Emergency and Continuous Exposure Guidance Levels for Selected Submarine Contaminants: Volume 3 (2009)

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.

4 Hydrogen Fluoride This chapter summarizes the relevant epidemiologic and toxicologic stud- ies of hydrogen fluoride. It presents selected chemical and physical properties, toxicokinetic and mechanistic data, and inhalation-exposure levels from the Na- tional Research Council and other agencies. The committee considered all that information in its evaluation of the U.S. Navy’s 1-h, 24-h, and 90-day exposure guidance levels for hydrogen fluoride. The committee’s recommendations for hydrogen fluoride exposure levels are provided at the end of this chapter with a discussion of the adequacy of the data for defining the levels and the research needed to fill the remaining data gaps. PHYSICAL AND CHEMICAL PROPERTIES Hydrogen fluoride is a corrosive, colorless gas that may fume in air (Budavari et al. 1989). Odor thresholds have been reported to range from 0.04 to 3 ppm (HSDB 2008). Like hydrogen chloride, hydrogen fluoride is highly solu- ble in water. Hydrofluoric acid is the term used to describe aqueous solutions of hydrogen fluoride. Selected physical and chemical properties are shown in Table 4-1. OCCURRENCE AND USE Hydrogen fluoride is used primarily to produce aluminum fluoride, syn- thetic cryolite, fluoropolymers, and chlorofluorocarbons (Lindahl and Mahmood 2005). It is also used in inorganic fluoride production, uranium enrichment, and fluorine production. Fluoride is found in some foods and beverages, particularly fish, seafood, gelatin, and tea; and many public water sources are fluoridated 70 

Hydrogen Fluoride 71 TABLE 4-1 Physical and Chemical Properties of Hydrogen Fluoride Synonyms Anhydrous hydrofluoric acid CAS registry number 7664-39-3 Molecular formula HF Molecular weight 20.01 Boiling point 19.51°C Melting point –83.55°C Flash point NA Explosive limits NA Specific gravity 1.002 at 0°C/4°C Vapor pressure 917 mmHg at 25°C Solubility Very soluble in water and alcohol; slightly soluble in ether; soluble in many organic solvents Conversion factors 1 ppm = 0.82 mg/m3; 1 mg/m3 = 1.22 ppm Abbreviation: NA, not available or not applicable. Sources: Budavari et al. (1989) and HSDB (2008). (ATSDR 2003). Ambient air concentrations of hydrogen fluoride are typically below the detection limit, although concentrations may be higher near industrial facilities that use or produce hydrogen fluoride (ATSDR 2003). Hydrogen fluoride has been measured on board submarines. NRC (1988) listed hydrogen fluoride as a potential contaminant of submarine air and re- ported a concentration of 0.3 ppm. No information was provided on sampling protocol, location, operations, or duration. Trials conducted on three nuclear- powered attack submarines did not detect hydrogen fluoride; the level of detec- tion was about 3 ppb (Hagar 2008). Whether the reported results are representa- tive of the submarine fleet is not known; few details were provided about the conditions on the submarines when the samples were taken. No other exposure data were located. Hydrogen fluoride emissions aboard submarines are thought to arise from decomposition of halogenated hydrocarbons and refrigerants (Hagar 2008). SUMMARY OF TOXICITY Hydrogen fluoride and its aqueous solutions present an acute hazard by inhalation or dermal exposure. The primary target of airborne gaseous hydrogen fluoride is the respiratory tract; however, injury to distant organs may also occur because of absorption of substantial amounts of fluoride. Acute effects of hydro- fluoric acid include damage to skin and lungs, including severe burns, and sys- 

