Point‐of‐use water filtration reduces healthcare‐associated infections in bone marrow transplant recipients

J.S. Cervia, B. Farber, D. Armellino, J. Klocke, R.‐L. Bayer, M. McAlister, I. Stanchfield, F.P. Canonica, G.A. Ortolano. Point‐of‐use water filtration reduces healthcare‐associated infections in bone marrow transplant recipients.
Transpl Infect Dis 2010: 12: 238–241. All rights reserved Abstract: Outbreaks of infection with gram‐negative bacteria (GNB) have been linked to hospital water. We sought to determine whether point‐of‐use (POU) water filtration might result in decreased risk of infection in hospitalized bone marrow transplant (BMT) recipients in the absence of any recognized outbreak. Unfiltered water was sampled from taps in the BMT unit of a major US teaching hospital, and cultured at a reference laboratory. POU bacterial‐retentive filters (0.2 μm) were installed throughout the unit, and replaced every 14 days. Infection rates were tracked over a 9‐month period, and compared with rates for a 16‐month period before POU filtration. Unfiltered water samples from 50% (2 of 4) outlets sampled grew P. aeruginosa (2 of 4) and Stenotrophomonas maltophilia (1 of 4). Clinical infection rates in the unit were significantly reduced from 1.4 total and 0.4 GNB infections per 100 patient days in the period before POU filtration to 0.18 total and 0.09 GNB infections per 100 patient days (P=0.0068 and 0.0431, respectively) in the 9‐month period for which filters were in place. Infections during the POU filtration period were due to non‐waterborne organisms. Point‐of‐use (POU) water filtration may significantly reduce infection rates in BMT recipients in the absence of any recognized outbreak.