The subcommittee concludes that the document is well written. The major comments include using a physiologically based pharmacokinetic (PBPK) model for AEGL-2 and AEGL-3, determining whether CNS effects (AEGL 2) and delayed liver toxicity (AEGL-3) are the most relevant end points or whether CNS be used for both end points, and being consistent with the use of uncertainty factors (UFs).
The subcommittee noted that the ambient air EPA’s proposed value of 0.061 µg/m3 is based on cancer. Therefore, the proposed AEGL values do not protect against excess calculated cancer risks. Cancer risk was not discussed adequately. Some consideration on how to address this public health concern is warranted. Reword sentences on page 58, lines 1966-1967.
There are inconsistencies regarding the use of UFs:
Page 29, lines 1218-1219. In the section titled Interspecies Variability, revise to read “Taken together the interspecies variability for acute lethal effects is limited and an interspecies uncertainty factor of 3 is considered adequate.”
Page 31, line 1236. In the section titled Intraspecies Variability, revise to read “Due to the lack of data, there was no basis for reducing the default intraspecies uncertainty factor.”
Page 32, lines 1287-1288. “A total uncertainty factor of 3 was used because for local effects, the toxicokinetic differences do not vary considerably within and between species.” The NAC needs to break this down to inter- and intra-, not combine them.
Page 34, lines 1359-1361. “An interspecies uncertainty factor of 1 was applied because metabolism in humans and rats is very similar, involving the same metabolic steps and intermediate metabolites (See Section 4.3.2)…” First, this should probably be Section 4.3.1, Pharmacokinetic Modelling. Second, the subcommittee did not see any documentation that the metabolic steps are similar. It is not that the subcommittee doubts it, only that there is no evidence presented in the toxicity summary document (TSD). There is a statement to that fact on page 29, lines 1216-1217, but Section 4.1 does not give data to support that statement.
The document needs to be consistent and in agreement with the Standard Operating Procedure (SOP) manual on the use of uncertainty factors (UFs).
The proposed AEGL-1 value is appropriate for contact irritation. A short discussion is needed for the selection of the lowest exposure to humans that caused eye irritation. The idea is that the AEGL is a discomfort number; thus, the question is how high can the concentration go for irritation before other effects become important? Would 200 or 300 ppm work (Silverman et al. 1946), since this is a reported 8 h value based on a 15 min exposure? The subcommittee understands that 5,500 ppm is too high because