May 2008
Feature Story

Karen Lusky

The evidence may not yet be sufficient to promote a best-practice approach to reining in nosocomial MRSA infection rates, but the stakes for U.S. hospitals in battling the superbug could be higher than ever.

MRSA is a growing public health menace that is drawing increasing attention from lawmakers, and all bets are off as to how payment will play out starting Oct. 1. That’s when Medicare will stop reimbursing hospitals for the extra costs associated with certain hospital-acquired infections—mediastinitis after coronary artery bypass graft surgery, cathet­er-­associated urinary tract infection, and vascular catheter-associated infection, all of which MRSA can cause.

In its fiscal 2009 hospital inpatient proposed rule published in April, the Centers for Medicare and Medicaid Services proposes additional hospital-acquired infections for which Medicare would not pick up the higher tab starting Oct. 1, 2008, if they are in the final rule. The proposed list includes ventilator-associated pneumonia, sur­gical site infections following certain elective procedures, and Staphy­lo­coccus aureus septicemia.

The Association for Professionals in Infection Control and Epidemiology “is feeling a bit suspicious that CMS may roll MRSA into the Staph aureus bloodstream infection never event, although in reading the proposed rule it doesn’t sound like that’s the intent,” says Denise Graham, executive vice president of the association.

Of course, where Medi­care goes, private insurers tend to follow. And a number of national health plans have announced already they won’t pay for the preventable health-care– associated conditions Medicare identified for 2008. Nancy Foster, vice president of quality and patient safety policy at the American Hospital Association, says private payers are all looking at not paying the additional costs associated with certain complications deemed preventable by, among others, the CMS and the National Quality Forum.

Hospitals are responding to the emerging MRSA challenge in various ways. But three U.S. hospitals that have put on the gloves (and gowns) and rolled out rapid MRSA screening and decolonization for all patients are emerging as winners, at least based on data reported to date.

Pitt County Memorial Hospital in Greenville, NC, which implemented a universal surveillance and decolonization program in February 2007, has a goal of zero MRSA transmission, says Keith Ramsey, MD, the hospital’s medical director for infection control. So far the results for 2008 have been encouraging, he says, with fewer than 10 cases in the first quarter.

Preliminary results show that Pitt County’s MRSA surveillance program, which uses the BD Gene­Ohm MRSA PCR-based test to screen all patients for MRSA and then topical decolonization (nasal mupirocin ointment and chlorhexidine antiseptic for bathing), lowered MRSA ventilator-associated pneumonia rates in the surgical intensive care unit by 68 percent over a one-year period. The hospital’s rates of MRSA ventilator-associated pneumonia decreased from 1.74 to 0.54 per 1,000 ventilator days, as reported in a poster session at the Society for Healthcare Epidemiology of America’s 2008 annual meeting.

Pitt County’s MRSA prevention program also reduced catheter-associated urinary tract infections by 60 percent, and central-line associated bloodstream infections by 29 percent, says Dr. Ramsey, lead researcher for the study and professor of medicine at Brody School of Medicine, East Carolina University. He declined to disclose the overall decrease in MRSA infections pending publication of the data.

The nares of patients admitted to Pitt County Memorial were tested by the GeneOhm rapid PCR test. The hospital preemptively isolates patients at high risk for MRSA, vancomycin-resistant Enterococcus, or both. (Patients are tested at admission for VRE also.) Other patients are not put into preemptive isolation because the hospital does five runs a day of the GeneOhm test for MRSA colonization. Thus, there’s never more than a three- to four-hour time lapse from when the nasal swab is done and the results are available.

Patients scheduled for surgery are tested in outpatient settings and decolonized before surgery.

The hospital finds that about eight percent of patients are colonized, which means, Dr. Ramsey says, that “one in 12 admissions are silently carrying MRSA. As high as 30 percent of high-risk patients are carriers, but the range on that varies.”

