Water–Vapor treatment of Shower Systems in healthcare facilities: An effective protocol to contain the risk of exposure to Legionella of at-risk patients

Study design
Shower systems were installed in replicate across 5 different settings. In the healthcare facility Lausanne (University hospital of Lausanne, Switzerland) and in the nursing home in St-Etienne (France) shower units were installed in triplicate for each treatment condition (no treatment, water-vapor treatment, or conventional thermal disinfection) and for each biofilm age targeted for sampling. In the domestic settings, three residential setting located in Nods, Nidau and Prêles (Switzerland) were selected based on the presence of at least two independent shower cabins per setting, allowing for paired comparisons between treated and untreated conditions. The frequency of use was controlled to ensure biofilm grow. At 5 weeks, 32 weeks, 52 weeks and 104 weeks, paired biofilms were either directly extracted or subjected to a single treatment either water-vapor treatment or thermal disinfection prior to extraction. To assess the effects of repeated weakly water-vapor treatments, biofilms aged 32 weeks were treated once per week for four consecutive weeks, one sample was extracted after each treatment.
The efficacy of the treatment in removing biofilm from hoses was visualized by scanning electron microscopy on 1x1 cm2 sections cut from both extremities and the center of each hose. These pieces were fixed in 2.5% Glutaraldehyde / 0.1 M Cacodylate buffer pH 7.2, dehydrated with increasing concentration of ethanol and carbon-coated by thermal evaporation. Biofilms from the remaining hose sections were extracted following previously validated protocol11. Total and viable cell counts were determined by flow cytometry passing SYBR green and PicoGreen-stained aliquots on a CYTOFLEX S2 (Beckman Coulter, US), and metal concentration by inductively coupled plasma mass spectrometry (ICP-MS Thermo iCAP TQ) following previously published protocol11,12. Microbiome community composition was assessed by 16S rRNA gene amplicon sequencing targeting the V3-V4 regions, using the Illumina MiSeq plateform, when sufficient DNA was available. ICP-MS analysis was performed by the Forensic Toxicology and Chemistry Unit (UTCF) at CHUV, and sequencing by Microsynth (Switzerland). Detailed protocol is provided in Annex 1.