Ethanol locks for the prevention of catheter-related bloodstream infection: a meta-analysis of randomized control trials

BACKGROUND: Current evidence regarding the efficacy of ethanol locks in preventing catheter-related bloodstream infection (CRBI) is inconclusive. METHODS: Electronic databases, including PubMed, Web of Science, Embase, and the Cochrane Library (until April 2018),were systematically searched for relevant studies. Two reviewers independently screened the retrieved records and identified RCTs that met the inclusion criteria. Relevant data were extracted for pooled analyses using Review Manager 5.3 software. Subgroup analysis was performed according to the study quality, duration of the ethanol lock, disease type and CRBI definition. Eggs’ method was applied to detect publication bias. Sensitivity analysis was conducted to check the stability of the meta-analysis results. RESULTS: Ten RCTs involving 2760 patients were included in the analysis. The overall pooled result indicated that ethanol locks significantly reduced the incidence of CRBI (RR 0.66, 95% CI 0.51–0.86). Subgroup analysis suggested that an ethanol lock significantly decreased the incidence of CRBI in patients with hematological diseases (RR 0.50, 95% CI 0.31–0.80). An ethanol lock significantly reduced the incidence of CRBI in a2-hour ethanol lock group (RR 0.49, 95% CI 0.33–0.73). The meta-analysis showed that an ethanol lock significantly reduced the incidence of CRBI according to analysis of high-(RR 0.66, 95% CI 0.47–0.94) or low-(RR 0.66, 95% CI 0.46–0.95) quality studies. Meta-analysis of studies with a strict CRBI definition showed that an ethanol lock can significantly prevent CRBI (RR 0.61, 95% CI 0.42–0.89). The results of sensitivity analysis suggested that the pooled result was stable. Meta-analysis of adverse events showed that an ethanol lock did not increase the incidence of thrombosis (RR 1.05, 95% CI 0.51–2.18) or mortality (RR 0.99, 95% CI 0.90–1.08) but did result in increased nausea (RR 1.54, 95% CI 1.01–2.35), dizziness (RR 4.21, 95% CI 2.40–7.39),elevated blushing rates (RR 3.27, 95% CI 2.05–5.22) and altered taste rates (RR 2.61, 95% CI 1.93–3.54). CONCLUSIONS: An ethanol lock may play a role in the prevention of CRBI, especially in immunocompromised patients with hematological diseases. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1186/s12871-018-0548-y) contains supplementary material, which is available to authorized users.