Evanston (Ill.) Northwestern Healthcare (ENH), the first hospital system in the nation to deploy universal MRSA surveillance and decolonization, has also reported robust results from its program. It includes nasal PCR testing of all admissions and then implementing contact precautions and topical decolonization for those who test positive for MRSA (a standard regimen of mupirocin applied to the nares for five days plus use of chlorhexidine during bathing on days one, three, and five).

As reported in the March 18 Annals of Internal Medicine, ENH’s three affiliated hospitals lowered the rate of hospital-acquired MRSA infections by 70 percent over a 21-month period compared with baseline MRSA infection rates measured for 12 months (Robicsek A, et al. Ann Intern Med. 2008;148:409–418). The reduction in infections occurred not only in the hospital but also for 30 days after discharge, says Lance Peterson, MD, director of microbiology and infectious disease research at ENH.

Patients at ENH’s three hospitals are tested at admission but are not put on contact precautions until the GeneOhm PCR test shows they are carrying MRSA. The average turnaround time for the test is a little under 16 hours. “The GeneOhm assay takes about two hours [to provide results] if you don’t batch it,” but ENH does run the test in batch mode, resulting in the longer turnaround time, Dr. Peterson says.

The hospital system achieved the impressive results in spite of the turnaround time. As Dr. Peterson explains: “We have models that show how turnaround time for MRSA screening impacts the percentage of days that people are on isolation if they turn out to be a MRSA carrier.” And “with a TAT of under 16 hours, you capture over 90 percent or more of the days you’d be isolating people. If we improved our TAT to about eight hours, we’d only improve isolation capture by about four or five percent. It’s unknown if that would make a big difference in the outcome of our program.”

Loyola University Hospital in Maywood, Ill., moved to universal screening of all inpatients using rapid molecular diagnostic testing on Nov. 27, 2007, after a six-week phase-in period during which the hospital screened all ICU patients using slower culture-based screening methods. Fully implemented, the program includes offering decolonization to those found to be carrying MRSA, says Jorge Parada, MD, MPH, Loyola’s medical director of infection control. “That being said, we found that in terms of health care providers writing orders for decolonization, we were well short of universally decolonizing patients.” This is important, Dr. Parada explains, because it suggests that the reduction in nosocomial MRSA rates associated with the program to date are attributable primarily to the universal surveillance program, not the decolonization treatments. And, he says, “It affords the potential for further improvement as we improve our decolonization of patients as the program continues.”

Loyola uses Cepheid’s GeneXpert rapid PCR test, the only other FDA-approved rapid PCR test for MRSA. The hospital has an average four-hour turnaround time from swabbing the patient’s nares to reporting results, says Paul Schreckenberger, PhD, director of the clinical microbiology laboratory and associate director of diagnostic molecular pathology at Loyola University Hospital.

Dr. Schreckenberger says the shorter turnaround time is important in assigning patients to rooms based on their MRSA status in a hospital like Loyola that, unlike ENH, has semi-private rooms. That’s because having to move a patient out of a semi-private room when his or her roommate tests positive for MRSA colonization can be stressful for patients and families, he says. And there’s a real concern about the patient who shared a room and bathroom with someone colonized with MRSA.

Data through March 2008 from the Loyola MRSA surveillance program show the hospital had the lowest hospital-acquired MRSA rates to date, Dr. Parada says. “Overall, our rates have dropped by two-thirds (67.6 percent) as compared to the rates during the six months pre-surveillance program,” he says.

Dr. Parada says the hospital’s baseline for the six months pre-surveillance was 0.529 MRSA nosocomial infections per 1,000 patient days. After implementing surveillance of all ICU patients in mid-October through late November 2007, that rate dropped to an average of 0.392 per 1,000 patient days. Now data through March 2008 show the rate to be 0.171 per 1,000 patient days.

