Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 1
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels BACKGROUND In 1991, the U.S. Environmental Protection Agency (EPA) and the Agency for Toxic Substances and Disease Registry (ATSDR) asked the National Research Council (NRC) to provide technical guidance for establishing community emergency exposure levels (CEELs) for extremely hazardous substances (EHSs) pursuant to the Superfund Amendments and Reauthorization Act of 1986. In response to that request, a subcommittee of the NRC Committee on Toxicology (COT) prepared a report titled Guidelines for Developing Community Emergency Exposure Levels for Hazardous Substances (NRC 1993). That report provides step-by-step guidance for the derivation of CEELs for EHSs. In 1995, EPA, several other federal and state agencies, and several private organizations convened an advisory committee—the National Advisory Committee on Acute Exposure Guideline Levels (AEGLs) for Hazardous Substances (referred to as the NAC)—to develop, review, and approve AEGLs (similar to CEELs) for up to 400 EHSs. AEGLs developed by the NAC have a broad array of potential applications for federal, state, and local governments and for the private sector. AEGLs are needed for prevention and emergency response planning for potential releases of EHSs either unintentionally from accidents or as a result of terrorist activities. THE CHARGE TO THE SUBCOMMITTEE The NRC convened the Subcommittee on Acute Exposure Guideline Levels to review the AEGL documents approved by the NAC. The subcommittee members were selected for their expertise in toxicology, pharmacology, medicine, industrial hygiene, biostatistics, risk assessment, and risk communication. The charge to the subcommittee is to (1) review AEGLs developed by the NAC for scientific validity, completeness, and conformance to the NRC (1993) guidelines report, (2) identify priorities for research to fill data gaps, and (3) identify guidance issues that may require modification or further development based on the toxicological database for the chemicals reviewed. This interim report presents the subcommittee’s comments concerning the NAC’s draft AEGL documents for 17 chemicals: hydrazine, ethylene oxide, propylene oxide, iron pentacarbonyl, nickel carbonyl, phosphine, boron trichloride, chlorine trifluoride, dichlorodimethyl silane, trichloromethyl silane, ethyleneimine, propylenimine, allyl alcohol, aniline, arsine, monomethylhydrazine, and dimethylhydrazine.
OCR for page 2
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels COMMENTS ON HYDRAZINE At its July 21–23, 2003 meeting, the subcommittee reviewed the AEGL document on hydrazine. The presentation was made by Robert Young of Oak Ridge National Laboratory. The subcommittee recommends a number of revisions. A revised draft should be reviewed by the subcommittee at its next meeting. Major Issues The uncertainty factors (UFs) are not applied consistently. For irritants and direct acting chemicals (and the case is made for irritation as the primary effect for AEGL-1 and AEGL-2 values), interspecies and intraspecies UFs generally applied were 3 each. This was done for the hydrazine AEGL-1. However, for AEGL-2 values, the statement is made, “An uncertainty factor of 10 for interspecies variability was applied to account for the high degree of variability in the data due to the extreme reactivity of hydrazine that compromised exposure concentration measurements.” There is no explanation of how the deposition of hydrazine on chamber surfaces and difficulties in concentration measurements relate to interspecies variability, especially when hydrazine appears to be direct acting agent. The study by Latendresse et al. (1995) does not appear to suffer from compromised exposure concentration measurements. If the NAC believes the overall uncertainty factor of 30 is appropriate, the subcommittee recommends using inter- and intraspecies UFs of 3 and 3, with a possible modifying factor of 10 for the compromised exposure concentration measurements, which will make the magnitude of all factors to be 100. However, that adjustment cannot apply to the Latendresse study. The subcommittee recommends using a smaller UF based upon the quality of the study and a possible modifying factor to account for the small database and numbers of animals. The section on level of distinct odor awareness (LOA) is misplaced in the Executive Summary after AEGL-1 and before the AEGL-2 values. The 10-min AEGL-3 is proposed by the NAC to be 63 ppm. How does the NAC explain this level compared with the odor threshold level of 3–4 ppm cited on page 30? From an emergency response perspective, the odor threshold is very important because this is the level at which people will start detecting exposure to hydrazine. A general statement regarding exposure concentrations should be included. In addition to being highly reactive, the subcommittee understands that hydrazine also adsorbs to most materials, including Teflon. Therefore, the subcommittee recommends that the NAC focus on the more recent studies that have presumably solved the technical problems with concentration measurements and use other studies as supporting information. In Section 2.2.2, the Morgenstern and Ritz study (it is only one) seems to have been added without considering what influence those data have on the rest of this section. For example, in the first paragraph, reference is made to a more recent study (1987) when the Morgenstern study was reported in 2001. Does the Morgenstern study follow up or include individuals from the previous studies? How are these studies linked, if at all? Delete the last paragraph. With the Morgenstern (2001) study, the information presented in the last paragraph does not seem to apply.
