Risk factors for catheter-related bloodstream infections in adult hospitalized patients — multicenter cohort study. (October 2013)

• Record Type: Journal Article
• Authors: Haga, Yoshio; Miyanari, Nobutomo; Takahashi, Tadateru; Koike, Shoichiro; Kobayashi, Ryozo; Mizusawa, Hiroya; Nakamichi, Chikaaki; Goto, Mataro
• Is Part Of: Scandinavian journal of infectious diseases. Volume 45:Number 10(2013:Oct.)
• Page Start: 773
• Page End: 779
• Publication Date: 2013-10
• Subjects: Communicable diseases -- Periodicals; Infection -- Periodicals; 616.9
• DOI: 10.3109/00365548.2013.807936
• Languages: English
• ISSNs: 0036-5548

Abstract

<abstract> <title>Abstract</title> <p> <italic>Background:</italic> Risk factors for catheter-related bloodstream infections (CRBSIs) may change over time with progress in infection control. This study was undertaken to explore the current risk factors for CRBSIs in hospitalized patients. <italic>Methods:</italic> Adult patients with non-tunneled central venous catheters (CVCs) in 12 Japanese referral hospitals were prospectively enrolled between December 2009 and January 2012. Patients were monitored for CRBSIs for up to 8 weeks from CVC insertion; data were collected regarding patient characteristics, the purpose of CVC insertion, insertion methods, mechanical complications during insertion, and post-insertion catheter care. <italic>Results:</italic> A total of 892 patients were enrolled in this study. The overall incidence of CRBSIs was 0.40 infections per 1000 catheter-days. Univariate analysis using the Fisher's exact test identified one of the participating hospitals (hospital A; <italic>p</italic> < 0.001), internal jugular vein catheterization (IJVC) (<italic>p</italic> = 0.0013), not using maximal sterile barrier precautions (<italic>p</italic> = 0.030), and the Seldinger technique for catheter insertion (<italic>p</italic> = 0.025) as significant risk factors for CRBSI. After excluding data from hospital A, only IJVC remained a significant risk factor for CRBSI (<italic>p</italic> = 0.025). The cumulative probability of remaining without CRBSI was significantly lower in patients with IJVCs than in patients with other catheter routes (<italic>p</italic> < 0.001; log-rank test). Similarly, the cumulative probability of remaining without catheter removal due to a suspected infection was significantly lower in patients with IJVCs (<italic>p</italic> = 0.034; log-rank test). <italic>Conclusions:</italic> The current study suggests that IJVC might be a risk factor for CRBSI under current infection control conditions.</p> </abstract>