FUNGAL PROPAGULES IN-HOUSE DUST .1. COMPARISON OF ANALYTIC METHODS AND THEIR VALUE AS ESTIMATORS OF POTENTIAL EXPOSURE

The presence of viable mold propagules in house dust was investigated by 10 different analytic methods, in order to determine to what extent different results are obtained when different analytic methods are us ed. Moreover, the Value of this measurement as an estimator of the pot ential exposure to fungi in epidemiologic studies was assessed. Floor and mattress dust was sampled in 60 homes in The Netherlands during au tumn 1990. For investigation of the variability in time, sampling was repeated in 20 homes after 6 weeks. Each analytic method is characteri zed by a unique combination of culture medium, suspension medium, and dilution step. The highest mean number of colony-forming units (CFU)/g dust was obtained by suspension of at least 100 mg dust in a peptone or sucrose solution in a ratio of 1:50 (w/w), followed by 10-fold dilu tion and plating on DG18 agar (geometric mean (GM) approximately 60 00 0 CFU/g dust). The lowest mean number of CFU/g dust was obtained by di rect plating of 30 mg dust on V8 agar (GM approximately 5300 CFU/g dus t). The mean coefficient of variation of duplicate analyses varied fro m 11%, for suspension in sucrose and plating on DG18 agar, to 27%, for suspension and dilution in sucrose in combination with V8 agar. The h ighest mean number of species isolated was obtained by direct plating of 30 mg dust on DG18 agar (mean number of species: 17). Suspension an d dilution on DG18 or V8 agars yielded an average of approximately six species. In duplicate analyses, the mean percentage of agreement for the species isolated varied from approximately 35%, for suspension and dilution, to 60%, for direct plating. The reproducibility of the numb er of CFU/g dust in time was better for mattress dust than for floor d ust; however, also for mattress dust, the predictive value of a single measurement was rather low. The variability in time in species isolat ed was substantial, both for mattress dust and floor dust. We conclude d that results of measurements of viable mold propagules in house dust , both quantitatively and qualitatively, depend greatly on the analyti c methods used. Furthermore, a single measurement of fungal propagules in settled house dust does not provide a reliable measure of potentia l exposure to fungi in indoor environments.