Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
6 C2-C9 Alkanes1 J. Torin McCoy Toxicology Group Habitability and Environmental Factors Division Johnson Space Center National Aeronautics and Space Administration Houston, Texas SUMMARY OF APPROACH A number of hydrocarbons fall under the category of C2-C9 aliphatic satu- rated alkanes. Rather than address each compound individually, this document initially sought to establish a group spacecraft maximum allowable concentra- tion (SMAC) for the C2-C9 alkanes in accordance with their similarities in toxic action and their physical and chemical properties. A SMAC for methane (C1) has already been established based on its lower limit of explosivity rather than on its chemical toxicity. Estimating health-protective environmental limits for groups of chemicals can be a useful and practical risk-assessment approach, as evidenced by its suc- cess in establishing guidelines for total petroleum hydrocarbon fractions (MA DEP 1994). Group SMACs can be an efficient way to evaluate spacecraft air concentrations, while helping to limit redundancy in developing individual SMACs. However, group SMACs lose some of their usefulness when there are broad variations in toxicity among group members or when the toxic effects produced by one member are poorly relevant to the remainder of the group. Some variation in toxicity among a group is unavoidable, and group SMACs often serve a screening function in the sense that the developed SMACs may be driven by one or two group representatives that exhibit the greatest toxicologic response for an end point that is relevant to the entire group. In this evaluation, we determined that it was neither realistic nor produc- tive to establish group SMACs for the C2-C9 alkane range. We based this con- clusion on two critical observations. First, for ethane (C2), propane (C4), and butane (C4), there is clearly limited potential for toxicity associated with envi- ronmentally relevant exposures to these compounds in air. Consistent with the 1 Not including n-hexane. 85
86 SMACs for Selected Airborne Contaminants approach taken in SMAC development for methane, individual SMACs for these compounds were based on 10% of the lower limits of explosivity for each of these gases. By maintaining concentrations below each chemicalâs limit, health concerns (including the potential for them to act as simple asphyxiants) are minimized. The C5-C9 alkanes are treated as a group, with the exception of n-hexane. Significant research and epidemiologic findings have established that n-hexane differs from the other alkanes (and from the other hexane isomers) in its potential to cause peripheral neuropathy (Spencer et al. 1980, Takeuchi et al. 1980, Frontali et al. 1981, Filser et al. 1996, ACGIH 2008). Because considera- tion of this specific effect for n-hexane would likely result in a group SMAC toxicologically irrelevant for most of members of the group, n-hexane is re- served for separate SMAC development and is not fully addressed in this docu- ment from a toxicologic standpoint. PHYSICAL AND CHEMICAL PROPERTIES Physical and chemical properties for the C2-C9 alkanes are presented in Table 6-1. Ethane (C2) through butane (C4) exist as gases at standard tempera- ture and pressure, whereas pentane (C5) to nonane (C9) are liquids. Various branched isomers exist for many of the n-alkanes, and the physical and chemical properties for these isomers may differ from those presented in Table 6-1, which are specific to the n-alkanes. OCCURRENCE AND USE The group of compounds that comprise the C2-C9 saturated aliphatic al- kanes (the linear and branched alkanes from ethane through nonane) are present in earth gases and crude oil and have a variety of commercial and industrial ap- plications. The gaseous alkanes within this group (ethane, propane, and butane) are principal components of natural gas and are used widely as fuels and propel- lants (Sandmeyer 1981). They are also greenhouse gases and contribute to the formation of ground-level ozone (Katzenstein et al. 2003). The liquid alkanes (pentane, hexane, heptane, octane, and nonane) also may be found at low con- centrations in natural gas but are known more for their applications as solvents and as important components of crude oil, diesel, and gasoline. n-Hexane is a widely used industrial solvent, as are pentane and heptane (Finkel 1983). Octane is used extensively as a preignition additive for high-compression engine fuels; other alkanes are either found in or are added to gasoline and other fuels. They are also formed and released after combustion of the fuels in automobiles, boil- ers, and other machinery (Sandmeyer 1981). The human body produces some of the more volatile alkanes (e.g., ethane and pentane) endogenously as a result of the breakdown of polyunsaturated fatty acids (Galvin and Marashi 1999); they can be measured at low concentrations in human breath (Frank et al. 1980).
TABLE 6-1 Physical and Chemical Properties of C2-C9 n-Alkanes C2 C3 C4 C5 C6 C7 C8 C9 Formula C2H6 C3H8 C4H10 C5H12 C6H14 C7H16 C8H18 C9H20 Name Ethane Propane Butane Pentane Hexane Heptane Octane Nonane CAS registry no. 78-84-0 74-98-6 106-97-8 109-66-0 110-54-3 142-82-5 111-65-9 111-84-2 Molecular weight 30.1 44.1 58.1 72.1 86.1 100.2 114.2 128.3 Boiling point (Â°C) â88.6 â42.1 â0.5 36.1 69 98.4 125.7 150.8 Lower explosive limit 3.2% 2.3% 1.9% 1.4% 1.2% 1.2% 1.0% 0.9% Upper explosive limit 12.5% 9.5% 8.4% 7.8% 7.8% 6.7% 4.7% 2.9% Vapor pressure (mm Hg) Gas Gas Gas 400 (18Â°C) 100 (20Â°C) 40 (25Â°C) 10 (19Â°C) 10 (38Â°C) Conversion factor 1.23 1.80 2.38 2.95 3.52 4.10 4.67 5.25 (ppm to mg/m3) 25Â°C Isomers 1 1 2 3 5 9 18 33 Abbreviations: mg/m3, milligrams per cubic meter; ppm, parts per million. Source: Sandmeyer 1981, ACGIH 1991. 87
88 SMACs for Selected Airborne Contaminants Given their widespread uses, these saturated hydrocarbons are commonly encountered in ambient air. For example, Hawas et al. (2001) found that C5-C9 aliphatic alkanes composed more than 80% of the total concentration of volatile organic compounds measured in ambient air from a light industrial area of Bris- bane, Australia. Individual alkanes were measured in ambient air at concentra- tions as high as 200 parts per billion (ppb) (n-pentane), with the highest three averages reported for n-pentane (70 ppb), 2,3-dimethylbutane (60 ppb), and n- hexane (19 ppb). The C2-C4 gases have been studied because of their impor- tance in the formation of ground-level ozone and were recently measured in the near-surface atmosphere of Texas at 34 ppb (ethane), 20 ppb (propane), and 13 ppb (butane) (Katzenstein et al. 2003). C2-C9 alkanes may be present in certain spacecraft payloads and are found in small amounts in certain flight hardware (e.g., detectors). Some mem- bers of this group have been measured in air samples collected onboard the In- ternational Space Station (ISS), although they typically occur at very low con- centrations. For example, Russian analysis of AK-1M tubes collected before the return of 6 Soyuz (Nov. 2003) reported that most C5-C9 alkanes were pre- sent at concentrations below 0.1 ppm. In 2003, Russian analysts reported much higher concentrations of alkanes in an air sample collected from the Progress resupply vehicle (total hydrocarbon concentrations of 34 milligrams per cubic meter [mg/m3], with pentane isomers as the primary components), although their source remains unclear. The C2-C9 alkanes either are not reported or are found in very low concentrations (e.g., 0.1-0.2 mg/m3) in U.S. analyses of ISS air sam- ples. Toxicokinetics Given that a group SMAC is being developed for C2-C9 alkanes, the toxi- cokinetics of these compounds is discussed as it applies to the class of com- pounds. Thus, the focus is on presenting an accurate general toxicokinetic pic- ture for this group rather than detailing the process for each member of the group. Because of the commercial importance of n-pentane and n-hexane, more specific information exists about their toxicokinetics than for some of the other alkanes. Because information on these two alkanes cannot necessarily be used to make general statements for the entire group, an attempt was made to describe how variations in carbon chain length or structure (e.g., branched isomers) may affect each area of toxicokinetics. Absorption There is some variation in respiratory absorption among the aliphatic C2- C9 saturated alkanes, but they are generally not as well absorbed as the unsatu- rated alkanes (Zahlsen et al. 1990). Among the C2-C9 saturated alkanes, absorp- tion into the bloodstream after inhalation exposures will generally be greater for
