Abstract
BACKGROUND: Neonatal healthcare-associated infections (HAIs) result in increased morbidity and mortality, as well as increased healthcare costs. Patient isolation measures, i.e. single-room isolation or the cohorting of patients with similar infections, remain a recommended and commonly used practice for preventing horizontal spread of infections in the neonatal intensive care unit (NICU). OBJECTIVES: Our primary objective was to assess the effect of single-room isolation or cohorting, or both for preventing transmission of HAIs or colonization with HAI-causing pathogens in newborn infants less than six months of age admitted to the neonatal intensive care unit (NICU). Our secondary objective was to assess the effect of single-room isolation or cohorting, or both on neonatal mortality and perceived or documented adverse effects in newborn infants admitted to the NICU. SEARCH METHODS: We searched the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, Embase, CINAHL, the WHO ICTRP and ClinicalTrials.gov trials registries. There were no restrictions to date, language or publication type. We also checked the reference lists of studies identified for full-text review. SELECTION CRITERIA: Types of studies: cluster-randomized or quasi-randomized trials at the level of the cluster (where clusters may be defined by NICU, hospital, ward, or other subunits of the hospital). We also included cross-over trials with a washout period of more than four months (arbitrarily defined). TYPES OF PARTICIPANTS: newborn infants less than six months of age in neonatal units that implemented patient isolation or cohorting as infection control measures to prevent HAIs. Types of interventions: patient isolation measures (single-room isolation or cohorting, or both of infants with similar colonization or infections) compared to routine isolation measures. TYPES OF OUTCOME MEASURES: the primary outcome was the rate of transmission of HAIs as estimated by the infection and colonization rates in the NICU. Secondary outcomes included all-cause mortality during hospital stay at 28 days of age, length of hospital stay, as well as potential adverse effects of isolation or cohorting measures, or both. DATA COLLECTION AND ANALYSIS: The standard methods of Cochrane Neonatal were used to identify studies and assess the methodological quality of eligible cluster-randomized trials. The certainty of the evidence was to be assessed by the GRADE method as evidence of high, moderate, low, or very low certainty. Infection and colonization rates were to be expressed as rate ratios for each trial and if appropriate for meta-analysis, the generic inverse variance method in RevMan was to be used. MAIN RESULTS: We did not identify any published or ongoing trials to include in the review. AUTHORS' CONCLUSIONS: The review found no evidence from randomized trials to either support or refute the use of patient isolation measures (single-room isolation or cohorting) in neonates with HAIs. Risks secondary to infection control measures need to be balanced against the benefits of decreasing horizontal transmission in the neonatal unit for optimal neonatal outcomes. There is an urgent need to research the effectiveness of patient isolation measures for preventing the transmission of HAIs in neonatal units. Well-designed trials randomizing clusters of units or hospitals to a type of patient isolation method intervention are warranted.