Loyola will not formally analyze the costs and benefits of its MRSA program until it has more established numbers. But Dr. Schreckenberger predicts the universal screening program will ultimately save the hospital money. Using a rapid MRSA screening method makes it possible for the hospital to avoid unnecessary preemptive isolation for patients at high risk for colonization with MRSA who turn out not to be carrying the organism in their nares. “The cost of an isolation room is about $200 a day” for disposables alone, such as gowns and gloves. “By using the rapid MRSA screening at admission, we avoid putting people in an isolation room who don’t need to be there,” Dr. Schreckenberger says.

ENH researchers found that depending on how they did the calculation, the hospital system is “at least breaking even and likely quite far ahead economically,” Dr. Peterson says, “in addition to having a major impact of reducing MRSA infections for our patients.”

Pitt County Memorial Hospital expects to “actually come out in the black in 2008” with its universal screening and decolonization initiative, Dr. Ramsey says. “We do bill insurers for the testing. The program cost $950,000 in 2007, and we got $700,000 from insurers for the testing. Conservatively, we saved about $175,000 in reduced numbers of infections, so it came out to be about a $75,000 investment. We bought five SmartCyclers, a one-time purchase [for running the GeneOhm test], and those came to about $175,000. And we hired three to four medical technologists. The rest of the expenditure is the cost of the kits.”

While Pitt County, ENH, and Loyola University hospitals have demonstrated dramatic successes with their universal surveillance and decolonization programs, two European studies recently found no benefit to using PCR-based MRSA screening for targeted—not universal—surveillance.

As reported in the British Medical Journal in April, a London teaching hospital on two sites over a 14-month period found no difference between use of rapid MRSA screening testing (BD’s GeneOhm assay) and conventional culture in preventing hospital-acquired MRSA (Jeyaratnam D, et al. BMJ. 2008 Apr 16; [Epub ahead of print]). The researchers used a cluster randomized crossover trial in medical, surgical elderly care and oncology wards. They screened patients at admission for MRSA; patients who tested negative were rescreened at discharge.

“Patients were screened at the nares, axillae, and groin; skin breaks; and clinically indicated sites,” the article says. All patients who tested positive for MRSA were put on contact precautions and received decolonization with chlor­hexidine-based cleansing solution and powder and, depending on sensitivity, mupirocin nasal ointment ( www.bmj.com/cgi/content/full/bmj.39525.579063.BEv1?q=rss_home).

Rapid tests reduced the turnaround time (median reporting time from admission) from 46 hours to 22 hours, but the rates of MRSA transmission, wound infection, and bacteremia were similar between the control and intervention arms (3.2 versus 2.8 percent, respectively).

Based on the data, the researchers concluded it is “unlikely that the increased costs of rapid tests can be justified compared with alternative control measures against MRSA.”

PCR testing did reduce the number of days of inappropriate preemptive isolation, affecting the use of resources. The researchers did not look at the cost-benefit of that finding but plan to do so, Dakshika Jeyaratnam, MSc, MB, BS, MRCPCH, MRCPath, lead author of the study, told CAP TODAY. “There is no formal cost-benefit analysis as part of the present publication.”

“A shorter turnaround time [for the testing], such as for point-of-care testing, might reduce MRSA infection rates,” with the caveat that it would have to be in conjunction with good general infection-control practices, Dr. Jeyaratnam says.

The other European study, reported in March in the Journal of the American Medical Association, created a stir at first among some proponents of universal MRSA screening in hospitals—until they took a closer look at it.

The article, “Universal Screening for Methicillin-Resistant Staphylococcus aureus at Hospital Admission and Nosocomial Infection in Surgical Patients,” found that rapid PCR screening for surgical patients at the University of Geneva Hospitals in Switzerland had no impact on the rates of nosocomial MRSA infection (Harbarth S, et al. JAMA. 2008;299:1149–1157).

The study used an in-house–developed rapid molecular test for all adult patients who had elective and nonelective surgery but not ambulatory surgical procedures. Admission screening included the “anterior nares and perineal region and other sites (catheter insertion sites, skin lesions, or urine) when clinically indicated,” the article says.

