In this issue of Critical Care
, AA Muller and colleagues present a study in which they assessed the efficacy of mupirocin nasal ointment in preventing methicillin-resistant Staphylococcus aureus
(MRSA) infections in an intensive care unit (ICU) [1
]. MRSA constitutes a special problem with regard to the prevention and treatment of infection. Studies show that MRSA carriers have a higher risk of nosocomial infection with this microorganism, and that those infected with MRSA have a greater morbidity and mortality than those infected with susceptible strains [2
]. It is therefore important to keep the prevalence of MRSA carriage and MRSA infections low. Efforts to achieve this should be supported.
Muller and colleagues performed a retrospective study in which they compared a 2-year MRSA control program with intranasal application of mupirocin, with a 2-year program in which mupirocin was not used. The MRSA control program consisted of screening for MRSA carriage at admission, and, in the event of MRSA carriage, hygienic measures to prevent cross-transmission were intensified. This study is clinically relevant, but the study design used is very susceptible to many biases and their results should therefore be interpreted with caution. Their findings, which suggest that intranasal mupirocin can prevent ICU-related MRSA infections, need confirmation in a well-designed clinical trial.
Results from recent clinical trials that studied the efficacy of mupirocin nasal ointment in preventing methicillin-susceptible S. aureus
(MSSA) infections can be extrapolated to MRSA infections. MSSA and MRSA are essentially the same microorganism, except that the latter is more difficult to treat with antibiotics. Several double-blind randomized placebo-controlled trials with mupirocin nasal ointment have been performed in both surgical and non-surgical patients, with S. aureus
infection as the outcome measurement [4
]. The results of these trials have been disappointing. So far, there is only evidence that mupirocin is beneficial for dialysis patients and general surgery patients [4
]. But can mupirocin nasal ointment prevent MRSA infections in ICUs?
In comparison with non-ICU patients, critically ill patients admitted to ICUs may receive more benefit from mupirocin, even when moderately efficacious, because the rate of S. aureus
infections in ICU patients is much higher. Corbella and colleagues reported a relative risk of acquiring nosocomial S. aureus
bacteremia of 59.6 (95% confidence interval 20.4 to 184.3) for nasal carriers after 14 days of ICU stay, compared with non-carriers [8
]. Another study found a relative risk for S. aureus
bacteremia of 3.9 (95% confidence interval 1.6 to 9.8) for MRSA carriers compared with MSSA carriers [2
]. These risk estimates in ICUs are significantly higher than a recently reported threefold increased risk for S. aureus
nasal carriers of acquiring nosocomial S. aureus
bacteremia in a general hospital population [9
]. Although the clinical trials described above may not be extrapolated to a high-risk setting, they do warn us not to have too high expectations of this drug.
Muller and colleagues also report the use of other preventive measures in the event of MRSA carriage. The extra hygienic precautions they implemented for this should, in our view, be standard care for any patient admitted to an ICU, irrespective of MRSA carriage. Currently there is a debate about whether infection control policies in an ICU can prevent cross-transmission of MRSA [10
]. However, history teaches us that MRSA prevalence is still low in countries where strict infection control policies were implemented soon after the first MRSA strains were detected [12
]. These policies also include health care workers and the environment, besides patients, as potential reservoirs of MRSA. Early identification, isolation and treatment of all MRSA carriers, including health care workers, and disinfection of contaminated environments are the main 'ingredients' of an effective 'search and destroy' program. The current study focuses only on the eradication of MRSA from colonized patients, which is just one part of an effective MRSA control program.
Since 2002, the first three vancomycin-resistant MRSA strains have been isolated in the USA [13
]. Preventing the spread of MRSA and preventing staphylococcal infections are therefore essential. The high usage of vancomycin in ICUs should alert us to the possibility of vancomycin-resistant strains. Effective strategies that can prevent the spread of MRSA and the development of MRSA infections still need to be developed. The use of mupirocin nasal ointment alone will not be sufficient. Increased usage of mupirocin in an MRSA endemic situation, such as ICUs, will lead to the selection of mupirocin-resistant strains [16
]. This may hamper the beneficial effects of mupirocin in other patient categories in the long term. There is no evidence yet that mupirocin is effective in preventing MRSA or MSSA infections in ICUs. An international MRSA prevention task force should be set up, with sufficient funding and expertise to develop the right strategies and validate these strategies in clinical trials. We still have a long way to go.