72 Exposure Guidance Levels for Selected Submarine Contaminants temic effects, such as cardiac arrhythmias and acute renal failure (see, for exam- ple, Sanz-Gallén et al. 2001; Björnhagen et al. 2003; Horton et al. 2004; Hol- stege et al. 2005; Mitsui et al. 2007; Vohra et al. 2008). Some of the systemic effects may be due to depletion of calcium and magnesium or hyperkalemia. The critical effects of inhalation exposure to hydrogen fluoride are respira- tory tract irritation and the induction of respiratory disease. Respiratory tract irritation is documented in animal models and has been observed in controlled human exposure studies. Marked sensory irritation can occur at exposures greater than 3 ppm for 1 h (Lund et al. 1997). Prolonged respiratory tract effects can occur after short-term exposure. To evaluate longer-term exposures or sys- temic effects, the total fluoride intake from all exposure routes (inhalation, der- mal, and ingestion) must be considered (EPA 1988; NRC 2006). Chronic expo- sure to hydrogen fluoride (with particulate fluorides) in the aluminum industry is associated with increased risk of asthma (Taiwo et al. 2006). The literature on the systemic toxicity of fluoride is voluminous and is not addressed in full detail here. NRC (2006) recently reviewed fluoride toxicity with an emphasis on chronic toxicity. Fluoride-induced effects include hormonal disturbances; renal damage; reproductive toxicity; skeletal changes, including fluorosis; and possible genotoxicity and cancer. Effects in Humans Accidental Exposures Several case reports of death after acute accidental exposure to hydrogen fluoride are available and have been extensively reviewed by ATSDR (2003) and NRC (2004). Most of the reports stem from accidents involving spills of hydrofluoric acid. Because of its high volatility, inhalation exposure to hydrogen fluoride results from spills of hydrofluoric acid. The degree to which hydroflu- oric acid-induced burns or dermal absorption of fluoride may have contributed to the death is not known. Nonetheless, the case reports indicate that lung injury, including pulmonary edema (with or without hemorrhage), is common after such accidents. An informative case report describes the delayed and prolonged chemical pneumonitis that developed in a woman after use of large amounts of 8% hydro- fluoric acid as a cleaner in an unventilated bathroom (Bennion and Franzblau 1997). Airborne hydrogen fluoride concentrations were unknown. Symptoms developed slowly in the days after the exposure and eventually necessitated oxygen therapy (100% O2; 10 cm H2O peak end-expiratory pressure) because of hypoxemia. Chest radiography indicated a lung infiltrate, and signs included diffuse rhonchi and wheezing in both lungs. Another case report describes a woman who used 8-9% hydrofluoric acid as a cleaner in a ventilated bathroom (Franzblau and Sahakian 2003). It was estimated that hydrogen fluoride in the bathroom may have exceeded 170 ppm. 

Hydrogen Fluoride 73 She developed breathing problems, such as persistent wheezing and difficulty taking a deep breath, over the 1-2 months after exposure. Examination at that time revealed a mild obstructive pattern. Several months after the exposure, she was diagnosed with reactive airways dysfunction syndrome (RADS); her inter- mittent wheezing on exertion persisted for at least 3 years. An industrial accident in Texas in 1987 resulted in the release of 24,000 kg of hydrogen fluoride and about 3,000 kg of isobutane over a small commu- nity (population, 41,000; Wing et al. 1991). The airborne hydrogen fluoride concentration 1 h after the accident was reported to be 10 ppm; 2 h after the ac- cident, concentrations were “minimal.” The report indicates that air sampling was performed at those times but provides no information on the analytic meth- ods used to determine hydrogen fluoride concentrations. A total of 939 people sought emergency care; common symptoms were eye irritation, throat irritation (burning), headache, and shortness of breath. Of those who sought care, 94 were hospitalized. Forced expiratory volume in 1 s (FEV1) was less than 80% of pre- dicted in one-third of the people who sought medical care and were not hospital- ized compared with half the people who were hospitalized. A follow-up study revealed that respiratory symptoms persisted in some people for at least 2 years, although much reduced (Dayal et al. 1992). The degree to which psychologic factors influenced the symptoms is unknown, but it is thought that the symptoms could not be explained entirely on the basis of psychologic stress (Dayal et al. 1994). In summary, respiratory tract injury appears to be the predominant re- sponse to accidental exposure to hydrogen fluoride. Respiratory tract effects include irritation, airway obstruction (as assessed with FEV1), and airway in- flammation. Upper airway symptoms may have occurred in some situations and gone unreported because they were overshadowed by the lower airway effects. There are suggestions that long-term respiratory tract effects may occur after exposure to hydrogen fluoride at high concentrations as indicated by the devel- opment of RADS in one subject and the presence of persistent respiratory symp- toms in the general population after the release of hydrogen fluoride during an industrial accident. The studies indicate that the respiratory tract may be a criti- cal target of hydrogen fluoride in the general population but do not provide in- formation on concentration-response relationships. Fluoride ion is rapidly and efficiently absorbed into the circulation after inhalation of hydrogen fluoride or airborne fluorides as indicated by increased blood or urinary fluoride concentrations (see, for example, Collings et al. 1951, 1952; Largent et al. 1951). Therefore, the possibility of systemic fluoride- induced injury after accidental exposure to hydrogen fluoride is important to consider. Little information is available on systemic effects after accidental in- halation exposure to fluoride, but accidental ingestion has been followed by itch- ing, rash, gastrointestinal symptoms, and numbing or tingling of extremities or the face (reviewed by NRC 2006). 