OCR for page 3
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels Section 2.5. See comment above. Are there one or more studies and how do they apply here? Section 4.4.1. As written, the text is unclear as to whether species variability or exposure concentration is the issue. The authors seem to be equally confused. Section 5.2. If hydrazine is a direct-acting toxicant and irritation is the primary adverse effect, then why is hydrazine discussed as a cumulative exposure? Why did the NAC use cumulative exposure and derive AEGL-1 values that are constant across time? At the end of this paragraph, add the human exposure level (not stated other than below the TLV of 0.1 at the time). Section 6.3. If the effect is irritation or a direct-acting effect, why not use a UF of 3 rather than 10? Derivation of an AEGL for irritants should be consistent, as has been the case for other chemicals. The use of the modifying factor appears inconsistent with the SOP, as the UF should be 10 with a modifying factor of 2, only because of the lack of data and the small study size of Latendresse study. Section 7.1. No mention has been made of human exposure levels. Section 7.3. The statement is made that the lethal effects of hydrazine appear to be more dependent on concentration than duration; therefore, exponential scaling was used. If the adverse health effects are more dependent on concentration than duration of exposure, then there is no basis for exponential scaling. Section 7.3, 5th paragraph. The subcommittee does not believe that the highly reactive nature of hydrazine that results in a compromised exposure-concentration measurement is a proper rationale for using an uncertainty factor of 10. This seems to be better addressed with a modifying factor. Section 8.1, 5th paragraph. Why was the HRC data set not included in the benchmark dose evaluation? Pages 10, 11, and 34. Rewrite the description of the paper by Sotaniemi et al. (1971). If the concentrations derived from the simulation of the fatal episode were likely to be 0.05 ppm as estimated, then the AEGL derivations could be incorrect. Since this study is not relied upon for the AEGL calculations, the reasons need to be fully stated. The actual dose received from a once-a-week, 6-month exposure is not known. This is a major variable and this should be acknowledged. Can dermal exposure in this case report be ruled out, or is it a possible route of systemic uptake? Pages 22–23. Delete tables and summarize the parenteral studies in a sentence or two at Section 3.3.1. Note the inability to scale the intraperitoneal doses to inhalation exposures with the information at hand. Page 32. Support the AEGL-1 value by brief discussion of the Koizumi results for the 12 workers exposed repeatedly at up to 0.12 ppm as these 8-hr TWA values were confirmed by personal sampling.
OCR for page 4
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels Page A-1. Add the calculation for a 10-min AEGL-1, as it is missing. Minor Issues In the AEGL derivations, the subcommittee recommends using ppm-hr and the SOP manual recommended significant figure issues. Use line numbering. Section 5.3, 2nd paragraph. Does the discussion on geometric mean contain the proper terminology and rationale? As 3.16 is the square root of 10, which makes it the geometric mean—but is that the reason a UF of 3.16 is used here? This is the first time it has been discussed and this is commended, but the rationale must be clear and understandable. Editorial Comments Add “planners and” before “responders” in all the LOA discussions. There are a number of blank pages. Section 4.4.3, 2nd line add “be” after “may”. Section 6.2, 1st paragraph, end of penultimate line change “question” to “questionable” and add “use”. Table 15. Spell out STPL. There are format problems with the carcinogenicity assessment section. Provide separate animal and human category plots. It is recommended that the NAC combine sections on odor threshold and LOA. These data are now separated and that format is confusing. Increase font size in Section B-1 for the equations. COMMENTS ON ETHYLENE OXIDE At its July 21–23, 2003 meeting, the subcommittee reviewed the revised AEGL document on ethylene oxide. The document was presented by Kowetha Davidson of Oak Ridge National Laboratory. The subcommittee recommends a number of revisions. The document can be finalized if the subcommittee’s recommended revisions are appropriately made.
OCR for page 5
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels Major Issues In general, the document should have a discussion of the analytical procedures, sterilizer type, and sterilant used. Historically, ethylene oxide was mixed with halons when it was used in hospital sterilizers. Some of the sterilizers also used steam to supplement the process. Another issue is the use of water aspiration to draw a vacuum. These factors are important because a common method of monitoring EtO in workplace air is infrared absorption. If halon was also used, the best wavelength for analysis of EtO is also the region for IR absorption of halon. The common solution is to move to an analytical wavelength where water is an interferent; this then becomes an analytical problem with water aspiration, steam, and any other source of water, but this is generally not a problem in dental laboratories where pure EtO (no halon) is dispensed from a disposable cartridge. A portable GC or FTIR eliminates these analytical problems. The subcommittee questions the statement that reported EtO concentrations measured at the floor were significantly higher than those measured closer to the source. While pure EtO is more dense than air, at concentrations that are in the range of parts per million, there is very little difference in density, and normal room air eddy currents mix the EtO; that mixing precludes any concentration gradients. Summary, 6th paragraph. The subcommittee understands the discussion that the mechanism of action is the same; however, wide range exists in lethal concentration of 660 ppm for death in female mice to 13,349 in rabbits with no effect. These observations seem to argue for a greater interspecies uncertainty factor or a more comprehensive discussion of the factors which explain these marked differences. Section 4.1. Expand the PBPK discussion. The subcommittee believes that the NAC used those data as rationale for stating later that humans are less susceptible to the adverse health consequences of inhaled EtO, as compared with laboratory rodents. Section 6.3, 2nd paragraph. Is a disturbance in gluthione-S-transferase activity associated with EtO developmental toxicity? If not, the justification for the intraspecies UF of 3 does not hold if the AEGL-2 is based upon the developmental toxicity study of Snellings. The subcommittee understands that the NAC followed the SOP manual for assigning the 10-min AEGL-2 the same value as the 30-min AEGL-2. However, the text is not clear in that it is acceptable to go to a 30-min or 1-hr AEGL from a 10-min exposure and from 8-hr to 30-min but not from 4-hr to 30-min? At what point in the SOP manual is this practice addressed? Section 7.3. Why was the Jacobson study used here and not the Nachreiner study? Section 7.2 points out the latter is a good study and that Jacobson used an unspecified rat strain. Section 7.3, 2nd paragraph. Use the PBPK discussion to support and explain the interspecies uncertainty factor. Section 8.2. The discussion of ERPG-3 requires refinement, as these are AEGL-2 effects. What study did the ERPG Committee use that the NAC did not or discounted? Figure 1. This is not a convincing figure, as it is essentially a straight line through two points.
