Presented as a poster at the Neonatal Society 2016 Summer Meeting.
Payne V, Smith S, Johnson MJ, Michael Hall
Department of Neonatal Medicine, University Hospital Southampton NHS Foundation Trust, UK
Background: Central Line Associated Bloodstream Infections (CLABSIs) are associated with increased mortality, prolonged hospitalisation and increased healthcare costs. Neonatal patients are at particular risk of CLABSI, with incidences of neonatal CLABSIs estimated to be between 3.2 – 21.8 per 1000 central line days. Care bundles have reduced CLABSIs in adult ICUs but replication in paediatric ICUs has been challenging, possibly due to small numbers and substantial variations in CLABSI rates. We carried out a systematic review to assess the evidence for the efficacy of care bundles in reducing CLABSIs in the NICU. We also aimed to establish which components were most commonly used.
Methods: MEDLINE, CINAHL and EMBASE were searched from January 2010 up to April 2016. The Cochrane Library, Web of Science, Zetoc and Ethos were searched for additional studies. In order to be eligible, studies had to have assessed the effect of the introduction of a care bundle compared to standard care. RCTs or observational studies were eligible. A meta-analysis was performed using Review Manager v5 with random effects modelling. Published studies and conference abstracts were considered separately using subgroup analysis. In addition, a post hoc subgroup analysis of UK based studies was carried out in order to consider the impact of care bundles in a national context.
Results: Forty-four observational studies, performed in Level II/III NICUs met inclusion criteria. Fifteen were published studies; only two were performed in Europe, and thirteen were in North America. The remaining 27 were conference abstracts. Common bundled interventions included education and training (100% of studies), audit and feedback (93%), a skin preparation protocol (80%), establishment of a multidisciplinary team (80%) and peer surveillance (73%). A meta-analysis revealed a statistically significant reduction in CLABSIs following the introduction of care bundles (Rate Ratio = 0.38, 95% CI 0.27-0.55, p<0.001); this effect remained when abstracts were included (RR 0.40, CI 0.33-0.47, p<0.001). There was minimal heterogeneity. Seven studies (one published article and six conference abstracts) were carried out in the UK. Sub-group analysis of these studies revealed a rate ratio of 0.39 in favour of care bundles (95% CI 0.22-0.67, p<0.001), though there was high heterogeneity amongst studies (I2=82%)
Conclusion: Care bundles appear to reduce neonatal CLABSIs, though it is unclear which elements may have the greatest impact. Generalisability is complicated by variations in CLABSI definitions, multiple interventions, and the organisational context. There is a need to reach a consensus definition of CLABSI across research groups, and to better understand which elements of the care bundles work best in specific settings.
Corresponding author: vp1m14@soton.ac.uk