Comparison of air sampling methods for aerosolized spores of B. anthracis Sterne

Bacillus anthracis Sterne spores were aerosolized within a chamber at concentrations ranging from 1 x 10³ to 1.7 x 10⁴ spores per cubic meter of air (particles (p)/m³) to compare three different sampling methods: Andersen samplers, gelatin filters, and polytetrafluoroethylene (PTFE) membrane filters. Three samples of each type were collected during each of 19 chamber runs. Chamber concentration was determined by an aerodynamic particle sizer (APS) for the size range of 1.114-1.596 μm. Runs were categorized (low, medium, and high) based on tertiles of the APS estimated air concentrations. Measured air concentrations and recovery efficiency [ratio of the measured (colony forming units (CFU)/m³) to the APS estimated (particles/m³) air concentrations] for the sampling methods were compared using mixed-effects regression models. Limits of detection for each method were estimated based on estimated recovery efficiencies. Mean APS estimated air concentrations were 1600 particles/m³, 4100 particles/m³, and 9100 particles/m³ at the low, medium, and high tertiles, respectively; coefficient of variation (CV) ranged from 25 to 40%. Statistically significant differences were not observed among the three sampling methods. At the high and medium tertiles, estimated correlations of measured air concentration (CFU/m³) among samples collected from the same run of the same type were high (0.73 to 0.93). Among samples collected from the same run but of different types, correlations were moderate to high (0.45 to 0.85); however, correlations were somewhat lower at the low tertile (-0.31 to 0.75). Estimated mean recovery efficiencies ranged from 0.22 to 0.25 CFU/particle with total CVs of approximately 84 to 97%. Estimated detection limits ranged from 35 to 39 particles/m³. These results will enable investigators to conduct environmental sampling, quantify contamination levels, and conduct risk assessments of B. anthracis.