OCR for page 6
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels The longitudinal study of approximately 600 or more individuals exposed to EtO years ago in Buffalo, N.Y., should be examined. The study is well-known, and many papers have appeared on various aspects of the study. Special attention was given to the incidence of cancer in this population and the value of SCE examinations as a marker for DNA damage and potential carcinogenesis. These data help to determine the relevance of animal data to human experience and the possibility that for EtO, the rodent may be an inappropriate model. Data are available on the off-gassing of EtO from plastic bags used to hold intravenous fluids. AEGL-1: It is acceptable that AEGL-1 is not set in this case. However, the wording for justification should be revised to that it is consistent with what is stated elsewhere in AEGL documentation for other chemicals in similar situations. AEGL-2: Why is a repeated exposure study used to set AEG1–2? Is not there an Oakridge study investigating effects following a single day exposure to EtO? If acute studies of relevance are nonexistent, this should be made clear. Further, only the key study should be presented and used to derive AEGL values. The other studies can be used to provide support or corroborate the derived AEGL values. The justification for an intraspecies uncertainty factor based on asthmatic people not responding differently (from the rest of population—to irritant effects) is unclear and should be improved. Section 6.3, 2nd paragraph. Is gluthione-S-transferase associated with developmental toxicity? If not, the explanation for using a UF of 3 for intraspecies extrapolation does not hold if the AEGL-2 is based upon developmental toxicity (Snellings study). AEGL-3: Even though the LC50/LC01 values are lower for the mice and dogs, the data from rats are used to derive AEGL-3. Why? Further, a UF of 3 is used on the basis that the rat is not the most sensitive species, which indicates the use of a UF of 3 would protect the known sensitive species such as mice and dogs. The presentation should be improved to clarify why the data from known sensitive species were not used and what the basis of the magnitude of UF was? Other Comments Measurement of adducts should be considered a parameter for exposure rather than the mechanism of toxicity: Formation of protein adducts might even be considered a way of detoxification: reactive intermediates are bound to proteins before they affect genetic material (cf. HbAiC: glucose adduct formation to hemoglobin in diabetics; ethyl adducts in alcohol consumers). It is remarkable that lethal accidents with EtO have never been described in the clinical toxicology literature, in spite of the widespread use of the substance. Apparently its acute toxicity is not very high. This should be mentioned in this document, especially because the NAC based acute exposure guidelines on long-term/chronic effects.
OCR for page 7
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels Exposure to EtO may be much more widespread than generally assumed. EtO sterilized food and medical appliances act as slow-release preparations. Are there any data on exposure levels from these sources? Section 4.1. Expand the PBPK discussion. This PBPK information appears to be the rationale for stating later that humans are less susceptible than animals. Section 7.3. Why was Jacobson used and not Nachreiner? Section 7.2 points out this is a good study and that Jacobson used an unspecified strain of rat. Section 7.3. Second paragraph, bring in the PBPK discussion in explaining the interspecies factor. Section 8.2. The discussion of ERPG-3 needs refinement. These are AEGL-2 effects. What study did they use that the NAC did not use or discounted? Figure 1. This is not a very convincing figure. It is essentially a straight line through two points. There is some confusion of units (regarding unit risk and slope factor). Minor issues Derivation of values—too many significant figures. The subcommittee commends the use of “(uncertainty factor)”. This helps the lay reader understand how the AEGL values were addressed. Summary. The odor threshold discussion does not belong in the first paragraph. Section 8.1, 5th line. Add “on” after “based”. COMMENTS ON PROPYLENE OXIDE At its July 21–23, 2003 meeting, the subcommittee reviewed the AEGL document on propylene oxide (PO). The document was presented by Claudia Troxel of Oak Ridge National Laboratory. The document can be finalized after the subcommittee’s recommended revisions have been made appropriately. General Comment Human data should be used for the derivation of AEGLs 1 and 2 values, but the animal data should be used for the derivation of AEGL 3 values with human data to support the value obtained. Furthermore, it appears that the same end point is being used for AEGLs 1 and 2. In fact, the effect used in support of the latter is really an effect that is more appropriate for the former. An explanation of this rationale is needed here.
OCR for page 8
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels Specific Comments Page 1, line 15. Change tense to “produced” from “will be produced.” Table 20–22. These tables should be deleted. They are likely to confuse the reader and are unnecessary. Page 28, lines 15–18. It should be pointed out here that PO reacts with glutathione (GSH) enzymatically (via GSH S-transferase) and non-enzymatically as well. It should also be pointed out that these reactions represent detoxification pathways, as PO is a direct alkylating agent. Page 29, lines 20–39. Have the toxicokinetic data described here been published in the peer-reviewed literature? Such information is pertinent to the conclusion (lines 1–3 of page 33) that species differences in susceptibility to PO are quite modest. Page 31, lines 2–19. Additional information should be provided on potential mechanism(s) of PO-induced preneoplastic changes and carcinogenicity in the respiratory tract. Rios-Blanco et al. (2003a, b), as described later in this critique, recently reported dose-dependent cellular proliferation and DNA binding in the nasal epithelium of rats inhaling a series of concentrations of PO over periods of 3 and 20 days. These publications resulted from work conducted as part of a robust, ongoing research program focusing on the mechanism of PO carcinogenicity in the rat. Page 33, lines 1–2: The introductory sentence is vague and should be modified to clarify that mice are the most susceptible species to acute lethality, but that mice, rats, dogs and humans may not differ by more than 3.5-fold in sensitivity to systemic effects (effects undefined, as there is little evidence of adverse effects other than those at the portal of entry). Swenberg and Filser (1998) predict a higher steady-state blood level and a longer t1/2 for PO in humans than in mice and rats. Page 34, lines 7–35. The authors of this document should consider using findings reported by CMA (1998a) as the basis for derivation of AEGL-1 values for PO. The eye irritation described by this group is an appropriate end point for AEGL-1. However, time scaling should not be performed for minor/modest mucus membrane irritation associated with PO exposures. Page 37, lines 1–25. As the observations in the CMA (1998a) report are appropriate for AEGL-1, data on more severe irritation, port of entry cytotoxicity, and/or systemic toxicity are needed to derive AEGL-2 values. There is very little relevant information for these end points in the current AEGL document. PO does not produce neurotoxically in monkeys or rats or recognizable histopathological changes. Swenberg and co-workers, however, recently published a well-conducted study that resulted in findings more applicable to AEGL-2. Rios-Blanco et al. (2003a) assessed histological changes and cellular
