Abstract:
This study compared three methods of breath sampling for exposure to volatile organic chemicals. Eleven healthy volunteers were exposed to four of these five chemicals: ethanol, acetone, methyl isobutyl ketone (MIBK), toluene, and halothane. After exposure, venous blood and mixed-exhaled breath, end-exhaled breath, and rebreathed gas were sampled ten times in four hours. The relationship between the concentrations of these chemicals in blood and each type of breath were calculated by linear regression. The calculated blood/gas partition coefficients and the goodness of fit for each breath sampling method were compared in order to evaluate which type of breath sample could best be used to predict blood concentrations. The effect of carbon dioxide correction and the effect of lung function on this relationship were investigated. Rebreathed gas concentrations predicted blood concentrations as well as end-exhaled concentrations and in most cases better than mixed-exhaled concentrations with or without carbon dioxide correction. Rebreathing gave the lowest calculated blood/gas partition coefficients, which were also closest to the literature values in vitro. Carbon dioxide correction lowered partition coefficients nearer to the end-exhaled and rebreathed coefficients for all but the most soluble chemical, but the coefficients did not match coefficients from the rebreathed and end-exhaled methods. Carbon dioxide correction did not generally improve the fit between blood and breath concentrations. The ratio between blood and breath concentrations is significantly lower for the rebreathed than for the mixed-exhaled methods, and the difference grows clearer as water solubility decreases. The ratio between blood and breath concentration is lower for the rebreathed than for the end-exhaled methods, which is more likely to be a significant difference between rebreathed and end-exhaled methods as water solubility decreases.