74 Exposure Guidance Levels for Selected Submarine Contaminants Experimental Studies Upper Airway Irritation There are several published controlled studies of short-term inhalation ex- posure to hydrogen fluoride in humans (Table 4-2). The exposure durations in the studies spanned from 1 min to 6 h/day for multiple days. All studies report upper airway irritation as the predominant symptom. The degree of upper airway irritation was reported as intolerable at 122 ppm for more than 1 min, marked at 61 ppm for several minutes, and mild at 32 ppm for several minutes (Machle et al. 1934). Lund et al. (1997) described a 1-h exposure with exercise at low concen- tration (0.2-0.7 ppm), intermediate concentration (0.9-2.9 ppm), and high con- centration (3.1-6.3 ppm). Exercise consisted of a fixed workload of 75 W on a bicycle ergometer for the last 15 min of exposure. There were no air-exposed control subjects, but baseline reporting of symptoms was conducted for all sub- jects before exposure. Subjects were men, 21-44 years old; persons with asthma or recent respiratory tract infection were excluded from the study, but the study group did include people with “hay fever.” Symptoms, including upper airway (nose or throat) itching and soreness, were reported during exposure on a scale of 0-5 (1 was very mild, and 5 was severe). More detail was not provided on the scaling; the authors report ratings of 1-3 as representing a “low” degree of irrita- tion and greater than 3 as representing a “high” degree of irritation. It seems reasonable to assume that low corresponds to mild irritation and high corre- sponds to moderate to marked irritation. In the low-concentration group, four of nine subjects reported mild upper airway irritation. In the intermediate- concentration group, six of seven reported mild irritation. In the high- concentration group, three of seven reported moderate to severe irritation, and the other four reported mild irritation. Thus, a clear concentration-response rela- tionship was observed in the study. Only mild irritation was reported at concen- trations as high as 2.9 ppm, whereas marked irritation was reported in some sub- jects at concentrations as low as 3.1 ppm. Thus, 3 ppm appears to reflect the demarcation between minimal and marked irritation, at least as determined by the small number of subjects in this study. In a later study with high concentra- tion (4.0-4.8 ppm), six of 10 subjects reported mild irritation, and one of 10 re- ported marked irritation (Lund et al. 2002)—essentially the same response pat- tern observed in their earlier study (Lund et al. 1997). There were no air- exposed control subjects, but baseline reporting of symptoms was conducted for all subjects before exposure. Largent (1961) performed a study with multiple 6-h exposures to hydro- gen fluoride 5 days/week for a total of 10-50 exposures. Again, upper airway irritation was experienced. One subject exposed at 1.4 ppm reported no symp- toms, and all five subjects exposed at 2.6-4.7 ppm reported the perception of 

TABLE 4-2 Effects of Hydrogen Fluoride in Controlled Human Studies Concentration (ppm) Duration Subjects and Effects Reference 32, 61,122 1 to several min Two healthy subjects, smoking status unknown Machle et al. Maximum tolerable level (1 min) at 122 ppm; marked conjunctival, nasal, and large 1934 airway irritation at 61 ppm (several minutes); mild conjunctival, nasal, and large airway irritation at 32 ppm (several minutes); sour taste detected at all concentrations 1.4, 2.6-4.7 6 h/day Five healthy subject, smoking status unknown Largent 1961 5 days/week No reported airway irritation in one subject exposed at average of 1.4 ppm; slight 10-50 days cutaneous (facial), ocular, and nasal irritation in all subjects at average concentration of 2.6-4.7 ppm daily for a total of 10-50 days; cutaneous erythema frequent (requiring face-cream application); increased symptoms in one subject who developed an upper respiratory tract infection during the protocol 0.2-6.3 1 h (with Twenty healthy, nonsmoking men (21-44 years old); persons with airway infection Lund et al. exercise) or history of asthma excluded; three exposure groups (low, 0.2-0.7 ppm; middle, 1997 0.9-2.9 ppm; high, 3.1-6.3 ppm); symptom scores reported as “low,” presumably mild (1-3 on scale of 0-5), and “high,” presumably moderate to marked (>3 on scale of 0-5). Mild upper airway irritation was reported in four of nine subjects in low- concentration group, six of seven in middle concentration group, four of seven in high-concentration group; moderate to severe upper airway irritation reported in three of seven in high-concentration group; mild lower airway irritation reported in two of seven and moderate to severe irritation in one of seven in high-concentration group; eye irritation (mild) reported in two subjects in each concentration group (Continued) 75 

