The rising incidence of multidrug-resistant Gram-negative bacterial (MDR-GNB) infections acquired in intensive care units has prompted a variety of patient-level infection control efforts. However, it is not known whether these measures are effective in reducing colonisation and infection. The purpose of this systematic review was to assess the efficacy of patient-level interventions for the prevention of colonisation with MDR-GNB and whether these interventions are associated with a reduction in the rate of infection due to MDR-GNB in the intensive care unit. Searches were conducted on PubMed, Cochrane, EMBASE and World of Science databases to identify comparative interventional studies on patient-level interventions implemented in the intensive care unit. Literature published in English, Spanish or French from January 1, 2000, until April 30, 2013, was searched. A total of 631 reports were found and we included and analysed 13 comparative studies that reported outcomes for an intervention compared with a control group. There were ten randomised and three observational interventional trials evaluating seven interventions. Overall, there was a reduction in colonisation (odds ratio [OR] 0.75; 95% confidence interval [CI] 0.66 to 0.85) and infection (OR 0.66; 95% CI 0.59 to 0.75) with MDR-GNB. This trend persisted after restricting pooled analysis to randomised controlled trials (pooled OR 0.66; 95% CI 0.57 to 0.76 and pooled OR 0.62; 95% CI 0.54 to 0.72, respectively). We identified a significant reduction in MDR-GNB colonisation and infection through the use of patient-level interventions. This effect was mostly accounted for by selective digestive decontamination. However, given the limitations of the analysed trials, adequately powered controlled studies are needed to further explore the effects of patient-level interventions on colonisation and infection with MDR-GNB.
Keywords: Gram-negative bacterial infections; decontamination; infection control; intensive care units.