OCR for page 9
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels proliferation in nasal and hepatic tissues of groups of male F344 rats exposed to a series of concentrations of PO vapor 6-hr daily over intervals of 3 and 20 days. It is noteworthy that no exposure-related changes in cellular proliferation were observed in the liver. The transitional epithelium of the anterior nasal passages showed the most pronounced changes in the target tissue, and it is noteworthy that the nasal tumors described in previous studies occurred in this region. Vapor concentration-dependent increases in cellular proliferation occurred in the same site in the rats that inhaled 300 or 500 ppm for 3 days. No such changes were manifest at 50 ppm. Alternatively, the NOAELs/LOAELs for histopathological changes following inhaled PO could be utilized to calculate AEGL-2 values. The ten Berge (1986) approach to time-scaling may not be appropriate for the portal-of-entry effects seen by Rios-Blanco et al. (2003a). The extent of nasal epithelial proliferation observed was no more serious after 20 days than after 3 days of their exposure regimen. Maples and Dahl (1993) (lines 5–8, page 29) found that circulating PO levels plateaued after the first 10-min of a 1-hr inhalation exposure of rats at 14 ppm. The severity of the aforementioned changes was quite modest, though cytotoxicity and compensatory cellular proliferation may very well be pertinent to the mechanism(s) of carcinogenicity of PO similar to that seen after inhaled formaldehyde, hydrazine, and other highly reactive gases. Accordingly, uncertainty and modifying factors (if required at all) for AEGL-2 derivation should be modest. Rios-Blanco et al. (2003b) reported finding DNA binding, an even more sensitive index of PO exposure, in a second recent publication. DNA binding was measured in the nasal epithelium, lung, and liver of the same groups of rats utilized for the Rios-Blanco et al. (2003a) publication, and N7-(2-hydroxypropyl) guanine formation was greatest in the nasal epithelium. Levels of the DNA adduct in the nasal epithelium after 3 daily 6-hr exposures increased linearly with the inhaled concentration. The binding of DNA at the target tissue at 5 ppm, the lowest concentration studied, should be pointed out. Page 37, lines 32–34. See comments previously (p. 33, lines 1 and 2) made about clarifying anticipated similarities in susceptibility of different species to PO toxicity. Page 39, lines 4–25. A mouse or rat LC50 value should be used as the basis for derivation of AEGL-3. Page 43, lines 22–36. It should be pointed out here that there is a paucity of data relevant to derivation AEGL-1 and -2. This section should be used to describe what data are needed, rather than to reiterate and rationalize the selection of AEGL calculations. Page 50. The key question in this section is whether cancer could result in human beings after a single 10-min to 8-hr inhalation exposure to PO. Sellakumar et al. (1987) demonstrated that repeated PO exposures of male Sprague-Dawley rats over periods of 8 or 30 days failed to elicit any evidence of carcinogenicity. These findings and their cancer risk implications of acute PO exposure should be the first topic addressed in this section.
OCR for page 10
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels It appears that PO may cause cancer by mixed modes of action (i.e., genotoxicity and cellular proliferation would both be required), as is the case for formaldehyde. The current document should address (1) the weight of scientific evidence for this versus other mechanisms; and (2) what type of biologically-based cancer risk assessment would be most appropriate in each instance. This discussion would include the present description of the linearized multistage model, though so much detail about the default EPA (1987) risk assessment paradigm is probably not necessary. Four more references (in addition to Rios-Blanco et al. 2003a & b) are included under Additional References, and those data may be of use in the document. AEGL-1 Values The total set of human data (13.2–15.20 ppm) should be presented to justify the identification of the NOAEL for AEGL-1 derivation. The n value should be 1, rather than 1.2, for irritant effect. The supporting data and corroborative studies should be presented in the text and not in tables along with an alternative set of AEGLs. AEGL-2 Values The derivation of alternative sets of AEGL values need not be done, since this will cause confusion (e.g., Tables 21 vs 22). The data and studies, supportive of the derived AEGL, can simply be presented and referenced in the text. Why was 15.20 ppm not used, even though it was associated with the same effect as 380–525 ppm (i.e., “irritation was not intolerable”)? AEGL-3 Values The end point selected for derivation of the AEGL-3 is the same as that for AEGL-2. More appropriate animal studies could be used to derive the AEGL-3 values, and those can be supported with the human data. In any case, the use of a single, free-standing human number (as presented in the text) is not convincing. Cancer Risk Assessment The potency of PO as a rodent carcinogen needs to be clearly documented. There are large populations of human veterinary pathologists, embalmers, and workers in biology museums, who were exposed to formaldehyde vapors for 30 to 40 years. This was before mandated air quality standards were applied only 30 years ago. Are there any epidemiologic data that demonstrate that formaldehyde is a human carcinogen? Since PO is compared with formaldehyde, what kinds of effects are expected? This section should put primary emphasis on the key short-term studies (8 days and 30 days). These studies should be presented in greater detail. The total cumulative dose associated with the observed outcome may be used to support the AEGL-3 values based on other effects.