76 TABLE 4-2 Continued Concentration (ppm) Duration Subjects and Effects Reference 0.2-6.3 1 h (with Nineteen healthy, nonsmoking men (21-44 years old); persons with airway Lund et al. exercise) infection or history of asthma excluded; three exposure groups (low, 0.2-0.7 ppm; 1999 middle, 0.9-2.9 ppm; high, 3.1-6.3 ppm). Publication presumably provides BAL results in subjects described in Lund et al. (1997); BAL performed 24 h after exposure; percentage of lymphocytes increased in both bronchial and bronchoalveolar portions of BAL with no apparent concentration-response relationship; increases appeared to be present in middle- and high-concentration groups; no observed changes in any other cell type; myeloperoxidase increased in bronchial portion of BAL with no apparent concentration-response relationship 4.0-4.8 1 h (with Ten healthy, nonsmoking men (21-44 years old); persons with airway infection or Lund et al. exercise) history of asthma excluded; all exposed at 4.0-4.8 ppm for 1 h with exercise; nasal 2002 irritation symptoms reported as “low,” presumably mild (1-3 on scale of 0-5), and “high,” presumably moderate to marked (>3 on a scale of 0-5). Mild irritation reported by six of 10 and marked irritation by one of 10 subjects; nasal lavage performed immediately and 90 min after exposure; lavage neutrophil count and lavage proteins (TNF-α, PGE2, LTB4, peptide LT) significantly increased; symptom score and lavage neutrophil counts correlated with each other 4.0-4.8 1 h (with Ten healthy, nonsmoking men (21-44 years old); persons with airway infection or Lund et al. exercise) history of asthma excluded; all exposed at 4.0-4.8 ppm for 1 h with exercise; nasal 2005 symptomology not measured; BAL performed 2 h after exposure No increase observed in differential cell count or in numerous mediators (interleukins, myeloperoxidase, eicosanoids, others); several significantly decreased. Abbreviations: BAL, bronchoalveolar lavage; LTB4, leukotriene B4; PGE2, prostaglandin E2; TNF-α, tumor-necrosis factor-alpha; peptide LT, peptide leukotriene. 

Hydrogen Fluoride 77 slight irritation. One subject developed an “upper airway cold” during the proto- col, at which time exposure at 3.4 ppm produced “considerable discomfort.” All subjects completed the multiple-exposure regimen—an indication that the de- gree of irritation was not sufficient to cause withdrawal from the study. Al- though a quantitative scaling of symptoms was not reported, comparison of the data with the symptoms reported in the studies of Lund et al. (1997, 2002) sug- gests that repeated exposure to hydrogen fluoride does not result in exacerbation of the irritation response and may actually lead to some degree of habituation. In summary, symptoms of upper airway irritation were uniformly reported in clinical studies. The threshold for mild irritation may be 0.5 ppm or less in some people. Given that the studies used small numbers of subjects, the database suggests that a significant fraction of subjects experience moderate to marked irritation at concentrations over 3 ppm but only mild irritation at lower concen- trations. Lower Airway Irritation Lower airway irritation has been reported in human subjects exposed to hydrogen fluoride but is generally of less magnitude than upper airway irritation. In the intermediate-concentration group (0.9-2.9 ppm) of the study of Lund et al. (1997), one of seven subjects reported mild lower airway irritation (chest tight- ness and soreness, coughing, expectoration, or wheezing) during the 1-h expo- sure (compared with six of seven reporting mild upper airway irritation). In the high-concentration group (3.1-6.3 ppm), two of seven reported mild lower air- way symptoms, and one of seven reported moderate to marked lower airway symptoms compared with three of seven reporting upper airway symptoms of this degree. A concentration-response relationship may be apparent, but the changes in the lower airway symptoms did not achieve statistical significance, and this led the study authors to conclude that lower airway symptoms were not reported to a significant degree in relation to exposure to hydrogen fluoride. The study design included forced expiration to assess lower airway physiologic changes. No consistent change was observed in forced vital capacity (FVC) or FEV1. Thus, mild symptoms of lower airway irritation occur at exposures as high as 2.9 ppm, and more marked symptoms may occur in some people ex- posed at higher concentrations, but such changes occurred in the absence of al- terations in airway function as assessed by forced expiration. Airway Inflammation Hydrogen fluoride exposure for 1 h results in airway inflammation as as- sessed by increases in inflammatory cells in nasal lavage or bronchoalveolar lavage (BAL) fluid. Exposure at 4.4 ppm for 1 h (range, 4.0-4.8) results in a significant increase in nasal lavage neutrophils and proinflammatory mediators, 

