AIRBORNE ENVIRONMENTAL ENDOTOXIN - A CROSS-VALIDATION OF SAMPLING ANDANALYSIS TECHNIQUES

A standard method for measurement of airborne environmental endotoxin was developed and field tested in a fiberglass insulation-manufacturin g facility. This method involved sampling with a capillary-pore membra ne filter, extraction in buffer using a sonication bath, and analysis by the kinetic Limulus assay with resistant-parallel-line estimation ( KLARE). Cross-validation of the extraction and assay method was perfor med by comparison with methanolysis of samples followed by 3-hydroxy f atty acid (3-OHFA) analysis by gas chromatography-mass spectrometry. D irect methanolysis of filter samples and methanolysis of buffer extrac ts of the filters yielded similar 3-OHFA content (P = 0.72); the avera ge difference was 2.1%. Analysis of buffer extracts for endotoxin cont ent by the KLARE method and by gas chromatography-mass spectrometry fo r 3-OHFA content produced similar results (P = 0.23); the average diff erence was 0.88%. The source of endotoxin was gram-negative bacteria g rowing in recycled washwater used to clean the insulation-manufacturin g equipment. The endotoxin and bacteria become airborne during spray c leaning operations. The types of 3-OHFAs in bacteria cultured from the washwater, present in the washwater and in the air, were similar. Vir tually all of the bacteria cultured from air and water, were gram nega tive composed mostly of two species, Deleya aesta and Acinetobacter jo hnsonii. Airborne countable bacteria correlated well with endotoxin (r (2) = 0.64). Replicate sampling showed that results with the standard sampling, extraction, and Limulus assay by the KLARE method were highl y reproducible (95% confidence interval for endotoxin measurement +/- 0.28 log(10)). These results demonstrate the accuracy, precision, and sensitivity of the standard procedure proposed for airborne environmen tal endotoxin.