The lowest acute exposure guideline level (AEGL) values are based on a 45-minute (min) no-observed-adverse-effect level (NOAEL) of 1.8 parts per million (ppm) in exercising adult asthma patients (Stevens et al. 1992). No uncertainty factors (UFs) were applied for inter- or intraspecies variability because the study population consisted of sensitive humans. The same 1.8-ppm value was applied across the 10- and 30-min and 1-, 4-, and 8-hour (h) exposure times, because mild irritance generally does not vary greatly over time, and because it is not expected that prolonged exposure will result in an enhanced effect.
The AEGL-2 for the 30-min and 1-, 4-, and 8-h time points was based on severe nasal or pulmonary histopathology in rats exposed at 1,300 ppm for 30 min (Stavert et al. 1991). A modifying factor (MF) of 3 was applied to account for the relatively sparse database describing effects defined by AEGL 2. The AEGL-2 values were further adjusted by a total UF of 10–3 for intraspecies variability, supported by the steep concentration- response curve, which implies little individual variability; and 3 for interspecies variability. Using the default value of 10 for interspecies variability would bring the total adjustment to 100 (total UF×MF) instead of 30. That would generate AEGL-2 values that are not supported by the total data set, including data on exercising asthmatic subjects, an especially sensitive subpopulation, because exercise increases HCl uptake and exacerbates irritation; no effects were noted in exercising young adult asthmatic subjects exposed to HCl at 1.8 ppm for 45 min (Stevens et al. 1992). A total UF of 10, accompanied by the MF of 3, is most consistent with the total database (see Section 6.3 for detailed support of uncertainty factors). Thus, the total factor is 30. Time-scaling for the 1-h AEGL exposure period was accomplished using the Cn×t=k relationship (C=concentration, t=time, and k is a constant), where n=1 based on regression analysis of combined rat and mouse LC50 data (concentrations lethal to 50% of subjects) (1 min to 100 min) as reported by ten Berge et al. (1986). The 4- and 8-h AEGL-2 values were derived by applying an MF of 2 to the 1-h AEGL-2 value, because time-scaling would yield a 4-h AEGL-2 of 5.4 ppm and an 8-h AEGL-2 of 2.7 ppm, close to the 1.8 ppm tolerated by exercising asthmatic subjects without adverse health effects. The 10-min AEGL 2 was derived by dividing the mouse RD50 (concentration expected to cause a 50% decrease in respiratory rate) of 309 ppm by a factor of 3 to obtain a concentration causing irritation (Barrow 1977). It has been determined that human response to sensory irritants can be predicted on the basis of the mouse RD50. For example, Schaper (1993) has validated the correlation of 0.03×RD50=