Airborne Endotoxin from Indoor and Outdoor Environments: Effect of Sample Dilution on the Kinetic Limulus Amebocyte Lysate (LAL) Assay

Abstract

Airborne endotoxins in occupational environments are a potential respiratory hazard to individuals. In this study, airborne endotoxins were collected using open-face and button aerosol samplers from inside animal housing units and downwind from agricultural production sites and a wastewater treatment plant. Filter extracts were then diluted to examine the effect of interfering substances on the kinetic Limulus amebocyte lysate (LAL) assay. In most cases, the overall endotoxin concentration was shown to decrease with increasing dilution up to 1000-fold, suggesting the presence of enhancing substances in the filter extracts. This dilution-dependent effect was most prominent in the open-face endotoxin samples, while button samples displayed little effect. Using a joinpoint regression model, it was determined that a dilution factor of 50 to 100 was generally sufficient to eliminate the presence of enhancing substances. After screening the data for dilution dependent effects, the airborne endotoxin concentrations were determined. The highest endotoxin concentrations, ranging from 2841 to 49,066 endotoxin units m⁻³ of air, were found inside swine farrowing and finishing barns. Airborne endotoxin concentrations were 10- to 100-fold lower inside a dairy barn and downwind of other agricultural production sites and the wastewater treatment plant. Examination of dilution-dependent effects should be considered essential when utilizing the LAL assay, especially if values are to be used for regulatory purposes.

Keywords:

• agriculture
• airborne
• dairy
• endotoxin
• Limulus amebocyte lysate assay
• swine

Notes

^A All dilution curve data was fit to a three-parameter single exponential decay curve, except the swine finishing barn sample, which was fit to a three-parameter single exponential growth curve.

^B Dilution factor at which the slope of two regression lines converge.

^C Too few data points available to run joinpoint model.

^A Average endotoxin concentration ± standard error of the mean.

^B Statistical differences determined using a two-sample t-test. P-values < 0.05 indicate a significant difference between the open-face and button endotoxin concentrations for that row only.