an intervention system that goes in concert with this new technology, then the technology will be relatively useless,” he observed. “When we first adopted this wonderful PNA FISH hybridization system, it was great that we had quick results, but patients were remaining on the same therapy because the results were just going into patients’ charts. It required an active intervention by a team of people to put that technology to its best use.”

Uettwiller-Geiger urged laboratorians to be full partners in the MRSA challenge. “It’s an opportunity for them to champion a program that contributes to patient safety and the hospital’s bottom line while meeting regulatory requirements. I couldn’t be more proud of something the lab can have such a big stake in.”