Evaluation of methods for enumerating microorganisms in filter samples from highly contaminated occupational environments

W Eduard,J Lacey,K Karlsson,U Palmgren,G Ström,G Blomquist

DOI: https://doi.org/10.1080/15298669091369899

Abstract:Scanning electron microscopy (SEM), light microscopy (LM), epifluorescence microscopy (FM), and culture were used to assess catches of microorganisms in parallel air samples on membrane filters from heavily contaminated working environments that differed in the relative abundance of bacteria, actinomycetes, and fungal spores. Except in pig houses, estimates by SEM and LM were similar, but those by FM and culture were smaller. However, in pig houses, the fluorescent stain enabled bacteria on skin scales, not seen by SEM or LM, to be counted. Although counts obtained by culturing were always smaller than those obtained by SEM or LM, they sometimes exceeded those obtained by FM. Counts suggested that 0.1-68% of bacteria + actinomycetes and 3-98% of fungal spores were viable. However, samples for culturing may have contained larger aggregates than parallel samples collected within a sampling apparatus. All spore types recognized by LM included aggregates--those of bacteria + actinomycetes sometimes exceeding 200 units, while Wallemia sebi spore aggregates were never larger than 3 spores. The size distributions of all types approximated to log-normal, although single spores and small aggregates of bacteria + actinomycetes were perhaps underrepresented. When spores were counted directly on the filter surface, as by SEM and LM, allowance was necessary for heavier deposition of particles near the center of filters by distributing counting fields systematically over the whole filter or a sector of it. Deposition was more uniform in graphite-filled polypropylene filter holders used open-faced. Losses within filter holders and during transportation from sampling site to laboratory were small. The precision of counting spore-containing particles by LM and SEM was better than that of counting individual spores. No such difference was found for FM because many large spore-containing particles were dispersed during preparation.