Patients who tested positive for MRSA colonization were slated to undergo decolonization with nasal mupirocin ointment and chlorhexidine body washing. Because of delays in notification of test results and emergency surgeries, 31 percent of 386 MRSA carriers weren’t identified until after surgery. Of the remaining 266 patients identified as colonized before surgery, 43 percent received perioperative prophylactic antibiotics targeting MRSA. The hospital did not preemptively isolate patients who had no history of MRSA colonization.

The researchers reported 22.5 hours as the median time from screening to notification of test results.

Fifty-three of 93 infected patients (57 percent) in the intervention arm who tested negative for MRSA at admission developed a MRSA infection during their hospital stay. Thus, the study authors concluded: “A universal, rapid MRSA admission screening strategy did not reduce nosocomial MRSA infection in a surgical department with endemic MRSA prevalence but relatively low rates of MRSA infection.”

“The headline was about universal surveillance but the study was actually targeted only to presurgical patients,” says Andy Guhl, vice president of health-care–associated infections, BD. And the Swiss hospital is held out as having the best hand hygiene in the world, Guhl says, and adds: “It’s interesting to see a study conclude that, even with fabulous hand hygiene, a hospital is still getting MRSA transmission.”

ENH’s Dr. Peterson says a long turnaround time for the MRSA screening could have had an effect on the study’s findings. The resear­ch­ers published their median turnaround time for the rapid PCR testing as 22.5 hours, he notes, but they didn’t do testing on Sundays. So the TAT is “probably twice that long. They also did not include the time up to 12 hours to collect specimens after admission.”

Dr. Peterson points to other shortcomings. “The test was 91 percent specific, which means that nine percent were false-positives, yet only five percent of total patients were reported as test positives. I’m surprised a reviewer didn’t pick that up. There was no confirmatory testing to demonstrate that their test was actually picking up MRSA,” he says.

The researchers tested only surgical patients, which means they were probably testing 20 percent to 25 percent of the hospital’s patients. “When we did surveillance in our ICUs alone, we found no benefit either,” Dr. Peterson says.

His conclusion, after reading the JAMA article, “assuming the testing is correct,” he says, “is that in a setting with modest levels of MRSA where you aren’t testing everyone and have a relatively slow TAT even though it’s PCR, targeted surveillance isn’t good enough, even with superb hand washing.”

Loyola’s Dr. Schreckenberger notes that the JAMA article authors say that none of the patients identified as MRSA-positive who received intervention decolonization and prophylactic antibiotics targeting MRSA before surgery developed an infection. In other words, when they screened and intervened based on a positive MRSA result, they had no infections, he says.

First author of the JAMA-reported study, Stephan Harbarth, MD, MS, associate hospital epidemiologist at Geneva University Hospitals and Medical Schools, said, when he spoke with CAP TODAY, that he was in the process of responding to several letters to the editor from colleagues who are proponents of universal MRSA screening.

“Of course,” he told CAP TODAY, “we were surprised by the findings of our study because our study hypothesis was that we could make a difference [with our interventions].”

Would a faster turnaround time for the MRSA test have made a difference, in his view? “To be honest, I don’t think it would have. Our basic message related to our study is that it was a real-life study.”

“And we have to be careful in comparing different settings,” he says, noting that although the University of Geneva Hospitals detected many MRSA carriers, its rates of MRSA nosocomial infection are one per 1,000 patient days. And its hand-hygiene compliance is “pretty good, although not perfect.”

And not to say that the Geneva hospital’s rate is “very, very low,” but large academic centers in the United States have higher rates than we have, Dr. Harbarth says. “Our low rate makes it more unlikely that we could make a significant difference [with our study intervention].”

Dr. Harbarth says if he has one lesson to impart about the study it’s this: “If a hospital has just a little money available, it’s probably better to start with a hand-hygiene campaign” before doing screening. If after some time, the hospital doesn’t achieve improvements with high rates of hand-hygiene compliance, it might add screening in selected wards and patient populations.

