Distribution of environmental L.pneumophila sg.1 strains with heterogeneous chlorine-susceptibility patterns inside hospital plumbing system

Abstract number: P862

Casini B., Baggiani A., Torracca F., Battisti F., Valentini P., Privitera G.

Objectives: To describe the distribution of Legionella strains and their evolution during the application of a five-year Water Safety Plan (WSP) in a university hospital based on a disinfection-filtration strategy.

Methods: Between March 2002 and June 2008, 224 out of 698 water samples were positive for Legionella spp.; 91 strains were selected on temporal and spatial criteria and were analysed applying both the comparison of the nucleotide sequences of six genes by SBT or the comparison of the genomic profile by PFGE. The chlorine-susceptibility of representative isolates was assessed according to the BS EN 1040:1997. Feasibility of a rapid assessment of chlorine-susceptibility using quantitative ATP detection was also evaluated.

Results: The majority of the isolates belonged to Legionella pneumophila serogroup 1. Molecular typing indicated the presence of three prevalent types of L. pneumophila Wadsworth (Type 1: 28/91. Type 2: 51/91. Type 3: 2/91) with different chlorine susceptibility. Type 1 and 2 were pre-existing chlorination, Type 1 being initially predominating, while Type 2 became widespread over the years. Type 3 was isolated only occasionally. Type 2 was fully susceptible to chlorine, while Type 1 and 3 were chlorine-tolerant. Following exposure to chlorine, ATP levels correlated well with the measured number of viable cells of the three Types, but the luminescence signals were significantly decreased only for Type 2, compared to untreated controls.

The WSP allowed to control the spread of Legionella spp. in the hospital water distribution system by the adoption of an integrated disinfection-filtration strategy, presumably due to the application of filters at those points-of-use where Type 1 persisted despite chlorination. The predominance of Type 2 in the last period of environmental monitoring could be explained either by the inability to obtain effective chlorine concentration at every point of use or by the emergence of a resistant Type 2 phenotype.

Conclusion: Standard environmental surveillance methods may not be sufficient to choose the most effective and efficient water safety plan, and should at least in some instances be accompanied by in vitro evaluation of the susceptibility of the environmental isolates to the sanitising agent considered.