OCR for page 11
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels Additional References Rios-Blanco, M.N., Ranasinghe, A., Upton, P., Lee, M.S., Filser, J.G., and Swenberg, J.A. (2002). Exposure-dependent accumulation of N-(2-hydroxypropyl) valine in hemoglobin of F344 rats exposed to propylene oxide by the inhalation route. J. Chromatogr. B 778: 383–396. Rios-Blanco, M.N., Yamaguchi, S., Dhawan-Robl, M., Kessler, W., Schoonhoven, R., Filser, J.G., and Swenberg, J.A. 2003a. Effects of propylene oxide exposure on rat nasal respiratory cell proliferation. Toxicol. Sci. in press. Rios-Blanco, M.N., Ranasinghe, A., Lee, M.S., Faller, T., Filser, J.G., and Swenberg, J.A. 2003b. Molecular dosimetry of N7-(2-hydroxypropyl) guanine in tissues of F344 rats after inhalation exposure to propylene oxide. Carcinogenesis 24:1233–1238. Schettgen, T., Eroding, H.C., Angerer, J., and Drexler, H. (2002). Hemoglobin adducts of ethylene oxide, propylene oxide, acrylonitrile and acrylamide-biomarkers in occupational and environmental medicine. Toxicol Lett. 134:65–70. Segerback, D., Plna, K., Faller, T., Kreuzer, P.E., Hakansson, K., Filser, J.G., and Nilsson, R. (1998). Tissue distribution of DNA adducts in male Fischer rats exposed to 500 ppm of propylene oxide: quantitative analysis of 7-(2-hydroxypropyl) guanine by 32P-postlabeling. Chem.-Biol. Interact. 115:229–246. Thier, R., Wiebel, F.A. and Bolt, H.M. (1999). Differential substrate behaviors of ethylene oxide and propylene oxide towards human glutathione transferase theta hGSTT1–1. Arch. Toxicol. 73:489–492. COMMENTS ON IRON PENTACARBONYL At its July 21–23, 2003 meeting, the subcommittee reviewed the AEGL document on iron pentacarbonyl. The document was presented by Robert Young of Oak Ridge National Laboratory. The subcommittee recommends a number of revisions. The revised document will be reviewed by the subcommittee at its next meeting. Major Issues The NAC must explain how exposures might occur to the public and to what compound. It is stated iron pentacarbonyl, released into air, will either spontaneously combust or undergo decomposition to iron nonacarbonyl and carbon monoxide or burn to ferric oxide. In the event of a release, what is the most likely compound to which the public will be exposed? If it is a decomposition product of the parent material and not iron pentacarbonyl, there is no reason to derive an AEGL for iron pentacarbonyl. This issue point needs to be resolved. How is iron pentacarbonyl produced and used in industrial plants? Is it consumed on site in an inert atmosphere? Is it transported to refineries and blended in gasoline? How is it used in making iron cores? Iron pentacarbonyl may also be used in making computer hard disk memory components, but the document suffers from the fact that the industrial use and handling practices for this material are given little attention.
OCR for page 16
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels of boron continuous filament fibers (whisker crystals) that are produced during thermal decomposition of boron trichloride. Boron filaments are stronger than carbon fibers and are incorporated into polymer and metal matrices used in both the aerospace and sporting goods industries. Boron trichloride is also used as a catalyst in polymerization of styrene and cyclopentadiene in aluminum refining, and it is used along with TiCl4 in zinc vapor to form a titanium-boron alloy applied to the carbon fiber surface to promote carbon diffusion and to promote fiber “wetting”. “Commercial BCl3 is shipped in steel cylinders (0.9–817 kg). The material can explode upon contact with water and at elevated temperature; it decomposes to produce chlorine gas.” Page 1, 1st paragraph. It would be helpful if the NAC would contact the sole production company (Kerr-McGee Chemical) and obtain details about how this substance is handled (e.g., exclusively closed-systems?), transported, and where and how many facilities use this material in production of aerospace/sporting goods. Additional information on the range of quantities commonly stored on-site or on whether the substance is used immediately upon receipt would be helpful. Page 1, Table 1: The dissociation constant should be added to the table. Page 2, lines 19–24: It should be noted in the study description of Vernot et al. (1977) that the lethality data are not shown in the report (only the LC50 and confidence limits are presented). The methods for exposure generation and measurement should be described. Page 2, line 26: The highest concentration for the two 7-hr daily periods in Stockinger and Spiegel (1953) is 85 ppm (not 80 ppm). It should also be noted that a “wet, sticky layer appeared on the cage floors and animal fur.” Page 3, line 5. The statements as written do not make sense. If all of the rats died after exposure to 20 ppm for 7 hr, then it does not follow that 30% mortality was found after similar exposure to 85 ppm for the same exposure duration (see also in Table 2). Page 3, line 8: Apparently only the rats showed swelling; double check as to whether mice also showed swelling. Pages 4: Why is it necessary to repeat the information presented in Sections 3.2.1, 3.2.2 and 3.2.3 that was already given in Sections 3.1.1, 3.1.2, and 3.1.3, respectively? These studies are not utilized to derive AEGLs. The presentation would be improved substantially by presenting all of the data (given the fact only two studies are available) in one section and then explain the mode of action and rationale for establishment of the AEGLs by analogy. Page 5, lines 2–3. Have other borates been considered by either IARC, NTP, or EPA? Page 5, line 21. Rephrase sentence to show it is compared with other guinea-pig study. Page 7, Section 4.2. This section and table need to provide more explanation. It should be clearly acknowledged that the mechanism of toxicity for BCl3 is not known. Further, what
OCR for page 17
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels is proposed is speculative and does account for the toxicity of boric acid or the interaction of boric acid and HCl. The argument presented here is less than convincing and not supported by the comparison of LC50 values shown in Table 3 (that is meant to be supportive). Table 3 shows LC50 values for boron trichloride that are higher than would be predicted given the mole equivalent to HCl. However, this is based on data from only one study using one species (Vernot et al., 1977). Using the proposed surrogate approach appears to provide conservative AEGLs, but the data are not convincing. This approach assumes that boric acid is not responsible for the observed toxicity or has a negligible contribution relative to HCl; there is no discussion to support or refute this. Page 7, lines 23–24. Delete. One cannot conclude that male animals are “more sensitive” than female animals from the LC50 values listed in Vernot et al. (1977) report. Note that the LC50 values for boron trifluoride are nearly identical for both genders under identical conditions used for the Vernot BCl3 study. Only if a true difference in “sensitivity” of the sexes can be demonstrated for HCl—which is considered by NAC as the active entity—can this argument be considered consistent with the weight-of-evidence for boron trichloride on this point. Page 11, Section 8.2. Given that no TLVs or other values are available for BCl3 but that occupational limits have been developed for other boron halides and HCl and AEGLs have been finalized for HCl, it is worthwhile to insert and compare that information here. It is noteworthy that the TLV ceiling values for BBr3 and BFl3 are identical (1 ppm) and are both based on avoidance of “even transient irritation and complaints.” Routine control of these materials at ~1/3 the TLV ceiling results in values not dissimilar from the AEGL-1 values presented in Table. 7. Pages A-3 to C-3. Delete. Refer the reader to the published AEGL volume of HCl documentation that contains the (AEGLs Volume 4, in preparation). Reduce the text and eliminate the reproduction of the HCl documentation. Condense the BCl3 documentation to address the parent material, its decomposition to HCl and boric acid, the issues surrounding boron, and refer the reader to the derivation of the AEGL values for HCl. Indicate clearly that the AEGL values for BCl3 were developed by analogy to HCl, just as was done by the TLV committee in the case of BBr3 on page 145 of the TLV 6th edition. The NAC should reproduce one copy of the final summary AEGL table for HCl for all time points, since these values were all utilized for BrCl3. COMMENTS ON CHLORINE TRIFLUORIDE At its July 21–23, 2003 meeting, the subcommittee reviewed the AEGL document on chlorine trifluoride. The document was presented by Sylvia Talmage of Oak Ridge National Laboratory. The subcommittee recommends one minor revision. The document can be finalized if the recommended revision is made appropriately.
