A controlled trial found that identifying nasal carriers of Staphylococcus aureus on admission to hospital using a rapid PCR test, followed by decolonisation with mupirocin and chlorhexidine soap, reduced the risk of surgical site infections, particularly deep infections.

Staphylococcus aureus is responsible for about a quarter of surgical site infections, and people who are high-level nasal carriers of the organism are at greatly increased risk of such infections. This study investigated whether the risk could be reduced by a strategy in which patients were screened for nasal S. aureus carriage and those identified were treated with a decolonisation regimen involving five days use of mupirocin nasal ointment and washing with chlorhexidine soap. Eligible patients were being admitted to general surgical or internal medicine units and were expected to stay for at least four days. They were screened using a rapid PCR test either on admission or in the week before admission, and those identified as carriers were randomised to treatment with active or placebo ointment and soap. If they remained in hospital, the treatment was repeated at three and six weeks. Primary outcome was the rate of S. aureus infection.

There were 6,771 patients screened of whom 1,251 were carriers. Of these, 353 were excluded (mainly due to ineligibility or unwillingness to participate), therefore 918 patients were randomised (917 analysed after one withdrawal of consent; active n=504, placebo n=413). Of these, 88.1% underwent some form of surgical procedure. The rate of S. aureus infection in the active group was less than half that of the placebo group - 3.4% vs. 7.7% (relative risk of infection, 0.42; 95% CI, 0.23 to 0.75). Secondary outcomes included infections at different sites - the difference was particularly marked for deep surgical-site infections (RR, 0.21; 95% CI, 0.07 to 0.62).

The authors conclude that the incidence of S. aureus infections in hospitalised patients can be reduced by rapid screening for nasal carriers of S. aureus, followed by decolonisation of those identified as nasal carriers using nasal mupirocin and chlorhexidine soap. They comment that the effect was most prominent in surgical patients, however medical patients were a minority and it is thus difficult to make inferences on effects in these patients.

An accompanying editorial discusses this study and the accompanying study on pre-operative skin cleansing. The author concludes that use of chlorhexidine-alcohol pre-operatively would be the more important intervention as it reduced the risk of infection due to all organisms; screening for S. aureus and decolonisation would be appropriate in those patients for whom a surgical site infection would be particularly adverse (e.g. open-heart surgery, implantation of any foreign body).