MRSA screening: Does it work?
By Robert Muder, MD
Hospital Epidemiologist
Pittsburgh VA Medical Center
Associate Editor, Infectious Disease Alert
Synopsis: Three recent reports yield different conclusions on the effectiveness of MRSA surveillance on the incidence of health care-associated MRSA infection.
Sources: Harbarth S, et al. Universal screening for methicillin-resistant Staphylococcus aureus at hospital admission and nosocomial infection in surgical patients. JAMA 2008; 299:1,149-1,157; Jeyaratnam D, et al. Impact of rapid screening tests on acquisition of methicillin-resistant Staphylococcus aureus: Cluster randomized crossover trial. Br Med J 2008; 336:927-930; Robicsek A, et al. Universal surveillance for methicillin-resistant Staphylococcus aureus in 3 affiliated hospitals. Ann Intern Med 2008; 148:409-418.
The utility of surveillance screening for MRSA on hospital admission remains controversial. Three recently published clinical trials attempt to assess the role of MRSA surveillance.
Harbarth and colleagues conducted a prospective, interventional study of universal MRSA screening on 12 surgical wards in a Swiss university hospital. Using a crossover design, they assigned half of the wards to universal MRSA surveillance for nine months and the remainder to control procedures. Following this, the ward assignment was reversed. Surveillance consisted of swabbing of anterior nares and perineum, with identification of MRSA by PCR. All units followed standard infection control procedures, including contact isolation for patients, identified as colonized by MRSA. MRSA-positive patients were treated with intranasal mupirocin and chlorhexidine baths for five days. Compliance with screening was 94%. On the intervention wards, most (65%) of the colonized patients would not have been identified by clinical cultures. The proportion of colonized patients identified by screening was 5%. There was no significant difference in MRSA infection during the intervention period (1.11/1000 patients days) and control periods (0.91/1000 patient days).
In a study from Great Britain, Jeyaratnam and colleagues performed a randomized, crossover trial of rapid screening for MRSA by PCR vs. conventional culture methods on general medical units. They randomized 10 units with endemic MRSA infection over 14 months. Patients were screened at nares, axillae, groin, and skin breaks. Patients identified as carrying MRSA were placed in contact precautions. There was no systematic attempt at decolonization. The median turnaround time for results from admission was 46 hours for culture and 22 hours for PCR. The rate of nosocomial acquisition of MRSA in the rapid test arm (4.4/1000 patient days) was not significantly different than the rate in the culture arm (4.9/100 patient days). Of note, the median rate of observed compliance with hand hygiene was 42%.
In the United States, Robicsek and colleagues conducted an observational study of the effect of MRSA surveillance in three university-affiliated hospitals. During the 12-month baseline period, there was no surveillance. During the subsequent 12 months, MRSA surveillance was conducted on all patients admitted to the ICU. For the next 21 months, surveillance was applied to all admitted patients in the three facilities. Surveillance was conducted by swabbing of the nares and detection of MRSA by PCR. Patients colonized with MRSA were placed in contact precautions. During the final period (universal surveillance period), patients identified as being colonized with MRSA were given topical mupirocin + chlorhexidine decolonization treatment. The main outcome measurement was total nosocomial MRSA infection. During baseline, ICU surveillance, and universal surveillance periods, the rates of MRSA infection were 0.89, 0.74, and 0.39 per 1000 patient days, respectively. The difference between the baseline and universal screening period infection rates was statistically significant (absolute change -0.50/1000 patient days; CI -0.66 to -0.35/1000).
Commentary
Despite the enrollment of large numbers of patients, with two of them being randomized, crossover designs, these three studies come to seemingly disparate conclusions, and are unlikely to settle the controversy over the utility of universal surveillance for MRSA in reducing the burden of MRSA infection.
The Swiss investigators evaluated the effect of rapid MRSA screening by PCR vs. no surveillance and found no change in infection rate despite detecting many MRSA colonized patients who would not have otherwise been identified. The British investigators did not find a change in the rate of MRSA transmission when culture surveillance was replaced by more rapid PCR testing. Both groups studied the effect of changes in surveillance strategies alone; there was no systematic change in other infection control practices. It's notable that hand hygiene compliance in the British study was a rather disappointing 42%. A recent report from the same Swiss university hospital found that hand hygiene compliance in the intensive care unit was only 39%.1 Clearly, knowledge of a patient's MRSA colonization status is not useful if it does not lead to a change in behavior on the part of health care workers.
The U.S. investigators, on the other hand, found that universal surveillance for MRSA on admission, supplemented by topical decolonization with mupirocin and chlorhexidine, was followed by a significant and sustained fall in nosocomial MRSA infection rate. This study had some methodological weaknesses, including a before-after design and the implementation of multiple interventions (i.e., surveillance plus decolonization).
These three studies highlight the formidable challenges inherent in studying the effectiveness of studying MRSA control strategies. One major problem is that individual interventions, such as universal surveillance, are unlikely to work as a single, effective measure. If patients are known to be colonized, but hand hygiene and isolation practices are not maintained at effective levels, there is not likely to be much benefit. Meaningful, sustained improvements in infection control practice require a major change in employee behavior and organizational culture. This often leads to multiple concurrent or simultaneous interventions, as in the U.S. study. Those studies are quite difficult to randomize, and it often is impossible to determine the relative effectiveness of each of the several interventions. In fact, it may be that most or all of the interventions may be needed for an effect on infection rates. One might postulate that a high level of compliance with surveillance, communication of the results to staff, and effective implementation of contact precautions are all critical. The latter requires availability of rooms for isolation or cohorting, adequate environmental disinfection, availability of gowns, gloves, and hand sanitizer and, last but certainly not least, a high level of staff compliance with precautions. Poor performance in any one of these areas may doom the intervention to failure.
The honest debate about the effectiveness of various interventions to control MRSA and other multiresistant organisms may well be superseded by state legislatures, several of which have passed, or are considering, legislation mandating admission surveillance for MRSA.2 It's far from clear that this legislation will be effective in reducing health care-associated MRSA infection. As our Swiss and British colleagues have demonstrated, surveillance, as a single intervention, is not likely to work.
References
- Traore O, et al. Liquid vs. gel handrub formulation: A prospective intervention study. Crit Care 2007; 11:R52.
- Weber SG, et al. Legislative mandates for use of active surveillance cultures to screen for methicillin-resistant Staphylococcus aureus and vancomycin-resistant enterococci: Position statement from the Joint SHEA and APIC Task Force. Infect Control Hosp Epidemiol 2007; 28:249-260.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content