OCR for page 18
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels Specific Comment Page 19, line 1. This statement is generic and does not apply specifically to this chemical as implied in this document. COMMENTS ON DICHLORODIMETHYL SILANE At its July 21–23, 2003 meeting, the subcommittee reviewed the AEGL document on dichlorodimethyl silane. The document was presented by Cheryl Bast of Oak Ridge National Laboratory. The subcommittee recommends the following revisions. A revised draft should be reviewed by the subcommittee at its next meeting. General Comment The AEGL values for both chlorosilanes should not be finalized until the AEGL document for HCl is finalized. The HCl AEGL document should be added as a reference and cited at several key places where the derivation is dependent. Specific Comments A 2-week nose-only inhalation study where male and female rats were exposed to dimethyldichloro silane at 15 ppm (nominal) concentration is cited in the ERPG document as reference 11 (Dow-Corning 1993). No effects were observed, and this fact should be added to the document. Although this study does not influence derivation of the AEGLs, it serves as a reference point and would support the conclusions presented in the document. AEGL-2, Page C-3. Delete reference to modifying factor for steepness to be consistent. Page 3, line 24 and page 10, lines 8 and 13. Round LC values to 4 significant figures. Page 4, line 9. The death of one of two animals is formated incorrectly as the fraction 1/2. Page 5, lines 33–34. “Carbon monoxide, carbon dioxide, and silicon oxide fumes may also be produced upon combustion or decomposition of dichlorodimethyl silane.” This statement may be true if the word decomposition is removed. However, this statement is not relevant to the discussion of “Mechanisms of Toxicity” as it relates to AEGLs. The subcommittee suggests removing the word “decomposition” and moving this discussion to the physical properties or to the introduction of the document. Related to the above discussion, the NAC should take a closer look at the references in any mention of combustion products. Phosgene may also be among the combustion products possible with the chlorosilanes.
OCR for page 19
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels Clarify the comment on Section 8.3 that “potential unidentified reactive intermediates…may be involved in the toxicity…” If other toxic decomposition products could result, then this reduces the confidence one can place in the AEGL derivations based on analogy to HCl. Page 26, lines 9–13. What are the details of the study? What strain of rats were used and what gross and microscopic pathology was observed? Were the lungs properly prepared for fixation and sectioning? Were any special studies on the tissues performed? AEGL-2. Delete the sentence on page 8, lines 39–40. “The modifying factor of 3 is considered insufficient due to the steepness of the concentration-response curve.” This is not needed as there are two other mentioned reasons; further, the slope of the curve can be argued as a reason for a larger uncertainty factor. The AEGL-2 and AEGL-3 were derived using the HCl data for time scaling, but in each case a different approach was taken with the same data. This is inconsistent. As the HCl document is being finalized, it may be necessary to make these two documents consistent. COMMENTS ON TRICHLOROMETHYL SILANE At its July 21–23, 2003 meeting, the subcommittee reviewed the AEGL document on trichloromethyl silane. The document was presented by Cheryl Bast of Oak Ridge National Laboratory. The subcommittee recommends the following revisions. A revised draft should be reviewed by the subcommittee at its next meeting. General Comment The AEGL for the chlorosilanes should not be finalized until the AEGL document for HCl is finalized. The AEGL document on HCl should be added as a reference and cited at several key places where the AEGL derivation for trichloromethyl silane is dependent on the AEGL document for HCl. Specific Comments Clarify the comment in Section 8.3 on page 12 that “potential unidentified reactive intermediate…may be involved in the toxicity…” If the toxic decomposition products could result, this undermines the AEGL derivations based on an analogy to HCl. Page 3, Table 2. Add the time to death for the one female that died when exposed to 1047 ppm. Page 4. The death of one of two animals is formated incorrectly as the fraction 1/2. Page 6, lines 10–12. See the statement above about dichlorodimethyl silane combustion/decomposition products, as this applies to trichloromethyl silane as well.
