American Association for Clinical Chemistry
Better health through laboratory medicine
May 2008 Clinical Laboratory News: Studies Take Conflicting Stances on Universal Screening for MRSA

CLN Banner Logo

May 2008: Volume 34, Number 5


Studies Take Conflicting Stances on Universal Screening for MRSA
By Deborah Levenson 
 

With MRSA-related deaths getting lots of attention in the press, contrasting conclusions about universal MRSA screening made in two recent papers underscore the need for labs and hospitals to carefully assess their individual MRSA epidemiologies and patient populations. While many institutions have adopted active surveillance strategies that aim to reduce transmission by separating incoming MRSA carriers from other patients, only a few U.S. hospitals have thus far adopted policies that call for screening all patients.

In a paper published in JAMA, Swiss authors concluded that universal rapid screening with a multiplex PCR test did not reduce nosocomial MRSA infection in a  teaching hospital’s surgical department, which had endemic MRSA prevalence, but relatively low infection rates to start (JAMA 2008; 299(10): 1149–1157). The study involved 21,754 surgery patients in 12 surgical wards with different specialties assigned to either rapid PCR screening upon admission and standard infection control procedures, or only those procedures for 9 months. During the next 9 months, the researchers switched the groups to the other intervention.

Overall, the strategy had little effect on MRSA surgical site infections and nosocomial MRSA acquisition. Screening upon admission identified 515 MRSA-positive patients, or 5.1% of subjects. While the majority of positive patients, 65%, had not been previously identified as MRSA carriers and would have been missed without systemic screening on admission, detecting one previously unidentified MRSA carrier upon admission would require screening 30 patients, the researchers estimated.

Surprisingly, universal screening didn’t markedly improve the rate of surgical MRSA infection and nosocomial transmission. During the intervention periods, 93 patients developed nosocomial MRSA infection, compared with 76 in the control periods, the authors noted. The researchers attributed the paltry improvements in part to the fact that initial MRSA rates were relatively low, making a significant improvement less likely.

Fifty-three of 93 infected patients (57%) in the intervention wards were MRSA-free on admission and developed MRSA infections during hospitalization. These figures demonstrate “the limited value of screening on admission for patients hospitalized for extended periods in surgical services that do not perform weekly surveillance cultures.”

Turnaround Time May Play a Role

Another finding suggests that lab efficiency or the flow of information within the hospital could have played a role in the lack of improvement. Although the Swiss hospital used rapid molecular tests, positive results for 31% of the patients were only available after surgery, and 34% of patients with a surgical site MRSA infection who could have benefited from antibiotic prophylaxis didn’t receive it.

An accompanying editorial in JAMA notes that active surveillance strategies remain controversial. Daniel J. Diekema, MD, of Iowa City Veterans Affairs Medical Center and Michael Climo, MD, of Hunter Holmes McGuire Veterans Affairs Medical Center wrote that such strategies can greatly increase the number of subjects placed in isolation, a situation that puts them at risk of reduced attention from healthcare workers and of depression and anxiety. Moreover, they noted, the local epidemiology of MRSA varies widely among hospitals, so there is no “one-size-fits-all solution to the problem of MRSA prevention.” Calling for more data on universal screening, Diekema and Climo suggest a tiered approach to curb the spread of MRSA until that data are available. First, hospitals should carefully assess their MRSA problems, and then adhere to established infection control principles and pursue patient safety initiatives known to reduce morbidity and mortality from all healthcare-related pathogens, they suggest.

A MRSA Screening Success Story

In contrast to the Swiss researchers, investigators from Evanston Northwestern Healthcare, a network of three hospitals in Chicago, found that universal screening for MRSA was associated with an impressive 70% reduction in disease related to the pathogen during admission and 30 days after discharge (Annals of Internal Medicine 2008; 148: 409-–418). But the team concluded that universal screening is something to consider rather than implement in all institutions.

The 3-year observational study compared rates of MRSA clinical disease during and after hospital admission in three consecutive periods: a 12-month baseline during which no efforts were made to identify newly admitted patients colonized with MRSA, a second, 21-month period with MRSA surveillance for all admissions to the intensive care unit (ICU), and a third, year-long term marked by universal MRSA surveillance for all hospital admissions. Patients received MRSA screening via PCR, and if needed, topical decolonization therapy and contact isolation. The researchers expressed MRSA rates as prevalence density of hospital-associated clinical MRSA disease, the number of infections per 10,000 patient-days. The study did not have an unscreened control group.

For the Evanston Northwestern group, this strategy got striking results. The number of infections of aggregate hospital-associated MRSA disease at all body sites per 10,000 patient-days dropped substantially during the universal screening period, but not during the ICU-only surveillance period, compared with baseline. Figures for universal screening, ICU-only surveillance, and baseline were 3.9, 7.4, and 8.9, respectively.

Yet the researchers conclude that it’s too early to say that all hospitals should test newly admitted patient for MRSA. Rather, they urge hospitals to consider such a policy.

Teasing Out Success Factors

An editorial that accompanies the Northwestern Healthcare study urges consideration of other infection control strategies that may have helped to reduce the MRSA rate. “The straightforward interpretation of the study results—screening for MRSA reduces MRSA infections—is not necessarily correct because the hospital implemented several co-interventions when the expanded surveillance on all admitted patients,” wrote Ebbing Lautenbach, MD, MPH of University of Pennsylvania School of Medicine. Because the study did not identify the most important intervention, widespread adoption of universal screening for MRSA via expensive PCR methods would be premature, he emphasized. Like Diekema and Climo, he cautioned against a “one-size-fits-all approach.” 

Kimberle C. Chapin, MD, Associate Professor of Pathology at Brown Medical School and Director of Microbiology for the Lifespan AMC Healthcare System in Providence, R.I. agreed, but she emphasized practical considerations including budgets and labs’ capacity to increase MRSA test volume. “I do 20,000 MRSA screens per year for high-risk patients. If my hospital had universal screening, that figure would jump to 86,000. So anyone contemplating universal screening should ask: Where will all those tests be done? What staff should do it? Who will pay for it?” Because MRSA screening is a bundled charge, it does not bring significant compensation, she added.

The divergent findings—as well as legislation in several states that mandates MRSA screening of all high-risk admissions and a new Medicare policy that withholds pay for care related to these infections—will likely prompt lab directors and other hospital administrators to consider how their current MRSA screening policies, infection control procedures, and supporting mechanisms work before making changes to deal with the ongoing spread of MRSA. Until more studies are published, it seems that neither molecular tests nor universal screening are the answer for all institutions, experts agree, adding that changes in testing strategies must also be accompanied by more vigilant infection control practices.