Published data implicate hospital water as a potential source of opportunistic fungi that may cause life-threatening infections in immunocompromised patients. Point-of-care filters are known to retain bacteria, but little is known about their efficacy in reducing exposure to moulds. We investigated the effect of point-of-use filters (Pall-Aquasafe) on the level of contamination of Aspergillus fumigatus and other filamentous fungi. The point-of-use filters were applied to several outlets (taps and showers) on the paediatric bone marrow transplantation (BMT) unit of the National Hospital in Oslo, Norway. In addition the efficacy was investigated using a test rig. The laboratory experiments showed that the filters were highly effective in reducing the number of colony-forming units for a period of at least 15 days. In the BMT unit the filters eliminated the fungi from the water on day 1 but due to particles present in the water the filters occluded, which prevented further evaluations. Our results show that point-of-use filtration might be an effective preventive measure to eliminate filamentous fungi at individual points of water use, thereby reducing patients' exposure.
In order to determine whether water or water-related surfaces are a reservoir for opportunistic filamentous fungi, water sampling in the paediatric bone marrow transplantation (BMT) unit of the National Hospital University of Oslo, Norway was performed. During a six-month period 168 water samples and 20 samples from water-related surfaces were taken. The water samples were taken from the taps and showers in the BMT unit and from the main pipe supplying the paediatric department with water. In addition, 20 water samples were taken at the intake reservoir supplying the city of Oslo with drinking water. Filamentous fungi were recovered from 94% of all the water samples taken inside the hospital with a mean colony forming unit (cfu) count of 2.7/500mL of water. Aspergillus fumigatus was recovered from 49% and 5.6% of water samples from the taps and showers, respectively (mean 1.9 and 1.0cfu/500mL). More than one third (38.8%) of water samples from the main pipe revealed A. fumigatus (mean 2.1cfu/500mL). All water samples taken at the intake reservoir were culture positive for filamentous fungi, 85% of the water samples showed A. fumigatus (mean 3.1cfu/500mL). Twenty-five percent of water-related surfaces yielded filamentous fungi, but A. fumigatus was recovered from only two samples. We showed that filamentous fungi are present in the hospital water and to a lesser extent on water-related surfaces. The recovery of filamentous fungi in water samples taken at the intake reservoir suggests that the source of contamination is located outside the hospital.