OCR for page 20
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels Page 8, line 21. The factor of 3 is not a modifying factor but an adjustment for molar equivalency. This needs to be revised to be consistent with pages A-2 and C-2. The AEGL-2 and AEGL-3 were derived using the HCl data for time scaling but in each case a different approach was taken with the same data. This is inconsistent because the AEGL document on HCl may soon be finalized with a similar issue, it will be necessary to make these two documents consistent. If the AEGL document for HCl is not finalized, then it is recommended that time scaling approaches using the same data should be consistently applied. COMMENTS ON ETHYLENIMINE At its July 21–23, 2003 meeting, the subcommittee reviewed the revised AEGL document on ethylenimine. The document was presented by Kowetha Davidson of Oak Ridge National Laboratory. The subcommittee recommends a number of revisions. General Comment There are relatively few publications available from which to derive AEGLs, but derivation of the AEGL-1 (page vi, line 4–5) is problematic. Upon what basis was an AEGL-2 value divided by one half to obtain an AEGL-1? Since no data were available for deriving AEGL-1 values (page vi, line 11), ethylenimine has poor warning properties (page vi, line d16) and death and other signs of poisoning can be delayed some 3 to 7-hr after exposure. Therefore, consideration should be given for not recommending an AEGL-1 for this material. In order to simply divide an AEGL-2 based on “extreme respiratory difficulty in guinea pigs” by two or any other value to obtain an AEGL-1 (notable discomfort and irritation), some indication of the slope of the concentration-response relation must be given. As written, derivation of the AEGL-1 does not appear founded in rigorous science. Specific Comments Page vi, lines 28–29. The NAC suggests that the “direct alkylating activity of ethylenimine is not expected to vary markedly among individuals in the population.” No empirical evidence or citations to peer-reviewed literature were supplied to support that contention. Even if the alkylating events proceed in a similar fashion and the time course at the nucleic acid and protein level between different people was identical, it is the repair of such molecular lesions that seems to account for different outcomes—unless of course the damage is so extensive that normal repair is overwhelmed. More extensive discussion of the basis for the NAC conclusion is needed here. Page 3, lines 40–43 and page 4, lines 1–28. The Weightman and Hoyle (1964) report does not add much to the TSD for ethylenimine. Exposure was to several chemicals, one of which is
OCR for page 21
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels ethylenimine. It needs to be explicitly stated that one cannot attribute the observed effects directly to ethylenimine. Page 3, line 40; page 19, line 24. Indicate whether these subjects were volunteers. Page 5, line 41. What is an “airline” respirator? Does this sentence mean to convey that these workers were equipped with personal protective clothing and an air-supplied respirator (as indicated on line 37)? Page 5, line 25. In what cells or tissues were the chromosome aberrations measured? Page 6, line 7. As written, the text suggests the odor of ethylenimine could be confused with that of ammonia? How does “because its odor is similar to that of ammonia” indicate poor warning properties for ethylenimine? Page 8, line 10. Note nasal irritation also occurred at 100 ppm or higher. Page 8, Carpenter et al. (1948) study. There is no discussion of effects at 10 ppm; presumably none were observed. Page 9, table 3. The 10 ppm group should be included. Page 11, line 11. State at what concentrations the kidney effects occurred. Page 12, lines 15–16. State the duration of administration of the compound in the diet. Page 15, line 23. The delayed onset of the inflammatory responses should be noted. Page 16, line 16. It should be clarified that based on lethality data, ethylenimine is more acutely toxic. Page 16, lines 15–16. The paragraph deals with the comparative potency of propylenimine compared to ethylenimine and states, “Ethylenimine is more toxic than either of the latter substances.” However, no indication of the magnitude of that difference was given. Presumably, this potency difference is restricted to lethality? Page 17, lines 3–4. Delete. The statement is speculation—particularly in light of the page 16, line 41 observations. Page 19, lines 24–33. It is not clear how specific substances in the Weightman and Hoyle (1964) study were ruled-out as causative. This section needs to acknowledge that the effects may not be attributed solely to ethylenimine due to the mixed exposure. Page 20, lines 16–21. Clarification is needed for the uncertainty factor of 3 for interspecies differences. Why would there be pharmacokinetic differences across species for an
OCR for page 22
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels alkylating agent with a portal of entry target (due to anatomical differences in the respiratory system)? Why is no difference in pharmacodynamics expected? Page 20, line 26. Clarification is also needed for the intraspecies uncertainty factor. Page 23, line 20. Add “not” between “do” and “take”, as the AEGL-3 does not take cancer risk into consideration. Page 24, lines 15–16. The suggestion concerning the skin notation is misguided. Given the results of the study described on page 11, lines 22–31, where no toxicity was evident after vapor exposure when inhalation was precluded, the suggestion is erroneous in the context of the AEGLs. The AEGLs are not analogous to the TLV in that the ACGIH skin designation is assigned when percutaneous contact contributes to systemic toxicity. Unless the NAC envisions public contact with liquid or neat ethylenimine under some unstated circumstance or release (page 1), the suggestion that a ‘skin designation’ be assigned to this material under the AEGL conditions is mistaken. Page 24, lines 16–17. Delete—this statement is redundant to page 23, lines 20–21. Page 25, line 14. Check the most recent TLV booklet to ascertain whether this material has been assigned the SEN (sensitizer) designation, as was suggested on page 24, line 35. Page vi, lines 29–31. The statement is unsupported. While the discussion seems to attempt to account for pharmacodynamic similarities, no mention is made of the range of pharmacokinetic parameters for this material. It may be worthwhile to cite reviews discussing accumulation of damage induced by nitrogen mustards (page 1, line 6) or other alkylating agents similar to ethylenimine to support the hypothesis presented. Page A-1, lines 19–22. Is this referring to the Weightman and Hoyle (1964) study? COMMENTS ON PROPYLENIMINE At its July 21–23, 2003 meeting, the subcommittee reviewed the AEGL document on propylenimine. The document was presented by Kowetha Davidson of Oak Ridge National Laboratory. The document could be finalized if the subcommittee is satisfied with the revisions made in response to its recommendations and if the AEGL document on ethylenimine is also approved by the subcommittee because the AEGL values for propylenimine are based on ethylenimine. General Comments Section 3, page 3. The studies are very old. What is the level of confidence that the reader can place on the results and the exposure concentrations?
OCR for page 23
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels If propylenimine is used to such great extent in modern industry, why is there such a dearth of data? Is this due to the systems? Explain the contemporary work practices used for this material and how it is handled and utilized in various industrial applications. The approach for propylenimine is based on ethylenimine; the AEGL document for propylenimine should not be finalized until the AEGL document for ethylenimine is finalized The rationale for selecting a relative potency of 5 for AEGL-2 needs to be clearly explained. Table 3 shows relative potency based on exposure duration (to same concentration and same percentage mortality) of 4–8. The choice of 5 seems arbitrary. Specific Comments Page 1, line 15. Is there a standard procedure for a relative toxicity approach in the SOPs? Page 8, line 21. The reason for a 2-fold modifying factor needs to be better explained. Editorial Comments Page v, lines 15–17. Upon what adverse health effect is the relative potency based? Page 5, lines 25–26. This sentence does not make sense. If the structural, chemical/physical, and toxicological properties are similar, does not it follow that the metabolic fate is the same? The sentence as written seems either redundant or obvious, and at a minimum the entry is confusing. Appendix A. Correct the number of significant figures. COMMENTS ON ALLYL ALCOHOL At its July 21–23, 2003 meeting, the subcommittee reviewed the AEGL document on allylamine. The document was presented by Claudia Troxel of Oak Ridge National Laboratory. The subcommittee recommends the following revisions. The document could be finalized if the subcommittee is satisfied with the revisions made in response to its recommendations. General Comment The regression analysis yielded n=0.78 and the NAC rounded it to 1. Yet, for propylenimine, the n value was 0.91 and it was not rounded. Therefore, the NAC has not been consistent with its approach. In the ten Berge analysis, there were only 20 compounds. The NAC has already evaluated many more materials; the NAC should contribute to the literature by publishing the procedures used and the factors which contribute to the values of n for time scaling.