John T. Mather Memorial Hospital, a 248-bed community hospital in Port Jefferson, NY, is taking an incremental approach to look at the impact of targeted surveillance before deciding whether to move to hospitalwide MRSA screening. The initiative, dubbed “The Bug Stops Here,” began early this year and entails active surveillance of ICU/CCU patients.

“We wanted to start small in order to gather and review our metrics, while developing a better understanding of what impact the surveillance program would have on the hospital, assignment of beds, and our patients,” says Denise Uettwiller-Geiger, PhD, DLM(ASCP), director of clinical laboratories.

The critical care nurses have been in-serviced on the technique for obtaining nasal swabs. “Appropriate specimen collection training is important because there can be preanalytical effects that may interfere—for example, heavy mucous or blood,” Dr. Geiger says. The program also focuses on infection control for environmental surfaces.

In the ICU and CCU, the hospital is using Cepheid’s GeneXpert, which is providing MRSA results in less than two hours. The fast TAT is a key factor in the hospital’s “guilty until proven innocent” of MRSA colonization strategy in which patients are assigned to a bed based on their MRSA status, Dr. Geiger says. Contact precautions are implemented until the results are available.

The hospital will soon begin active surveillance for presurgical patients scheduled for joint replacements. “We will attempt to decolonize joint-replacement patients pre­surgically,” using a protocol of seven days of mupirocin ointment to the nares before surgery and for seven days after surgery, along with showering with chlorhexidine once a day for two days before surgery, Dr. Geiger says. Contact isolation will be enforced strictly for all patients found to be colonized.

Many in the United States are looking to an initiative requiring rapid PCR MRSA screening in the Veterans Affairs hospital system to provide multi-center data about the impact of such an approach, which does not involve routine decolonization.

Veterans Integrated Service Network (VISN) 12, a seven-hospital system headquartered in Chicago, is performing rapid PCR screening (Gene­Xpert) on all admissions with the exception of mental health patients, says Bruce Dunn, MD, chief pathologist of VISN 12. In addition, they just started MRSA screening of nursing home patients.

Dr. Dunn says the Department of Veterans Affairs required all hospitals to implement a MRSA screening program by March 15, 2007 to assess all ICU patients. “Then by Oct. 1, 2007, we phased in screening of all acute medicine and surgery patients” as part of a national mandate based on success at the Pittsburgh VA in particular, he says.

The VA program also involves a strict hand-washing program for which VISN 12 hospitals are reporting 90 to 100 percent observed compliance. All patients who test positive for MRSA go on contact precautions. Decolonization is not done routinely, but clinicians have the option of ordering it for patients before selected types of surgery.

The average TAT from specimen receipt in the lab and availability of results in the LIS in a typical VISN 12 lab is less than 100 minutes. “Conventional cultures are performed on PCR-positive specimens so that isolates can be typed using molecular techniques if requested,” Dr. Dunn says.

The initiative isn’t inexpensive: The annualized cost of reagents for the MRSA testing alone in the seven-hospital network is more than $1 million, Dr. Dunn says. “Another challenge is that we do not have enough private rooms in which to house all of our MRSA-colonized patients,” requiring the hospitals to group MRSA-colonized patients together.

The VA rolled out a national database into which all sites are to report their MRSA data, beginning in October 2007. At this point, it’s too early to draw general conclusions about how well the program is doing in VISN 12, Dr. Dunn says. But one hospital in VISN 12 has reported a “very clear and marked reduction in nosocomial transmission of MRSA in its ICUs and in acute medicine and surgery patients.” And one of the smaller hospitals in the system has not reported a single case of hospital-transmitted MRSA in the past two years.

Donna Wolk, PhD, D(ABMM), clinical director of microbiology at the Southern Arizona VA Health Care System and University Medical Center, Tucson, says the VA hospital, which implemented active surveillance in high-risk patient care units in January 2005 and achieved hospitalwide MRSA screening in January 2007, is today near zero transmission of MRSA infections. All members of the VA health care team worked together to re-engineer the hospital environment to make laboratory reporting, isolation precautions, and hand hygiene work in concert, Dr. Wolk reports. The hospital uses anonymous observers to assess hand-hygiene compliance.

As for how and when VA hospitals will begin to report formally the outcomes of their MRSA prevention efforts, Robert R. Muder, MD, chief of the infectious disease section for the VA Pittsburgh Healthcare system, expects the data to be reported for the VA as a whole. Given that acute-care facilities moved to whole-house surveillance six months ago, he believes another year’s worth of data will be needed before the findings can be reported.

The VA data will not shed light on all of the questions hospitals would like answered about how to stamp out spread of MRSA. For example, says Graham of the Association for Professionals in Infection Control and Epidemiology, many hospitals don’t have enough private rooms to effectively isolate people. And there are questions about whether to group people with the same strain of MRSA, which requires additional testing, she adds.

Decolonization also remains controversial. Graham says the association “would have loved to have included that recommendation in our MRSA implementation guideline for our membership based on what was happening at ENH, but the data aren’t there and we can’t promote it as a best practice.”

The problem with decolonization, says Loyola’s Dr. Schreckenberger, is that it’s an effective short-term strategy but not shown to always last because many people become recolonized. Studies underway are looking at the role of MRSA decolonization for hospital patients, he says.

Based on the data to date, the CDC guidance recommends that all health care facilities implement a multifaceted MDRO prevention program and carefully monitor the impact, says the CDC’s John Jernigan, MD, MS, deputy branch chief, Prevention and Response Branch, Division of Healthcare Quality and Promotion. The components of the initial program may vary from facility to facility depend­ing on the char­ac­teristics of patient population, he adds. If facilities find their selected approach is not having the desired effect, they should modify their approach according to specific CDC recommendations, “which in some settings would include use of active surveillance in high-risk populations,” Dr. Jernigan says.

The Association for Professionals in Infection Control and Epidemiology also advocates a broad-based approach where requirements for MRSA control allow hospitals the flexibility to choose strategies based on an institution’s risk assessment. “Why,” asks Graham, “would a facility that has driven its MRSA rates way down want to be mandated to spend money on that issue when, based on their risk assessment, its problem might be tuberculosis or something else depending on the patient mix?”

Each hospital can be viewed as having its own uniquely configured MRSA battleground where the organism is more problematic in certain areas than others. Based on that concept, MultiGen Diagnostics of San Diego will begin this summer to offer a comprehensive service that includes a “point determinant” analysis to identify and monitor genes/mutations associated with MRSA and other nosocomial infections in hospitals, using its proprietary multi-target sequencing technology, says T.V. Moorthy, PhD, president and CEO of the company. “The goal is to help the hospital do a comprehensive analysis to generate scientific evidence as the basis for an intervention program tailored to that specific hospital,” he says.

Whatever hospitals do about MRSA, they shouldn’t take the low road on the issue, says Dr. Schreckenberger. “We really need to get people thinking about doing MRSA surveillance and other types of surveillance at admission....If hospitals don’t do that, then legislatures may step in and make laws forcing them to do so.” A handful of states, including Illinois, have legislation requiring MRSA screening of some or all patients. (To see a map showing the status of MRSA legislation on the APIC Web site, go to http://www.apic.org/am/images/maps/mrsa_map.gif.) A federal bill, S. 2525, has been introduced that would essentially mandate MRSA surveillance in all U.S. hospitals.

The bottom line on MRSA prevention in hospitals, in the view of the VA’s Dr. Wolk, is that no one should get sicker because he or she visited a hospital for care. Yes, the jury may still be out on the effectiveness of some active surveillance practices, she agrees. However, “in the real world, you can’t always wait for that evidence to be 100 percent conclusive. You try to do the right and logical thing for your own patient population.”

Related Links

• Staphylococcus aureus screening methods, today and to come