C2-C9 Alkanes 89 the linear members of this group and for those with higher molecular weights (Galvin and Marashi 1999). Perbellini et al. (1985) conducted in vivo studies of human tissues and blood and found that alkane solubility in blood typically in- creased with increasing molecular weight, as evidenced by blood:air partitioning coefficients of 0.4, 0.8, and 1.9 for n-pentane, n-hexane, and n-heptane, respec- tively. n-Hexane reached steady-state blood concentrations within 100 min dur- ing a limited 4-h inhalation study with human volunteers (Veulemans et al. 1982). In a longer study, Zahlsen et al. (1990) observed that peak blood concen- trations of n-nonane occurred within the first day (12 h) of exposure during a 14-d inhalation test with rats. Distribution Once dissolved in the bloodstream, the saturated alkanes can be distrib- uted to various organ systems. In general, in accordance with the lipid solubility of the alkanes, most can easily cross biologic membranes, and the group has an affinity for lipid-soluble tissues (Perbellini et al. 1985, Robinson 2000). Perbel- lini et al. (1985) found that n-pentane was distributed to a significant degree to adipose, brain, liver, and kidney tissue. The higher molecular weight alkanes would be expected to move from blood into these organs even more readily be- cause of their greater solubility in lipids. That was observed in a study of n- nonane by Zahlsen et al. (1990) where, for example, n-nonane accumulated in the brain at concentrations twice as high as trimethylbenzene, although n-nonane was present in the blood at only a third of the concentration of trimethylbenzene. Peak brain tissue concentrations of n-nonane were reached after 12 h of inhala- tion exposure. This transfer from the blood occurs in spite of the possibility of size restriction, which can slow the absorption of high molecular weight com- pounds across the blood-brain barrier (Hau et al. 2002). Metabolism and Excretion Frank et al. (1980) studied the rate of elimination of n-pentane in rats after inhalation and reported an inhalation half-life of 2.3 h. They found that the addi- tion of peroxidation inhibitors (dithiocarb or ethanol) significantly increased the half-life. Consistent with similar observations reported by Allerheilingen et al. (1987), who observed that clearance decreased in the presence of carbon tetra- chloride (which destroys cytochrome P-450), it appears that metabolism by the mixed-function oxygenase system plays a significant role in the elimination of n-pentane and other saturated alkanes. A closed-chamber test with radiolabeled n-pentane revealed that 50% of the radioactivity was excreted as CO2 after 8 h, with 7.6% metabolized and released in urine (Daugherty et al. 1988). The au- thors assumed that most of the remainder (about 40%) was primarily excreted unchanged in the breath. This percentage released in the breath is probably an overestimate for the alkanes with lower volatility and increased blood solubility.
90 SMACs for Selected Airborne Contaminants For example, no more than 20% was estimated to be excreted unchanged in the breath for n-hexane (ATSDR 1999). With respect to the fraction of inhaled alkanes absorbed into the blood- stream and metabolized, there are important metabolic distinctions among the individual C2-C9 alkanes. To a significant degree, it is thought that a major elimination pathway for the n-alkanes is chain breakdown and elimination of the carbon as expired CO2 (Daugherty et al. 1988). Dahl (1989) exposed rats to ra- diolabeled n-octane and found that to be the case. However, Dahlâs studies of a branched alkane isomer (isooctane) produced very different results. Instead of being metabolized to CO2, the branched structure prevented normal chain break- down and resulted in significant excretion as urinary metabolites (e.g., alcohols). Instead of being released as CO2, alkanes can be oxidized by microsomal P-450 activity and undergo aliphatic hydroxylation to form an alcohol (Sand- meyer 1981). Studies with n-pentane (Frommer et al. 1970) have shown that 2-pentanol is the major metabolite (89%) formed by rat liver microsomes, with 3-pentanol as a minor metabolite (11%), consistent with the aliphatic hydroxyla- tion process. However, as is observed with n-hexane, the alcohol can also be further oxidized to form ketones, diketones, diols, and carboxylic acids (ATSDR 1999). These intermediates can be released in urine or exhaled (EEMA 1995). Some of these metabolites have exhibited specific toxicity that is not relevant to other metabolites of the same parent hydrocarbon. In particular, exposure to n- hexane has been associated with the development of neuropathy that is mediated by the formation of a Î³-diketone (2,5-hexanedione) following metabolism of the parent hydrocarbon (Schaumburg and Spencer 1976). The same effect is not observed with 2,4- or 3,5-diketones, and it is thought that the neurotoxicity of the Î³-diketone is due to its high water solubility and ability to form stable conju- gated Schiff bases (Graham and Abou-Donia 1980). TOXICITY SUMMARY Available data suggest that a distinction can be made between the gaseous (C2-C4) and liquid (C5-C9) saturated alkanes in terms of their toxicity. Because of the widespread harmless exposures associated with their presence as compo- nents of natural gas, ethane, propane, and butane are generally viewed as having extremely low chemical toxicity (Sandmeyer 1981; Finkel 1983). One of the main hazards with these gases is their potential to cause asphyxiation after their release in an enclosed environment. These light gases can dilute available con- centrations of oxygen, thus decreasing oxygen uptake by the lung (Galvin and Marashi 1999). That is not to say that there is no toxicity associated with these alkanes. For example, in summarizing the toxicity of these gases, the European Agency for the Evaluation of Medicinal Products reported the following effects: (1) propane concentrations of 100,000 parts per million (ppm) (10%) caused respiratory depression and bronchospasm in exposed mice; (2) isobutane re- sulted in increased pulmonary resistance and depressed minute volume in rats
C2-C9 Alkanes 91 exposed at 50,000 ppm; (3) in humans, 100,000 ppm of n-butane caused vertigo after 2 min of exposure; (4) 350,000 ppm of isobutane had an anaesthetic effect in mice after 25 min; and (5) 520,000 ppm of isobutane was a 2-h LC50 (dose lethal to 50% of subjects) for mice (EEMA 1995). These concentrations are not environmentally relevant and confirm only that there is very limited potential for these gases to be toxic. However, these gases do have significant potential to explode if allowed to accumulate to sufficient concentrations (lower explosive limits of 1.9% to 3.2%), and consideration of these hazards appears to be more than adequately protective of any potential toxicity (Sandmeyer 1981). This conclusion does not extend to the C5-C9 saturated alkanes because a number of toxicologic effects at more realistic concentrations of exposure have been reported in the scientific literature (although there is still variability in the toxic potential among members of this group). Those acute, subchronic, and chronic effects of particular relevance to SMAC development are more fully described in the following sections (refer to Table 6-2 for study details). When possible, general trends in toxicologic effect are noted among the different members of the series of C5-C9 saturated alkanes. Knowledge of these general trends is useful, especially when scanty toxicity data are available for a specific alkane (Robinson 2000). Acute Exposures As might be predicted based on their physical and chemical properties, the central nervous system (CNS) is a main target for the C5-C9 saturated alkanes (Hau et al. 2002). Adverse CNS effects may include narcosis and anesthetic effects, and have been reported in humans in the form of headaches, exhilara- tion, nausea, a loss of fine motor skills, difficulty concentrating, confusion, and loss of appetite (Sandmeyer 1981, Finkel 1983). Swann et al. (1974) noted that an increase in the length of the carbon chain among the C5-C9 alkanes typically results in a greater potential for narcotic effects, although exceptions have been cited (Nilsen et al. 1988). For example, in mice exposed to n-pentane, concentra- tions of at least 32,000 ppm were necessary to produce light anesthesia, although 8,000 ppm of n-heptane was capable of producing the same effect (Swann et al. 1974). Branched alkanes may not have the same potential for anesthetic effects as their linear alkane counterparts (Sandmeyer 1981) because isooctane did not result in anesthesia in this study, even at lethal concentrations of 32,000 ppm. A number of studies have evaluated the acute toxicity of n-pentane (com- prehensive review by Galvin and Marashi 1999). It is clear from reviewing the results of these studies that n-pentane has very low acute toxicity, because many studies failed to find any adverse effects, and those that did typically required extremely high concentrations (32,000-100,000 ppm). Carpenter et al. (1978) evaluated the acute toxicity of n-nonane vapors in rats. The LC50 (after 14 d) for rats subjected to inhalation exposures to n-nonane for 4 h was 3,200 ppm.
92 TABLE 6-2 Toxicity Summary for C5-C9 Saturated Alkanes (Excluding n-Hexane) Effect Concentration (ppm) Exposure Duration Species Effects Reference n-Pentane 3,000 16 wk (12 h/d) Wistar rats No neurotoxic effects. Takeuchi et al. 1980 3,000 30 wk (9 h/d, 5 d/wk) Sprague-Dawley rats No neurotoxic effects. Frontali et al. 1981 4,500 (50:50, pentane: 13 wk (6 h/d, 5 d/wk) Fischer 344 rats No reported kidney effects. Aranyi et al. 1986 butane) 7,000 90 d (6 h/d, 5 d/wk) Sprague-Dawley rats No effects (including McKee et al. 1998 survival, body weight changes, organ weight, tissue pathology [including reproductive tissue], blood chemistry). 10,000 2 wk (6 h/d, 5 d/wk) Charles River rats No effects besides reversible Stadler et al. 2001 and slight clinical pathology changes. 10,000 Gestation day 6-15 Pregnant Charles River rats No developmental toxicity. Hurtt and Kennedy 1999 32,000 5 min Swiss mice Light anesthetic effects. Swann et al. 1974 128,000 5 min Swiss mice Respiratory arrest (1/4 Swann et al. 1974 mice). 2-Methylpentane 1,500 14 wk (9 h/d, 5 d/wk) Sprague-Dawley rats No neurotoxic effects. Frontali et al. 1981 3-Methylpentane 1,500 14 wk (9 h/d, 5 d/wk) Sprague-Dawley rats No neurotoxic effects. Frontali et al. 1981 n-Heptane 800 28 d (6 h/d) Long Evans rats Ototoxicity NOAEL. Simonsen and Lund 1995
1,500 30 wk (9 h/d, 5 d/wk) Sprague-Dawley rats No neurotoxic effects. Frontali et al. 1981 3,000 16 wk (12 h/d) Wistar rats No neurotoxic effects. Takeuchi et al. 1980 4,000 28 d (6 h/d) Long Evans rats Increased auditory Simonsen and Lund 1995 threshold, ototoxicity. 6,700 10 min CF-1 mice 10% reduced respiratory rate Kristiansen and Nielsen (RD10 a) calculated from 1988 concentration-response relationship. 8,000 5 min Swiss mice Anesthetic effects. Swann et al. 1974 17,400 10 min CF-1 mice 50% reduced respiratory rate Kristiansen and Nielsen (RD50 b) calculated from 1988 developed concentration- response relationship. 48,000 5 min Swiss mice Respiratory arrest Swann et al. 1974 (3/4 mice) Isooctane 16,000 5 min Swiss mice Respiratory arrest Swann et al. 1974 (1/4 mice). 32,000 5 min Swiss mice Respiratory arrest Swann et al. 1974 (4/4 mice). n-Octane 3,800 10 min CF-1 mice 10% reduced respiratory rate Kristiansen and Nielsen (RD10 a), calculated from 1988 concentration-response relationship. (Continued) 93
94 TABLE 6-2 Continued Effect Concentration (ppm) Exposure Duration Species Effects Reference n-Nonane 590 13 wk (6 h/d, 5 d/wk) Wistar rats NOAEL for reduced weight Carpenter et al. 1978 gain; also, no shorter term effects that were noted in the 7-d study at 1,500 ppm. 1,100 10 min CF-1 mice 10% reduced respiratory rate Kristiansen and Nielsen (RD10 a), calculated from 1988 concentration-response relationship. 1,500 7 d (6 h/d) Wistar rats Mild tremors, coordination Carpenter et al. 1978 loss, irritation. 1,600 13 wk (6 h/d, 5 d/wk) Wistar rats Mild tremors, coordination Carpenter et al. 1978 loss, salivation, and significantly reduced weight gain over the exposure duration relative to controls 2,414 8 h (14 d observed Sprague-Dawley rats NOAEL for CNS effects. Nilsen et al. 1988 3,200 4 h (14 d observed) Wistar rats LC50 Carpenter et al. 1978 3,560 8 h (14 d observed) Sprague-Dawley rats LOAEL, gross ataxia, Nilsen et al. 1988 spasms, loss of Purkinje cells. a RD10 is the concentration expected to result in a 10% decrease in respiratory rate relative to controls. b RD50 is the concentration expected to result in a 50% decrease in respiratory rate relative to controls. Abbreviations: LOAEL, lowest-observed-adverse-effect level; NOAEL, no-observed-adverse-effect level.
C2-C9 Alkanes 95 Rats exposed to n-nonane at 1,500 ppm for 6 h/d for 7 d exhibited mild tremors, loss of coordination, and slight irritation. The same effects were exhibited in a longer term experiment with n-nonane at 1,600 ppm, which was part of the Car- penter et al. study, but the effects were not observed with exposures to n-nonane at 590 ppm. Nilsen et al. (1988) evaluated the toxicity of n-nonane and other higher al- kanes in association with inhalation exposures of 8 h followed by a 14-d obser- vation period in Sprague-Dawley mice. They focused on evaluating CNS and respiratory system effects. They evaluated four exposure groups, with concen- trations ranging from 2,414 to 5,289 ppm, and found a good dose-response relationship with tremors, ataxia, spasms, and limb paralysis after 2-4 h of expo- sure. An 8-h LC50 of 4,500 ppm was estimated for n-nonane, with death occur- ring as a result of cardiopulmonary insufficiency (the authors were not able to distinguish whether it was due to CNS depression or to direct action on the heart or lungs). One interesting finding was the report of neurotoxicity in mice surviving exposure to n-nonane at 4,438 ppm. On autopsy (14-d postexposure), surviving rats (6/10) were found to have a significant loss of Purkinje cells and a marked increase in damaged cerebellar neurons compared with controls. No neuropathic changes were found in any of the mice dying before the 14-d observation period, nor was it observed with C10-C13 alkanes tested. The authors suggested that this effect is likely specific to n-nonane and is indicative of its neurotoxicity. However, they also cautioned that the effect could be attributable to general hy- poxia. As described later in this document, the neuropathy reported by Nilsen et al. (1988) is inconsistent with the current weight of evidence for the C5-C9 al- kanes, which points only to n-hexane as having specific neurotoxicologic prop- erties that distinguish it from the other members of this group. In the same study, anesthetic effects were not noted, even at the highest tested concentrations for n-nonane. This observation can be partially explained by other studies that have noted slower absorption into the brain for the larger alkanes despite their lipophilicity (Hau et al. 2002). However, as other signifi- cant CNS effects were noted in this 8-h study, more research is needed to fully understand the factors controlling the anesthetic properties of these alkanes. Swann et al. (1974) also observed that mice commonly exhibited respira- tory irregularities progressing to arrest after inhalation exposures to saturated alkanes. One in four mice experienced respiratory arrest in association with 5 min of exposure to n-pentane at 128,000 ppm, although only 16,000 ppm pro- duced the same effect with isooctane. All four mice went into arrest within 3 min at 32,000 ppm. The authors suggested that these observed respiratory ef- fects may not simply be due to direct alkane action on CNS depression and noted that the observed respiratory changes corresponded to irritation (deter- mined by assessing body movements). This irritation increased in severity and sensitivity with increasing carbon chain length of the tested alkanes. Other authors have studied specific respiratory changes in laboratory animals in response to exposure to saturated alkanes in the context of deter-
96 SMACs for Selected Airborne Contaminants mining the potential for these compounds to act as sensory irritants (Kristian- sen and Nielsen 1988). Alarie (1973, 1981) suggested that the trigeminal nerve endings in the nasal mucosa would respond to sensory irritants (resulting in a burning sensation in the mucous membranes of the eyes, nose, and throat) and that this response could be measured by a decrease in respiratory rate. This type of sensory irritation would affect the upper respiratory tract, as distin- guished from pulmonary irritants, which act on the lower respiratory tract (Alarie 1973). To study these effects, Kristiansen and Nielsen (1988) used an inhalation test chamber to expose groups of four CF-1 mice for 30 min to a range of C7-C11 alkane concentrations (Table 6-2). Certain groups of mice were exposed through a trachea cannula, thus bypassing the trigeminal nerve to assess nonsensory irritation. The authors measured the response at each exposure by evaluating the peak percentage change in the average respiratory rate of each group of four mice relative to their average preexposure level (as measured during a 10-min control period). On the basis of this testing, they determined a relationship between exposure concentration and the reduction in respiratory rate relative to controls for each alkane (Figure 6-1). They applied the regression equation for this relationship to predict RD0, RD10 (the concen- tration expected to result in a 10% decrease in respiratory rate relative to con- trols), and RD50 (the concentration expected to result in a 50% decrease in res- piratory rate relative to controls) for each alkane. Pulmonary irritation seemed not to be a significant factor with the alkanes. This study affirmed previous observations (Swann et al. 1974, Sandmeyer 1981) about the relationship be- tween carbon chain length and the potential for irritation, with n-nonane re- sulting in respiratory rate reduction at the lowest concentration of the alkanes tested that were relevant to this group SMAC. Table 6-3 presents the RD0, RD10, and RD50 values for the C7-C9 alkanes (with data for C10 and C11 al- kanes also provided for comparison) based on results observed within the first 10 min of the test (when effects could be most clearly attributed to sensory irritation). 60 Respir atory Rate- % De crease 40 n-heptane n-octane 20 n-nonan e 0 2 2.5 3 3.5 4 4.5 5 Log Concentr ation (ppm) FIGURE 6-1 Concentration-response slopes for decreases in respiratory rate after expo- sures to n-heptane, n-octane, and n-nonane (Kristiansen and Nielsen 1988). The slopes extend to an RD50, but only n-heptane exhibited an RD50 response at the concentrations tested.
C2-C9 Alkanes 97 TABLE 6-3 Predicted RD0, RD10, and RD50 Values for C7-C11 Alkanes from Sensory Irritation Investigations Compound RD0 (ppm) RD10 (ppm) RD50 (ppm) n-Heptane 6,000 6,700 17,400 n-Octane 3,000 3,800 18,100 n-Nonane 620 1,100 62,200 n-Decane 110 420 >1 Ã 106 Undecane Not predicted Not predicted because Not predicted because because of lack of of lack of significance of lack of significance significance of the of the regression of the regression regression Source: Kristiansen and Nielsen 1988. Reprinted with permission; copyright 1988, Ar- chives of Toxicology. An important finding is that, among the alkanes and concentrations tested, only n-heptane exhibited enough of a sensory response to reduce respi- ration by 50% relative to controls. The authors postulated that this response may be because of stabilization of the receptor-substance complex or because the greater water solubility of the shorter chain alkanes may allow them more contact with receptors deep within the nasal mucosa. Thus, these findings sug- gest that exposure to n-nonane may cause sensory irritation at the lowest air concentrations (among the C5-C9 alkanes), although the severity of the re- sponse above that threshold may not be drastic (see Figure 6-1). Subchronic and Chronic Exposures Several longer term studies and reviews of n-pentane toxicity described in the literature consistently have shown n-pentane to have very low toxicity. McKee et al. (1998) conducted a 90-d inhalation study in which Sprague- Dawley rats were exposed to n-pentane at concentrations of 0, 1,700, 3,500, and 7,000 ppm. Various end points were examined, including weight gain, changes in various organ weights, and organ morphology. No treatment-related effects were observed at any concentration. Stadler et al. (2001) also exposed rats through inhalation to n-pentane for 2 wk to concentrations as high as 10,000 ppm. They observed subtle changes in clinical pathology at the 3,000- and 10,000-ppm dosing concentrations (changes in serum calcium and phosphorous concentrations), but these changes were slight, reversible, and within the range of variation seen in historical controls. They observed no outward signs of toxic- ity in any of the rats, and histopathologic evaluation of organs and the assess- ment of body weight showed no adverse effects after the n-pentane exposure. Taken as a whole, the scientific evidence suggests that n-pentane has little po- tential for adverse health effects, especially in association with exposures to environmentally relevant concentrations (Galvin and Marashi 1999).
98 SMACs for Selected Airborne Contaminants Subchronic and chronic exposures to n-hexane have been shown to result in a specific polyneuropathy in human epidemiologic studies as well as in stud- ies with laboratory animals (Yamada 1967, Takeuchi et al. 1980, Frontali et al. 1981, Filser et al. 1996). This has been found to be attributable to the neurotoxic potential of a specific metabolite of n-hexaneâ2,5-hexanedione, which is a Î³- diketone (DiVincenzo et al. 1980). Similar metabolites (for example, 2,4- hexanedione) do not exhibit the same neurotoxicity (Graham and Abou-Donia 1980). For many of the epidemiologic cohorts, it was not possible to rule out the presence of other alkanes besides n-hexane, and there has been some discus- sion about whether other alkanes could be metabolized to form sufficient quanti- ties of neurotoxic Î³-diketones (Truhaut et al. 1973, 54 Fed. Regist. 2424). However, convincing work (Takeuchi et al. 1980, Frontali et al. 1981, Filser et al. 1996) suggests that the formation of necessary Î³-diketones is favored by n- hexane metabolism, whereas it is not favored for other closely related alkanes. Therefore, the weight of evidence suggests that n-hexane has significantly greater potential to cause neurotoxicity than the other C5-C9 alkanes. For this reason, n-hexane is not included with the other C5-C9 alkanes in the develop- ment of a group SMAC; a separate SMAC will be developed in another docu- ment. Hydrocarbon-induced nephropathy in rodents has been widely ad- dressed in the scientific literature, especially for the C5-C9 hydrocarbons that are present in gasoline vapors (Aranyi et al. 1986, Halder et al. 1986). This nephrotoxicity (observed in male rats only) includes progressive renal tubular disease and primary renal neoplasms (Dahl 1989); several C5-C9 saturated alkanes have been shown to exhibit this type of nephrotoxicity (e.g., isooctane, 2,3,4-trimethylpentane). However, it has been accepted that this effect is likely mediated by binding to Î±2Âµ-globulin (Short et al. 1986, Loury et al. 1987, Dahl 1989), a protein unique to male rats and therefore irrelevant to SMAC development. Simonsen and Lund (1995) conducted a study to evaluate the potential for exposure to aliphatic saturated alkanes (n-heptane) to result in ototoxicity. Pre- vious work (Pryor et al. 1984, Rebert et al. 1991) showed auditory impairments in rats after exposure to other organic solvents (e.g., toluene, trichloroethylene). Simonsen and Lund tested whether n-heptane inhalation could reduce auditory thresholds in mice. They used implanted electrodes and needle electrodes to measure auditory brain stem responses elicited by an instrument that emitted sounds at frequencies of 3, 8, 16, and 32 kilohertz (kHz) and eight levels of sound intensity ranging from 25 to 95 decibels (dB). The rats were exposed through inhalation to n-heptane at 0, 800, and 4,000 ppm 6 h/d for 28 d. For the 4,000-ppm exposure group, effects were observed on the auditory brain stem response, which was most marked at the higher sound intensities. These effects translated into an increase in the auditory threshold by approximately 10 dB for the rats in this exposure group (although the effect was observed at all frequen- cies, it was statistically significant for controls only at 8 and 16 kHz). Rats in the 800-ppm exposure group showed no effects that were significantly different
C2-C9 Alkanes 99 from controls, and this concentration was proposed as a no-observed-adverse- effect level (NOAEL) for the study. Although this study was not designed to elucidate the mechanism of the observed ototoxicity, other studies have sug- gested that the effects may be related to a loss of outer hair cells in the coch- leaâcells that aid in frequency selectivity (Pryor et al. 1984, Yano et al. 1992). With respect to studies involving longer term inhalation exposures to al- kanes, Carpenter et al. (1978) exposed rats to n-nonane vapors 6 h/d, 5 d/wk for 13 wk at concentrations of 0, 590, and 1,600 ppm. They observed excess saliva- tion and lacrimation throughout the study for the rats in the highest exposure group. These rats also experienced a significant decrease in body weight gain relative to controls. These effects were not observed in the 590-ppm exposure group, and microscopic histopathology revealed no lesions or other effects at- tributable to the n-nonane exposure. Thus, 590 ppm was suggested as a NOAEL for these effects with n-nonane. Genotoxicity and Mutagenicity Limited information is available on the genotoxicity or mutagenicity of the C5-C9 alkanes in the scientific literature. Some information is available on ef- fects due to n-hexane, especially its 2,5-hexanedione metabolite, but evaluation of this information is not relevant to this document. For n-pentane, no evidence of mutagenicity was observed in five species of Salmonella in the in vitro Ames assay (Kirwin et al. 1980). n-Pentane was also not mutagenic in a dominant le- thal study when it was injected into the peritoneum of male mice (Epstein et al. 1972). A chromosome aberration test in Chinese hamster ovary cells conducted by McKee et al. (1998) produced equivocal results, especially given the high concentrations used (>10 millimolar) in the testing. Given these results, McKee et al. (1998) also conducted an in vivo bone marrow micronucleus test by exam- ining the bone marrow from the rats exposed to n-pentane at up to 7,000 ppm for 90 d (another experiment reported in the same paper). They reported no increase in micronuclei formation and no evidence of cytotoxicity in the bone marrow. Brooks et al. (1988) evaluated the potential for genetic toxicity with n- heptane by means of a bacterial mutation assay, a yeast assay for mitotic gene conversion, and cultured mammalian cells (rat liver and Chinese hamster ovary). n-Heptane did not produce a positive mutagenic response in any of the assays. Developmental and Reproductive Toxicity With respect to developmental toxicity, Hurtt and Kennedy (1999) ex- posed groups of eight pregnant female rats to n-pentane at 0, 1,000, 3,000, and 10,000 ppm by inhalation for 6 h/d from gestational days 6 to 15. They observed no maternal toxicity or adverse developmental effects on the fetuses. McKee et al. (1998) examined the reproductive tissues in the rats exposed to n-pentane as part of their 90-d inhalation study (exposures as high as 7,000 ppm) and found
100 SMACs for Selected Airborne Contaminants no differences in organ weight or pathology relative to controls. Although rele- vant reproductive and developmental studies for the other alkanes were not found (n-hexane was not included in this SMAC evaluation), the lack of adverse effects in n-pentane, the negative findings from mutagenicity and genotoxicity assays, and the rapid clearance of the C5-C9 alkanes from the body suggest that these are not likely to be critical toxicologic end points for these compounds. Interactions with Other Chemicals No specific data were found indicating that the C5-C9 saturated alkanes would interact with other compounds to modify their toxic effect. RATIONALE FOR ACCEPTABLE CONCENTRATIONS Table 6-4 presents occupational exposure limits set by other organizations for the C2-C9 saturated alkanes. In general, occupational exposure limits de- crease with increasing length of the carbon chain for this class of compounds. n- Hexane is a specific exception because lower limits are provided to address the potential for exposure to result in neurotoxicity. For the American Council of Governmental Industrial Hygienists (ACGIH) and the Occupational Safety and Health Administration (OSHA), the limits for the remaining C2-C9 saturated alkanes are set primarily to minimize their potential for irritation or narcosis. However, the National Institute for Occupational Safety and Health (NIOSH) takes a different approach and does not view n-hexane as the sole member of this chemical class with the potential to cause neuropathy. Russian PDK (Rus- sian acronym representing their long-term air quality guideline) values are also provided for comparison purposes. Table 6-5 includes the SMACs for the C2-C9 saturated alkanes established by the National Aeronautics and Space Administration through the evaluation presented in this document. For the C2-C4 alkane gases (ethane, propane, and butane), 10% of each chemicalâs lower explosive limit was established as the SMAC. As described earlier in this document, it was determined that these gases were not likely to result in toxicity in association with environmentally relevant exposures, and ensuring that levels of explosivity are not approached would be adequately protective of any direct toxicity associated with these compounds. The C2-C4 alkanes were not grouped with the C5-C9 alkanes for that reason. A literature review identified three general categories of effects potentially relevant to the C5-C9 alkanes: sensory irritation, CNS effects, and auditory tox- icity. These effects were further considered for acceptable concentration (AC) development and are discussed later in this document (for a summary of ACs, see Table 6-6). Because of the lack of individual studies for each C5-C9 alkane, data were not available for each effect. Accordingly, an attempt was made to use available data for conservative representatives for the C5-C9 group that could adequately represent C5-C9 alkane toxicity for each effect without underesti-
C2-C9 Alkanes 101 mating health risks. For example, studies suggest that sensory irritation may be a concern with the C5-C9 alkanes. Although sensory irritation data on every al- kane of interest were not available, the data that were available suggested that n- nonane was a conservative representative for the group because it was predicted to have the lowest threshold for sensory irritant effects. TABLE 6-4 Exposure Limits Set by Other Organizations ACGIH-TLVa,d OSHA PELb,d NIOSH RELb, d Russian PDKc ppm ppm ppm (360 d), ppm Ethane 1,000 â â â Propane 1,000 1,000 1,000 â Butane 1,000 â 800 â Pentane 600 1,000 120 3 n-Hexane 50 500 50 1 Other hexane 500 â 100 1 isomers 1,000 STEL/CEIL â 510 STEL/CEIL â Heptane 400 500 85 2 500 STEL/CEIL â 440 STEL/CEIL â Octane 300 500 75 â Nonane 200 â 200 â a Source: ACGIH 2008. b Source: NIOSH 2005. c International Space Station Medical Operations Requirements Document (MORD), Johnson Space Center 50620, National Aeronautics and Space Administration. d Value in column is a time-weighted average (TWA) unless otherwise specified. Note: Limits apply to all isomers unless otherwise specified. Abbreviations: â, not applicable; PEL, permissible exposure limit; REL, recommended exposure limit; STEL/CEIL, short-term exposure limit/ceiling; TLV, threshold limit value. Source: ACGIH 2008. Reprinted with permission; copyright 2008, American Conference of Industrial Hygienists. TABLE 6-5 Spacecraft Maximum Allowable Concentrations C2-C9 Alkanes (ppm) 1h 24 h 7d 30 d 180 d Ethanea 3,200 3,200 3,200 3,200 3,200 Propanea 2,300 2,300 2,300 2,300 2,300 Butanea 1,900 1,900 1,900 1,900 1,900 C5-C9b 150 80 60 20 3 (target (irritation, (irritation, (CNS effects) (ototoxicityc, (ototoxicityc) toxicity) CNS effects) CNS effects) CNS effects) a 10% of the LEL for this compound is set as the SMAC. b Includes all isomers and members except n-hexane, which has its own SMAC. c Overall ototoxicity risk will depend on noise levels in the specific environment, because hearing damage may be caused by both mechanical and chemical injury. Abbreviation: LEL, lower explosive limit.
TABLE 6-6 Acceptable Concentrations 102 Uncertainty Factors Acceptable Concentration (ppm) Space- End Point Data and Reference Species NOAEL Time Species flight 1h 24 h 7d 30 d 180 d Sensory n-Nonane as surrogate; CF-1 mice 1 1 3 1 150 150 â â â irritation 95% lower confidence limit on the RD10; 440 ppm, 0-10 min; Kristiansen and Nielsen 1988 CNS effects n-Nonane; 2,414 ppm Sprague- 1 1 (1 h) 10 1 240 80 â â â (NOAEL), 8 h; Nilsen et Dawley rats 8/24 h (24 al. 1988 h) n-Nonane; 590 ppm Wistar rats 1 1 10 1 â â 60 60 60 (NOAEL), 6 h/d, daily evaluation over 13-wk study; Carpenter et al. 1978 Auditory n-Heptane; 800 ppm Long 1 6/24 h 10 1 â â â 20 3 toxicity NOAEL, 6 h/d, 28 d; Evans rats 28/180 d Simonsen and Lund 1995 SMACs 150 80 60 20 3
C2-C9 Alkanes 103 Efforts were made to eliminate differences that might limit the usefulness and applicability of a group SMAC. Thus, n-hexane has its own SMAC in ac- cordance with the understanding that its neurotoxicity is not relevant to the other C5-C9 alkanes. Even with these efforts, however, it is clear that the remaining alkanes are not equal in terms of their potential for adverse health effects. Ac- cordingly, the measurement of an exceedance of the C5-C9 SMAC over a par- ticular time frame should prompt a closer examination of the specific alkane to ensure that any actions taken are commensurate with the actual risk that com- pound poses. For example, if the group SMAC is being applied to evaluate n- pentane measurements, slight exceedances of the SMAC would be viewed with less health significance than if n-nonane measurements were being evaluated. Although there are limitations inherent in establishing group SMACs, it can often be a useful and practical approach to risk assessment when reliable data are scanty for all group members but there is a reasonable basis and ability to predict toxicologic response within a grouping (Robinson 2000). For exam- ple, establishing a reference compound that can represent a broader group of chemicals plays a central role in some advocated approaches to risk assessment for total petroleum hydrocarbonfractions. The Massachusetts Department of Environmental Protection has recommended grouping C9-C17 alkanes and iden- tifying a reasonably conservative reference compound (in this case, n-nonane) for the purposes of establishing a reference dose (MA DEP 1994). Benchmark dose analysis was considered but was determined to be inap- propriate for the end points and studies evaluated. Specifically, the approach taken in developing sensory irritation ACs utilized a statistical procedure that precluded benchmark dose analysis. For other end points, critical studies did not include sufficient dosing groups to allow meaningful modeling of the dose- response relationship. Sensory Irritation The 1-h and 24-h ACs for sensory irritation were calculated for C5-C9 al- kanes based on the work of Kristiansen and Nielsen (1988). These sensory irrita- tion effects are evidenced as a burning sensation in the eyes, nose, or throat that is mediated by receptors on the trigeminal nerve endings within the nasal mu- cosa (Alarie 1981). The effect is measured by evaluating exposure concentra- tions and corresponding reductions in respiratory rate in exposed mice. Pre- dicted concentration-response relationships were calculated for n-heptane, n- octane, and n-nonane and were expressed in regression equations developed for each alkane. To use these study results in SMAC development, it was necessary to identify a specific degree of sensory irritation that is consistent with the level of irritation allowable for varying exposure durations. With guidance from the National Research Council committee, we used the 95% lower confidence limit on the RD10 (lower confidence limit on the concentration that corresponds to a 10% reduction in the respiratory rate relative to controls). We found this level of
104 SMACs for Selected Airborne Contaminants response to be appropriately conservative for use in evaluating short-term expo- sures and generally consistent with the guidelines used in establishing a point of departure for benchmark dose analysis (that is, a lower confidence limit on a specific response rate is identified). Among the alkanes evaluated, n-nonane produced an irritant response at the lowest concentration. Thus, we used n-nonane as a conservative representa- tive compound in setting ACs for sensory irritation for the alkanes. To better reflect sampling variability inherent in the RD10 estimate, we used a bootstrap analysis to generate a distribution of possible RD10 values for n-nonane, with 5th, 50th, and 95th percentile estimates of the RD10 of 440 ppm, 1,095 ppm, and 1,700 ppm, respectively (90% confidence interval of 440-1,700 ppm) (Efron and Tibshirani 1993). We applied the lower 95% confidence limit on the RD10 (440 ppm) in establishing the AC for this end point. Although these effects were measured in mice, the few human observa- tions available seem to correlate reasonably well with these predictions (Alarie 1981, Kristiansen and Nielsen 1988), and mice are not thought to underpredict the human health risks for sensory irritants (Buckley et al. 1984, Schaper 1993). Therefore, we applied an uncertainty factor (UF) of only 3 to account for neces- sary species extrapolation. It was not necessary to further adjust the RD10 esti- mate when determining a 24-h AC, because any sensory irritant effects are likely to be exhibited during the first few minutes of exposure and are not ex- pected to occur at progressively lower concentrations in conjunction with longer term exposures. Also, because n-nonane is being used as a basis for the AC, any short-term irritation would be extremely mild and would also be acceptable for 24 h of exposure. Sensory irritation: 1- and 24-h (AC sensory irritation) = 440 ppm (lower 95% confidence limit on the RD10) Ã 1/3 (species factor) = 147 ppm, rounded to 150 ppm CNS Effects Available studies addressing CNS effects related to the C5-C9 alkanes are relevant to the 1-h, 24-h, 7-d, 30-d, and 180-d ACs. With respect to the 1-h and 24-h ACs, Nilsen et al. (1988) reported CNS effects including ataxia, focal seizure, and spasms in rats exposed to 3,560 ppm, 4,438 ppm, and 5,280 ppm of n-nonane for 8 h. Effects were observed within 2- 4 h of exposure, and they proposed a NOAEL of 2,414 ppm. In calculating the ACs, a UF of 10 was applied to account for necessary species extrapolations. Although CNS effects are generally more dependent on the attainment of critical blood concentrations than on exposure duration, a time adjustment was applied for the 24-h AC to account for the possibility that enough time may not have passed during the 8-h study for the CNS effects to be fully realized (Zahlsen et al. 1990).
C2-C9 Alkanes 105 CNS: 1-h AC (CNS) = 2,414 ppm (NOAEL) Ã 1/10 (species factor) = 241 ppm, rounded to 240 ppm 24-h AC (CNS) = 2,414 ppm (NOAEL) Ã 1/10 (species factor) Ã 8 h/24 h (time extrapolation) = 80 ppm The CNS effects reported by Carpenter et al. (1978) after inhalation expo- sures to n-nonane were used as the basis for the 7-d, 30-d, and 180-d ACs. Rats exposed to 1,500 ppm of n-nonane (6 h/d) were observed to exhibit mild tremors and coordination loss. However, in a separate experiment described in the same paper, no such effects were reported in rats exposed to n-nonane at 590 ppm and evaluated daily over 13 wk (6 h/d, 5 d/wk), although rats in the 1,600-ppm ex- posure group did show such effects. Thus, 590 ppm was applied as a NOAEL for these CNS effects in developing the 7-d, 30-d, and 180-d ACs. A UF of 10 was applied to account for species extrapolation. Time adjustments were not considered to be necessary for these longer term ACs because CNS effects are generally expected to be more dependent on the attainment of critical blood con- centrations. This is supported by the observations of Carpenter et al. (1978) that CNS effects occurred within the first few days of exposure and that these effects did not appear to worsen with longer exposures. This is further corroborated by modeling predictions (Robinson 2000) and measured data on rat inhalation of n- nonane (Zahlsen et al. 1990). It was observed that peak brain concentrations of n-nonane, which are thought to be proportional to the potential for hydrocarbon- induced CNS effects (Baker et al. 1985), are likely attained within the first day of exposure. CNS: 7-d AC (CNS) = 590 ppm (NOAEL) Ã 1/10 (species factor) = 59 ppm, rounded to 60 ppm 30-d AC (CNS) = 590 ppm (NOAEL) Ã 1/10 (species factor) = 59 ppm, rounded to 60 ppm 180-d AC (CNS) =590 ppm (NOAEL) Ã 1/10 (species factor) = 59 ppm, rounded to 60 ppm Ototoxicity In a study of inhalation exposures of rats to n-heptane, Simonsen and Lund (1995) observed ototoxicity in rats as measured through changes in audi-
106 SMACs for Selected Airborne Contaminants tory brain stem responses. They exposed groups of rats to n-heptane at 0 ppm, 800 ppm, and 4,000 ppm for 6 h/d over 28 d. Rats in the 4,000-ppm exposure group experienced changes in auditory brain stem responses that corresponded to an approximate 10 decibel (dB) increase of the auditory threshold. Rats in the 800-ppm group did not experience significant differences relative to controls. Thus, 800 ppm was taken as a NOAEL for this effect, and 30-d and 180-d ACs were calculated. A UF of 10 was applied to account for species extrapolation. In addition, a time adjustment was made to account for the less-than-continuous exposure regimen (6 h/d). In extending the 28-d findings to calculate a 180-d AC, an additional time adjustment (180 d/28 d) was also applied to account for the shorter exposure. There is some uncertainty in terms of the need for this ad- justment for exposure duration. Simonsen and Lund (1995) noted that n-heptane is a relatively weak ototoxic agent that would likely act similarly to trichloro- ethylene and toluene (Pryor et al. 1984, Rebert et al. 1991), which generally exhibit a threshold in animal testing below which even long-term exposures do not result in ototoxicity. However, scanty data are available on the role of expo- sure duration in the ototoxicity of the saturated alkanes. Also, epidemiologic data from occupational cohorts suggest that length of exposure can be a contrib- uting factor to ototoxicity for certain solvents (Morata et al. 2002, Sliwinska- Kowalska et al. 2004), although confounding factors make it difficult to assess fully the relevancy of these findings in interpreting animal study results. Overall risk of hearing damage will depend on noise levels in the actual space environment because damage may be caused by both mechanical and chemical injury. Noise is an important aeromedical factor for both ISS and Shut- tle operations, and flight rules limit the amount of noise that can be tolerated for 24 h (65 dB for ISS and 74 dB for Shuttle). The more stringent noise require- ment for ISS is partially because of the longer period of crew habitation than for Shuttle. Measurements above these limits prompt mitigation efforts to reduce noise as well as precautionary actions such as the use of hearing protection. Shorter term noise limits have also been established to guide decision making (e.g., Shuttle flight rules state that noise shall not exceed 86 dBA for any time period). Although not specifically intended to address cumulative effects due to noise and chemical exposures, limiting noise levels plays an important role in minimizing the potential for auditory damage. Ototoxicity: 30-d AC (ototoxicity) = 800 ppm (NOAEL) Ã 1/10 (species factor) Ã 6 h /24 h (time extrapolation) = 20 ppm. 180-day AC (ototoxicity) = 800 ppm (NOAEL) Ã 1/10 (species factor) Ã [6 h/24 h Ã 28 d/180 d] (time extrapolation) = 3 ppm.
C2-C9 Alkanes 107 BOOTSTRAPPING THE SENSORY IRRITATION DATA This section describes the nonparametric bootstrap method used to esti- mate confidence intervals for the regression on the sensory irritation data for n- nonane reported by Kristiansen and Nielsen (1988). Bootstrapping represents a statistical analysis that allows for better characterization of the sampling vari- ability inherent in estimating an RD10 concentration (Efron and Tibshirani 1993). From the authors: % decrease relative to controls n-Nonane concentration (ppm) (0-10 min) 1,159 8.77 1,521 17.18 2,873 21.97 3,443 21.25 4,439 20.54 4,862 12.71 5,433 27.70 6,182 29.95 6,358 34.37 Starting with the linear model: y = b0 + b1 Ã log10(x). log (RD10) = log10(x) = (y â b0)/b1 = (10 â b0)/b1. The goal of the bootstrap method is to generate bootstrap sample data [x, y] from the original data, (xi, yi), i = 1, 2, ..., 9. Specifically, bootstrap sample data are data sets of the same size (n = 9) sampled from the original data set with replacement. An example of a bootstrap resample of the data could be as fol- lows: 1 (x3, y3) 2 (x1, y1) 3 (x4, y4) 4 (x7, y7) 5 (x7, y7) 6 (x2, y2) 7 (x3, y3) 8 (x2, y2) 9 (x5, y5)
108 SMACs for Selected Airborne Contaminants For each bootstrap sample data set, the linear model is refit to obtain b0, b1; hence, log10(x10). From 1,000 bootstrap samples, we acquire 10,000 corre- sponding estimates of log10(x 10). The quantiles of this empirical distribution for n-nonane are as follows: 5% 50% 95% log10(x10) 2.6 3.0 3.2 x10 440.5 1095.9 1723.5 The 90% bootstrap confidence interval of log RD10(x10) is [2.6, 3.2] and the 90% confidence interval of the RD10 is [440, 1723]. REFERENCES ACGIH (American Council of Governmental Industrial Hygienists). 1991. Documenta- tion of the Threshold Limit Values and Biological Exposure Indices, 6th Ed. American Council of Governmental Industrial Hygienists, Cincinnati, OH. ACGIH (American Council of Governmental Industrial Hygienists) 2008. Guide to Oc- cupational Exposure Values. American Council of Governmental Industrial Hy- gienists, Cincinnati, OH. Alarie, Y. 1973. Sensory irritation by airborne chemicals. CRC Crit. Rev. Toxicol. 2(3):299-363. Alarie, Y. 1981. Dose-response analysis in animal studies: Prediction of human re- sponses. Environ. Health Perspect. 42:9-13. Allerheiligen, S.R., T.M. Ludden, and R.F. Burk. 1987. The pharmacokinetics of pen- tane, a by-product of lipid peroxidation. Drug Metab. Dispos. 15(6):794-800. Aranyi, C., W.J. OâShea, C.A. Halder, C.E. Holdsworth, and B.Y. Cockrell. 1986. Ab- sence of hydrocarbon-induced nephropathy in rats exposed subchronically to vola- tile hydrocarbon mixtures pertinent to gasoline. Toxicol. Ind. Health 2(1):85-98. ATSDR (Agency for Toxic Substances and Disease Registry). 1999. Toxicological Pro- file for Hexane. Agency for Toxic Substances and Disease Registry, Atlanta, GA. Baker, E.L., T.J. Smith, and P.J. Landrigan. 1985. The neurotoxicity of industrial sol- vents: A review of the literature. Am. J. Ind. Med. 8(3):207-217. Brooks, T.M., A.L. Meyer, and D.H. Hutson. 1988. The genetic toxicology of some hy- drocarbon and oxygenated solvents. Mutagenesis 3(3):227-232. Buckley, L.A., X.Z. Jiang, R.A. James, K.T. Morgan, and C.S. Barrow. 1984. Respira- tory tract lesions induced by sensory irritants at the RD50 concentration. Toxicol. Appl. Pharmacol. 74(3):417-429. Carpenter, C.P., D.L. Geary, R.C. Myers, D.J. Nachreiner, L.J. Sullivan, and J.M. King. 1978. Petroleum hydrocarbon toxicity studies: XVII. Animal response to n-nonane vapor. Toxicol. Appl. Pharmacol. 44(1):53-61. Dahl, A.R. 1989. The fate of inhaled octane and the neprotoxicant, isooctane in rats. Toxicol. Appl. Pharmacol. 100(2):334-341. Daugherty, M.S., T.M. Ludden, and R.F. Burk. 1988. Metabolism of ethane and pentane to carbon dioxide by the rat. Drug Metab. Dispos. 16(5):666-671.
C2-C9 Alkanes 109 DiVincenzo, G.D., W.J. Krasavage, and J.L. OâDonoghue. 1980. Role of metabolism in hexacarbon neuropathy. Dev. Toxicol. Environ. Sci. 6:183-200. EEMA (European Agency for the Evaluation of Medicinal Products). 1995. Summary Report: Propane, n-butane, Isobutene. Committee for Veterinary Medicinal Prod- ucts, European Agency for the Evaluation of Medicinal Products, London. Efron, B, and R.J. Tibshirani. 1993. An Introduction to the Bootstrap. New York: Chap- man & Hall. Epstein, S.S., E. Arnold, J. Andrea, W. Bass, and Y. Bishop. 1972. Detection of chemical mutagens by dominant lethal assay in the mouse. Toxicol. Appl. Pharmacol. 23(2):288-325. Filser, J.G., G.Y. Csanady, W. Dietz, W. Kessler, P.E. Kreuzer, M. Richter, and A. Stormer. 1996. Comparative estimation of the neurotoxic risks of n-hexane and n- heptane in rats and humans based on the formation of the metabolites 2,5- hexanedione and 2,5-heptanedione. Pp. 1002-1023 in Biological Reactive Inter- mediates V: Basic Mechanistic Research in Toxicology and Human Risk Assess- ment, R. Snyder, J.J. Kocsis, I.G. Sipes, G.F. Kalf, D.J. Jollow, H. Greim, T.J. Monks, and C.M. Witmer, eds. New York: Plenum Press. Finkel, A. 1983. Aliphatic hydrocarbons. Pp. 245-255 in Hamilton and Hardyâs Industrial Toxicology, 4th Ed. London: John Wright PSG. Frank, H., T. Hintze, D. Bimboes, and H. Remmer. 1980. Monitoring lipid peroxidation by breath analysis: Endogenous hydrocarbons and their metabolic elimination. Toxicol. Appl. Pharmacol. 56(3):337-344. Frommer, U., V. Ullrich, and H. Staudinger. 1970. Hydroxylation of aliphatic compounds by liver microsomes. I. The distribution pattern of isomeric alcohols. Hoppe Sey- lers Z. Physiol. Chem. 351(8):903-912. Frontali, N., M.C. Amantini, A. Spagnolo, A.M. Guarcini, M.C. Saltari, F. Brugnone, and L. Perbellini. 1981. Experimental neurotoxicity and urinary metabolites of the C5- C7 aliphatic hydrocarbons used as glue solvent in shoe manufacturing. Clin. Toxi- col. 18(12):1357-1367. Galvin, J.B., and F. Marashi. 1999. n-Pentane. CAS No. 109-66-0. J. Toxicol. Environ. Health A 58(1-2):35-56. Graham, D.G., and M.B. Abou-Donia. 1980. Studies of the molecular pathogenesis of hexane neuropathy. 1. Evaluation of the inhibition of glyceraldehydes-3-phosphate dehydrogenase by 2,5-hexanedione. J. Toxicol. Environ. Health 6(3):621-631. Halder, C.A., G.S. Van Gorp, N.S. Hatoum, and T.M. Warne. 1986. Gasoline vapor exposures. Part II. Evaluation of the nephrotoxicity of the major C4/C5 hydrocar- bon components. Am. Ind. Hyg. Assoc. J. 47(3):173-175. Hau, K.M., D.W. Connell, and B.J. Richardson. 2002. A study of the biological partition- ing behavior of n-alkanes and n-alkanols in causing anesthetic effects. Regul. Toxicol. Pharmacol. 35(2 Pt. 1):273-279. Hawas, O., D. Hawker, A. Chan, D. Cohen, E. Christensen, G. Golding, and P. Voweles. 2001. Characterization and identification of sources of VOCs in an industrial area of Brisbane. Australian Nuclear Science and Technology Organization Publication. Hurtt, M.E., and G.L. Kennedy. 1999. A limited developmental toxicity study of pentane by inhalation in the rat. Food Chem. Toxicol. 37(5):565-567. Katzenstein, A.S., L.A. Doezema, I.J. Simpson, D.R. Blake, and F.S. Rowland. 2003. Extensive regional atmospheric hydrocarbon pollution in the southwestern United States. Proc. Natl. Acad. Sci. 100(21):11975-11979. Kirwin, C., W. Thomas, and V. Simmon. 1980. In vitro microbiological mutagenicity studies of hydrocarbon propellants. J. Soc. Cosmet. Chem. 31:367-370.
110 SMACs for Selected Airborne Contaminants Kristiansen, U., and G.D. Nielsen. 1988. Activation of the sensory irritant receptor by C70- C11 n-alkanes. Arch. Toxicol. 61(6):419-425. Loury, D.J., T. Smith-Oliver, and B.E. Butterworth. 1987. Assessment of the covalent binding potential of 2,2,4-trimethylpentane to rat alpha2u-globulin. Toxicol. Appl. Pharmacol. 88(1):44-56. MA DEP (Massachusetts Department of Environmental Protection). 1994. Interim Final Petroleum Policy: Development of Health-Based Alterative to the Total Petroleum Hydrocarbon Parameter. Massachusetts Department of Environmental Protection, Boston, MA. McKee, R., E. Frank, J. Heath, D. Owen, R. Przygoda, G. Trimmer, and F. Whitman. 1998. Toxicology of n-pentane (CAS No. 109-66-0). J. Appl. Toxicol. 18(6):431- 442. Morata, T.C., A.C. Johnson, P. Nylen, E.B. Svensson, J. Cheng, E.F. Krieg, A.C. Lindblad, L. Ernstgard, and J. Franks. 2002. Audiometric findings in workers ex- posed to low levels of styrene and noise. J. Occup. Environ. Med. 44(9):806-814. Nilsen, O.G., O.A. Haugen, K. Zahlsen, J. Halgunset, A. Helseth, H. Aarset, and I. Eide. 1988. Toxicity of n-C9 to n-C13 alkanes in the rat on short-term inhalation. Pharmacol. Toxicol. 62(5):259-266. NIOSH (National Institute for Occupational Safety and Health). 2005. NIOSH Pocket Guide to Chemical Hazards: Carbon dioxide. NIOSH Publication No. 2005-149. National Institute for Occupational Safety and Health [online]. Available: http://www.cdc.gov/niosh/npg/npgd0103.html [accessed August 26, 2008]. Perbellini, L., F. Brugnone, D. Caretta, and G. Maranelli. 1985. Partition coefficients of some industrial aliphatic hydrocarbons (C5-C7) in blood and human tissues. Br. J. Ind. Med. 42(3):162-167. Pryor, G.T., C.S. Rebert, J. Dickinson, and E.M. Feeney. 1984. Factors affecting toluene- induced ototoxicity in rats. Neurobehav. Toxicol. Teratol. 6(3):223-238. Rebert, C.S., V.L. Day, M. Matteucci, and G.T. Pryor. 1991. Sensory-evoked potentials in rats chronically exposed to trichloroethylene: Predominant auditory dysfunction. Neurotoxicol. Teratol. 13(1):83-90. Robinson, P. 2000. Pharmacokinetic Modeling of JP-8 Jet Fuel Components. I. Nonane and C9-C12 Aliphatic Components. AFRL-HE-WP-TR-2000-0046. U.S. Air Force Research Laboratory, Wright Patterson AFB OH. Sandmeyer, E.E. 1981. Aliphatic hydrocarbons. Pp. 3175-3220 in Pattyâs Industrial Hy- giene and Toxicology, Volume II B, 3rd Rev. Ed., G.D. Clayton, and F.E. Clayton, eds. New York: John Wiley & Sohn. Schaper, M. 1993. Development of a database for sensory irritants and its use in estab- lishing occupational exposure limits. Am. Ind. Hyg. Assoc. J. 54(9):488-544. Schaumburg, H.H., and P.S. Spencer. 1976. Degeneration in central and peripheral nerv- ous systems produced by pure n-hexane: An experimental study. Brain 99(2):183- 192. Short, B.G., V.L. Burnett, and J.A. Swenberg. 1986. Histopathology and cell proliferation induced by 2, 2, 4-trimethylpentane in the male rat kidney. Toxicol. Pathol. 14(2):194-203. Simonsen, L., and S.P. Lund. 1995. Four week inhalation exposure to n-hexane causes loss of auditory sensitivity in rats. Pharmacol. Toxicol. 76(1):41-46. Sliwinska-Kowalska, M., E. Zamyslowska-Szmytke, W. Szymczak, P. Kotylo, M. Fiszer, W. Wesolowski, M. Pawlaczyk-Luszczynska, M. Bak, and A Gajda-Szadkowska. 2004. Effects of coexposure to noise and mixture of organic solvents on hearing in dockyard workers. J. Occup. Environ. Med. 46(1):30-38.
C2-C9 Alkanes 111 Spencer, P.S., H.H. Schaumberg, M.I. Sabri, and B. Veronesi. 1980. The enlarging view of hexacarbon neurotoxicity. Crit. Rev. Toxicol. 7(4):279-356. Stadler, J.C., A.J. OâNeill, G.S. Elliot, and G.L. Kennedy. 2001. Repeated exposure inha- lation study of pentane in rats. Drug Chem. Toxicol. 24(2):75-86. Swann, H.E., B.K. Kwon, G.K. Hogan, and W.M. Snellings. 1974. Acute inhalation toxi- cology of volatile hydrocarbons. Am. Ind. Hyg. Assoc. J. 35(9):511-518. Takeuchi, Y., Y. Ono, N. Hisanaga, J. Kitoh, and Y. Sugiura. 1980. A comparative study of the neurotoxicity of n-pentane, n-hexane, and n-heptane in the rat. Br. J. Ind. Med. 37(3):241-247. Truhaut, R., P. Laget, G. Piat, P.L. Nguyen, H. Dutertre-Catella, V.N. Huyen, F. FrÃ©dÃ©ric, and E. Shechter 1973. Preliminary electrophysiologic results following experimen- tal poisoning with technical hexane and heptane in white rats [in French]. Arch. Mal. Prof. Med. Trav. Secur. Soc. 34(7-8):417-426. Veulemans, H., E. Van Vlem, H. Janssens, R. Masschelein, and A. Leplat. 1982. Experi- mental human exposure to n-hexane. Study of the respiratory uptake and elimina- tion, and of n-hexane concentrations in peripheral venous blood. Int. Arch. Occup. Environ. Health 49(3-4):251-263. Yamada, S. 1967. Polyneuritis in workers exposed to n-hexane, its causes and symptoms [in Japanese]. Jpn. J. Ind. Health 9:651-659. Yano, B.L., D.A. Dittenber, R.R. Albee, and J.L. Mattsson. 1992. Abnormal auditory brainstem responses and cochlear pathology in rats induced by an exaggerated sty- rene exposure regimen. Toxicol. Pathol. 20(1):1-6. Zahlsen, K, A.M. Nilsen, I. Eide, and O.G. Nilsen. 1990. Accumulation and distribution of aliphatic (n-nonane) aromatic (1,2,4-trimethylbenzene) and naphthenic (1,2,4, trimethylcyclohexane) hydrocarbons in the rat after repeated inhalation. Pharma- col. Toxicol. 67(5):436-440.