OCR for page 24
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels COMMENTS ON ANILINE At its July 21–23, 2003 meeting, the subcommittee reviewed the AEGL document on aniline. The document was presented by Sylvia of Oak Ridge National Laboratory. The subcommittee recommends that the following revisions be made to the document. A revised draft should be reviewed by the subcommittee at its next meeting. Comments Development of 10-min AEGLs for Aniline When the AEGL program was initiated, the plan was to develop AEGL values for exposure durations of 30 min, 1 hr, 4 hr, and 8 hr, which the subcommittee reviewed for the initial group of hazardous chemicals (aniline, arsine, chlorine, fluorine, 1,2-dichloroethane, phosphine, methylhydrazine, and dimethylhydrazine). AEGLs for four of these chemicals were published in the first volume of the subcommittee’s series Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 1 (NRC 2000). Subsequently, in response to a request from the sponsors, the subcommittee developed 10-min AEGL values for those chemicals, and 10-min AEGLs for all chemicals beginning with the subcommittee’s fifth meeting in 1999. However, this is the first revision of any previously established AEGL document, therefore, the NAC must ensure that all data published since the publication of AEGLs Volume 1 have been reviewed, and that no recent data are available that warrant changing the adopted values. Since the material presented in these documents is an exact copy of the summary of documents in Volume 1 (NRC 2000) and the data in the tables are identical to those in Volume 1 (except for the addition of the 10-min AEGL values), the subcommittee requires assurance that such a search has been carried out, and that all relevant new data have been considered. The subcommittee concludes that 10-min AEGLs derived for aniline are appropriate. However, the NAC should explain the scientific rationale to support linear extrapolation of the 8-hr values to a 10-min duration of exposure. Expanded details should be provided for the supporting studies, particularly, that of Kakkar et al. study. Page 31, paragraph 2. Is there a reason to believe that a 15,302 ppm exposure will not result in mortality in a typical acute lethal study design (i.e., a 10-min exposure followed by a 14-day observation period)? Did the study protocol include measurement of circulating methemoglobin? If not, this study should be viewed cautiously in supporting the 10-min AEGL-3 values. Page 33, line 2–3. The text should read: Following a 10-min exposure, the concentration of methemoglobin in blood is unlikely to reach equilibrium values, as typically seen 6–8 hr after the initiation of exposure. Page 33, line 4. “very…. 10 min” should be deleted.
OCR for page 25
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels The justification on page 33 is appropriate for the 10-min AEGL values. The subcommittee recommends that the entire revised aniline document be included in the next AEGL volume. COMMENTS ON ARSINE At its July 21–23, 2003 meeting, the subcommittee reviewed the AEGL document on arsine. The document was presented by Robert Young of Oak Ridge National Laboratory. The subcommittee recommends that the following revisions be made. This document can be finalized if the recommended revisions are made appropriately. Comments Development of 10-min AEGLs for Arsine The subcommittee’s concern for arsine is the same as previously discussed for aniline. The revised document is identical to that published in AEGLs Volume 1 (NRC 2000) and the derivation of the AEGLs in the appendix is also identical to that in Volume 1 except for addition of 10-min AEGL values. The subcommittee must be assured that all relevant new data have been considered and that there are no new data that would alter the previously recommended AEGLs for arsine. Page 3. A decreased hematocrit is not a sensitive indicator of arsine toxicity because it reflects chronic anemia. An increase in plasma hemoglobin (or potassium) is a better indicator for acute hemolysis. Page 5. AEGLs-2 end point: How can absence of an effect be consistent with arsine toxicity? High-susceptibility groups have to be defined (sickle cell anemia; G6PD-deficiency). The garlic odor of arsine was used until 1840 without too many casualties. It would be relevant to know how the odor threshold compares with the proposed AEGLs. The document should provide the odor threshold value of arsine in air. Delete calculation for AEGL-1 values. COMMENTS ON MONOMETHYLHYDRAZINE At its July 21–23, 2003 meeting, the subcommittee reviewed the revised AEGL document on monomethylhydrazine. The revised document is identical to that published in AEGLs Volume 1 (NRC 2000), and the derivation of the AEGLs in the appendix is also identical to that in Volume 1 except for addition of 10-min AEGL values. The subcommittee must be assured
OCR for page 26
Tenth Interim Report of the Subcommittee on Acute Exposure Guideline Levels that all relevant new data have been considered and that there are no new data that would alter the previously recommended AEGLs for monomethylhydrazine. This document can be finalized if the recommended revisions are made appropriately. COMMENTS ON DIMETHYLHYDRAZINE At its July 21–23, 2003 meeting, the subcommittee reviewed the revised AEGL document on dimethylhydrazine. The document was presented by Robert Young of Oak Ridge National Laboratory. The revised document is identical to that published in AEGLs Volume 1 (NRC 2000), and the derivation of the AEGLs in the appendix is also identical to that in Volume 1 except for addition of 10-min AEGL values. The subcommittee must be assured that all relevant new data have been considered and that there are no new data that would alter the previously recommended AEGLs for dimethylhydrazine. This document can be finalized if the recommended revisions are made appropriately
Representative terms from entire chapter: