Table 3

Review of literature on the single room isolation and/or hand hygiene practice as the predominant measures to control the spread of methicillin-resistant Staphylococcus aureus in adult intensive care unit.

Study [reference]

Design and setting

Main intervention

Major outcome


Cepeda JA

et al (2005)


Prospective 12-month study in the ICU of 2 teaching hospitals (18-bed for hospital A & 10-bed for hospital B), London, UK

Phase 1 (6-month): all MRSA-positive patients were moved to single room or cohort nursed

Phase 2 (7-12 month): all MRSA-positive patients were not moved or cohort nursed

Other measures: (i) admission and weekly screening for MRSA colonization; (ii) hand hygiene was encouraged and compliance audited

MRSA acquisition rates in ICU were similar in phase 1 and 2

Suboptimal patient screening, delay in the availability of MRSA results, and low adherence to hand hygiene (21%)

Huang SS

et al (2006)


Retrospective 9-year study in 8 ICUs in an 800-bed hospital, Boston, US

Phase 1 (since 1 Sept 2000): campaign for sterile CVC placement

Phase 2 (since 1 Sept 2001): institution of alcohol-based hand rubs

Phase 3 (since 1 Jul 2002): hand hygiene campaign

Phase 4 (since 1 Sept 2003): routine admission and weekly screening for MRSA colonization and initiation of contact isolation precaution

Significant reduction in MRSA bacteremia by 75% (p = 0.007) in ICU during phase 4

Other interventions were not associated with a significant change in MRSA bacteremia

Bracco D

et al (2007)


Prospective 30-month study in a 18-bed medico-surgical ICU (6 single-bed rooms plus a 6-bed and 2-bed bay room), Montreal, Canada

Placement of patients into single room or bay room according to the availability of place

Other measures: (i) admission and weekly screening for MRSA colonization; (ii) hand hygiene practice with alcohol-based hand solution

The rate of MRSA acquisition was significantly lower in single room (1.3 per 1000-patient-days) than bay room (4.1 per 1000-patient-days) (p < 0.001)

Placement in single room may reduce MRSA cross-transmission in the institution where MRSA is not hyperendemic

Gastmeier P

et al (2004)


Questionnaire surveillance to 212 ICUs participating in KISS

To enquire the infection control practice in preventing nosocomial MRSA infection; univariate and multivariate analyses to identify risk factors for nosocomial MRSA infection

164 (77.4%) ICUs response; placement in isolation rooms or cohorts was found to be a protective factor (OR, 0.36; CI95, 0.17-0.79) in multivariate analysis

Up to 34% of the German ICUs have not isolated MRSA patients in single rooms or cohorts

Harrington G et al (2007)


Prospective 40-month study in a 35-bed ICU, Melbourne, Australia

Introduction of antimicrobial hand hygiene gel with the consumption of hand hygiene product increased from 78.1 liters per 1000-patient-days to 102.7 liters per 1000-patient-days

Other measures: MRSA surveillance feedback program using statistical process control chart

The rate of MRSA acquisition was significant lower in post-intervention (6.7 per 100 patient admission) than baseline (9.3 per 100 admission) (p = 0.047)

No admission and weekly screening for MRSA; no placement of MRSA patient in single room

Souweine B

et al (2009)


Prospective 4-month study in 2 ICUs (10-bed in a University hospital and 8-bed in a non-teaching hospital), France

Provision of alcohol-based hand rub during the intervention period

Other measures: (i) admission and discharge screening for MRSA; (ii) decolonization of MRSA patients with mupirocin nasal ointment

No significant reduction in MRSA colonization and infection after intervention

The sample size was underpower to estimate the difference

Note. CI95, 95% confidence interval; CVC, central venous catheter; ICU, intensive care unit; KISS, Krankenhaus Infektions Surveillance System (German Nosocomial Infection Surveillance System); MRSA, methicillin-resistant Staphylococcus aureus; OR, odd ratio

Cheng et al. BMC Infectious Diseases 2010 10:263   doi:10.1186/1471-2334-10-263

Open Data