FIELD-EVALUATION OF ENDOTOXIN AIR SAMPLING ASSAY-METHODS

This study tested the importance of filter media, extraction and assay protocol, and bioaerosol source on the determination of endotoxin und er field conditions in swine and poultry confinement buildings. Multip le simultaneous air samples were collected using glass fiber (GF) and polycarbonate (PC) filters, and these were assayed using two methods i n two separate laboratories: an endpoint chromogenic Limulus amebocyte lysate (LAL) assay (QCL) performed in water and a kinetic chromogenic LAL assay (KQCL) performed in buffer with resistant-parallel line est imation analysis (KLARE). In addition, two aqueous filter extraction m ethods were compared in the QCL assay: 120 min extraction at 22 degree s C with vigorous shaking and 30 min extraction at 68 degrees C with g entle rocking. These extraction methods yielded endotoxin activities t hat were not significantly different and were very highly correlated. Reproducibility of endotoxin determinations from duplicate air samplin g filters was very high (Cronbach alpha all > 0.94). When analyzed by the QCL method GF filters yielded significantly higher endotoxin activ ity than PC filters. QCL and KLARE methods gave similar estimates for endotoxin activity from PC filters; however, GF filters analyzed by th e QCL method yielded significantly higher endotoxin activity estimates , suggesting enhancement of the QCL assay or inhibition of the KLARE a ssay with GF filters. Correlation between QCL-GF and QCL-PC was high ( r = 0.98) while that between KLARE-GF and KLARE-PC was moderate (r = 0 .68). Analysis of variance demonstrated that assay methodology, filter -type, barn-type, and interactions between assay and filter-type and b etween assay and barn-type were important factors influencing endotoxi n exposure assessment.