Airborne endotoxin concentrations in homes burning biomass fuel

Abstract

Background: About half of the world's population is exposed to smoke from burning biomass fuels at home. The high airborne particulate levels in these homes and the health burden of exposure to this smoke are well described. Burning unprocessed biological material such as wood and dried animal dung may also produce high indoor endotoxin concentrations.

Objective: In this study we measured airborne endotoxin levels in homes burning different biomass fuels.

Methods: Air sampling was carried out in homes burning wood or dried animal dung in Nepal (n = 31) and wood, charcoal, or crop residues in Malawi (n = 38). Filters were analyzed for endotoxin content expressed as airborne endotoxin concentration and endotoxin per mass of airborne particulate.

Results: Airborne endotoxin concentrations were high. Averaged over 24 hr in Malawian homes, median concentrations of total inhalable endotoxin were 24 endotoxin units (EU)/m(3) in charcoal-burning homes and 40 EU/m(3) in wood-burning homes. Short cooking-time samples collected in Nepal produced median values of 43 EU/m(3) in wood-burning homes and 365 EU/m(3) in dung-burning homes, suggesting increasing endotoxin levels with decreasing energy levels in unprocessed solid fuels.

Conclusions: Airborne endotoxin concentrations in homes burning biomass fuels are orders of magnitude higher than those found in homes in developed countries where endotoxin exposure has been linked to respiratory illness in children. There is a need for work to identify the determinants of these high concentrations, interventions to reduce exposure, and health studies to examine the effects of these sustained, near-occupational levels of exposure experienced from early life.

Figure 1

Box plot of airborne total inhalable endotoxin concentrations by fuel type during cooking in Nepalese homes. The line inside the box represents the median value, the lower and upper box lines represent the limits of the interquartile range (25th and 75th percentiles), and the “whiskers” represent the 5th and 95th percentiles of the distribution. Difference in means p < 0.01.

Figure 2

Box plot of airborne endotoxin by fuel type during cooking per PM mass on the filter in Nepalese homes. The line inside the box represents the median value, the lower and upper box lines represent the limits of the interquartile range (25th and 75th percentiles), and the “whiskers” represent the 5th and 95th percentiles of the distribution. Circles indicate outlier observations with values 1.5–3.0 times the interquartile range from the 25th or 75th percentile. Difference in means p = 0.024.

Figure 3

Box plot of 24-hr airborne respirable endotoxin concentrations by fuel type in Malawian homes. The line inside the box represents the median value, the lower and upper box lines represent the limits of the interquartile range (25th and 75th percentiles), and the “whiskers” represent the 5th and 95th percentiles of the distribution. The circle indicates an outlier observation as described in Figure 2; the asterisk indicates an observation more than three times the interquartile range from the 25th or 75th percentile. Difference in means p = 0.647.

Figure 4

Box plot of 24-hr airborne respirable endotoxin by fuel type per PM mass on the filter in Malawian homes. The line inside the box represents the median value, the lower and upper box lines represent the limits of the interquartile range (25th and 75th percentiles), and the “whiskers” represent the 5th and 95th percentiles of the distribution. The circle indicates an outlier observation as described in Figure 2; the asterisk indicates an observation more than three times the interquartile range from the 25th or 75th percentile. Difference in means p = 0.808.