Research News

Screening for MRSA and isolating carriers doesn’t reduce ICU infections

BMJ 2013; 346 doi: (Published 04 June 2013) Cite this as: BMJ 2013;346:f3581

A trial randomised 43 US hospitals to one of three strategies for combating infections in intensive care units (ICUs). The first screened for meticillin resistant Staphylococcus aureus (MRSA) and isolated those who screened positive (group 1). The second added decolonisation of MRSA carriers with intranasal mupirocin and chlorhexidine bathing for five days (group 2). The third comprised decolonisation, with intranasal mupirocin over five days and chlorhexidine bathing for the entire stay, of all patients, without screening (group 3).

Nearly 50 000 patients were treated in the participating intensive care units during a one year baseline period and nearly 75 000 in an 18 month intervention period. The units had three months in between to adapt to the new regimens.

In group 1, the intervention had no effect on any of the outcomes compared with the baseline period. This included MRSA clinical isolates (3.2 v 3.4 events/1000 days), bloodstream infection with MRSA (0.6 v 0.7), and bloodstream infection with any pathogen (4.1 v 4.2). In group 2, reductions were seen for MRSA clinical isolates (3.2 v 4.3) and bloodstream infections with any pathogen (3.7 v 4.8), but not for bloodstream infection with MRSA (0.6 v 0.5). Universal decolonisation had no effect on this last outcome either (0.5 v 0.6), but effect sizes for the other two outcomes were greatest with this intervention. It reduced MRSA positive cultures from 3.4 events per 1000 days in the baseline period to 2.1, and bloodstream infections with any pathogen from 6.1 to 3.6.

To prevent one MRSA positive culture, 181 patients needed to undergo decolonisation; for one case of bloodstream infection avoided, 54 patients needed to be decolonised. Only seven patients had adverse events—mild pruritus or rash due to chlorhexidine, which resolved after discontinuation. Still, the editorialists warn that widespread use of universal decolonisation needs to be considered with caution, owing to emerging resistance to mupirocin and possibly chlorhexidine (doi:10.1056/NEJMe1304831).


Cite this as: BMJ 2013;346:f3581