Controlling the Ever-Present Methicillin-Resistant Staphylococcus aureus

abstract & commentary

Source: Chaix C, et al. Control of endemic methicillin-resistant Staphylococcus aureus: A cost-benefit analysis in an intensive care unit. JAMA 1999;282:1745-1751.

Henri mondor hospital near paris has been afflicted with endemic methicillin-resistant Staphylococcus aureus (MRSA) since the 1970s, a problem familiar to many American medical centers. In view of this, Christian Brun-Buisson and colleagues implemented a control program in the early 1990s that included selective screening in selected high-risk units like ICUs. A 30% reduction in the incidence of MRSA followed this effort.1

Few American hospitals have adopted the practice of selected screening for MRSA, perhaps because the cost-benefit was unclear or because the commitment to control was lacking. Now, Chaix and colleagues have published their cost analysis in a major American journal, the Journal of the American Medical Association.

Quite simply, the keystone to the control program featured cultures of several sites plated to selective media for specific detection of MRSA in ICU patients, first on admission and then weekly thereafter. As soon as a patient was found to be colonized or infected, the patient was placed in contact isolation in a single room. Isolation was continued until discharge or eradication of loss of colonization. Treatment of colonized patients included chlorhexidine body washes on alternate days. Mupirocin was used to decolonize the nares in those patients with only nasal carriage.

Models for cost determination have been well worked out by Chaix et al. Of note for the reader, the average nurse’s salary was $46,000 for 1700 hours of work, or $27 per hour, while the average physician’s salary was $70,000 for 1700 hours, or $41 per hour. Isolation costs were determined over the duration of the patient’s stay. The overall intent of the study was to determine the ranges of probabilities that being an MRSA carrier at the time of ICU admission that would make the strategy of targeted screening (in the terms of sensitivity analysis) favorable.

The results were compelling. During the study from 1993-1997, 85 patients became infected with MRSA and 27 were chosen for analysis compared to controls. Compared to controls, MRSA cases were more likely to have a longer stay, have a fatal outcome, and undergo multiple procedures. The medical costs for the MRSA patients were $9550 (mean) vs. $6040 for controls (P = 0.007), and a total cost of $30,225 for MRSA patients vs. $20,950. The total extra cost for contact isolation was $655-$705 for patients isolated for an average of 20 days. Using sensitivity analysis, if the MRSA carriage rate on admission was 4%—as Chaix et al had previously determined—the transmission of MRSA was reduced 15-fold by the isolation precautions. The control measures became cost beneficial when the targeted reduction of MRSA infection was reduced only by 14%.

Some other data generated by the study are interesting. The time per day attributable to isolation precautions was 20 minutes if there was full compliance with isolation precautions. For a given patient, the total nursing time for contact isolation was 3 h 25 min to 6 h 40 min. The total cost of contact isolation and screening per patient was $365-$705 if there was full compliance. These costs compare to excess medical and total costs of $3500 and $9275, respectively, for each MRSA infection.

Comment by joseph f. john, md

There was a time (that none of us can remember) when all strains of S. aureus were penicillin susceptible. That only lasted a year or two after the introduction of this antibiotic! Now there are few no penicillin-susceptible strains. There was a time when most S. aureus were susceptible to semisynthetic penicillins. That lasted a bit longer but, eventually, the gene labeled mecA has entered a large majority of hospital strains of S. aureus and almost all nosocomial strains of coagulase-negative staphylococci. The result of the spread of mecA has been a huge increase in the use of vancomycin during the last decade. In fact, in 1998, vancomycin accounted for 10.4% of all IV antibiotic use when 11,980,614 patients received some form of IV antibiotics.2

As the epidemic of MRSA has progressed, thousands of papers, clinical and basic, have been written about MRSA. Control of MRSA has been elusive at best, with some centers, in fact, during the 1990s just throwing up their hands in dismay and ignoring attempts at control. Yet, in the last few years, several papers—including one from Chaix et al, who also authored the current paper—have appeared showing that, with proper methodology and resources, control was possible.1 Skeptics awaited data showing that successful control programs were also cost effective.

The current paper by Chaix et al does just that, the setting again an ICU in one of France’s largest hospitals. The largest issue for American hospitals is likely who is going to pay for the admission (and possibly the weekly) cultures. The admission culture cost the French hospital only $15. Moreover, for a figure of costs of control of $500, the savings was $3500. Estimates may be excessive due to a lower rate of carriage on admission (unlikely at most of our large medical centers) or an extremely low rate of transmission (unlikely from the overall rates of infection at most of our medical centers). Even if savings were less than seven-fold times the cost of the MRSA control, it is hard to make an argument for not implementing this type of control.

So how will U.S. hospitals infested with MRSA respond to the French challenge? My guess is that we will continue to support the traditional assumption that isolating patients upon accidental discovery of MRSA carriage or infection does in fact suffice as control. Implementation of the French approach may be somewhat painful at first, but Chaix et al emphasize that their hospital workers became facile at the control methodology—perhaps a confounding variable in their study—to implement selective screening.

The French program worked for a hospital that had only 15-30% of all S. aureus isolates determined to be methicillin resistant. Many American centers now have 60% or even more MRSA, so the benefits to a selective screening approach should be even more impressive and, certainly, more cost beneficial.

About the only aspect of the control program that dulls my enthusiasm is the inclusion in the French approach of whole-body washings of those patients colonized with MRSA with 4% chlorhexidine on alternate days. Since the nares are by far the most common site of colonization and probably the major site for perpetuation of MRSA, I would recommend that some centers may choose to use nasal mupirocin on nasal MRSA carriers at a first attempt to lower MRSA infection rates. For those hospitals willing to move to whole-body washings with chlorhexidine, however, the payoff will likely be more immediate and more financially gratifying.


1. Girou E, et al. Selective screening of carriers for control of methicillin-resistant Staphylococcus aureus in high-risk hospital areas with a high level of endemic MRSA. Clin Infect Dis 1998;27:543-550.

2. Lavin BS. Antibiotic cycling and marketing into the 21st century: A perspective from the pharmaceutical industry. Infect Control Hosp Epidemiol 2000;21: S32-35.

What is the average reduction in medical costs that a MRSA selective screening program will realize for each MRSA infection prevented?

a. $1500

b. $2500

c. $3500

d. $4500