78 Exposure Guidance Levels for Selected Submarine Contaminants including tumor-necrosis factor-alpha, prostaglandin E2, and leukotriene B4 (Lund et al. 2002). Although hydrogen fluoride clearly induced upper airway inflammation, the exposure was not debilitating, nor would any long-term ef- fects be expected to result from a response of this nature. Exposure to hydrogen fluoride for 1 h also results in inflammatory cell changes in the lower airways as assessed by BAL (Lund et al. 1999). A significant correlation between increases in BAL lymphocyte numbers (but not neutrophil or eosinophil numbers) and increased exposure concentrations was observed 24 h after exposure (Lund et al. 1999, which involved the same subjects described in Lund et al. 1997). It is dif- ficult to discern precisely the concentration-response relationships from the data presented, but apparently no alteration occurred in the low-concentration (0.2- 0.7 ppm) group; increases in BAL lymphocyte number occurred only in the in- termediate (0.9-2.9 ppm) and high (3.1-6.3 ppm) groups. The changes were ob- served 24 h but not 2 h after the 1-h exposure (Lund et al. 1999, 2005). Al- though lavage neutrophil numbers were not significantly increased, a slight increase in the myeloperoxidase content in the bronchial portion of the BAL fluid was observed; this suggests that subtle recruitment or activation of neutro- phils occurred. The absence of an overt increase in neutrophils or overt symp- toms suggests that the responses would not result in short-term or long-term health impairment. Other Irritation Effects Cutaneous irritation and ocular irritation have been reported in subjects exposed to hydrogen fluoride. In the study of Machle et al. (1934), two subjects exposed to hydrogen fluoride at 32 ppm or higher for several minutes reported cutaneous, ocular, and respiratory tract irritation. Ocular irritation was reported during 1-h exposures in the study of Lund et al. (1997), but the degree of ocular irritation was less than that of upper airway irritation. Largent (1961) used mul- tiple 6-h exposures and found that cutaneous irritation was experienced at 2.6- 4.7 ppm. Subjects applied cream to alleviate symptoms. Cutaneous erythema was common in the five subjects although reported to be without discomfort. One subject experienced peeling of the skin in the third week of exposure. It should be noted that the sour, pungent taste of hydrogen fluoride can be detected during exposure at above 3 ppm. Amoore and Hautala (1983) reported the odor threshold at below 1 ppm. Systemic Effects Exposure to hydrogen fluoride or other airborne fluorides may result in absorption of fluoride ion (see, for example, Collings et al. 1951, 1952; Largent et al. 1951; Waldbott and Lee 1978); thus, the potential for fluoride-induced systemic effects should be considered. For example, given a ventilation rate of 

Hydrogen Fluoride 79 15 m3/day (EPA 1997) for a 70-kg man and 100% deposition and absorption, a 1-h exposure to hydrogen fluoride at 3 ppm results in systemic absorption of 1.5 mg of fluoride (0.02 mg/kg).1 Few experimental studies involving human expo- sure to hydrogen fluoride or inhaled fluorides are available; experimental studies and case reports involving ingestion of fluoride have reported antithyroid effects (0.03-0.14 mg/kg-day for 20-245 days; Galletti and Joyet 1958) and hypersensi- tivity or reduced tolerance to fluoride (0.02 mg/kg-day for short-term exposures; Grimbergen 1974; Waldbott 1956, 1958). One study reported a threshold of hydrogen fluoride for the light-adaptive reflex2 of 0.04 ppm (Sadilova et al. 1965) in three subjects exposed to hydrogen fluoride at 0.02, 0.04, or 0.07 ppm (exposure duration not available), but it is difficult to evaluate the underlying experimental work (Smith and Hodge 1979), and the toxicologic relevance of the response is unknown (ATSDR 2003). The absorbed doses at those exposures are likely to be so low that it is difficult to attribute the neurologic effect to absorbed fluoride itself. Occupational and Epidemiologic Studies A variety of occupational and epidemiologic studies of airborne and in- gested fluoride have been conducted; many of them have been reviewed by NRC (2006) and ATSDR (2003). The following paragraphs discuss respiratory symptoms (asthma), renal damage, endocrine effects, increased risk of bone fracture, and bone and joint pain (skeletal fluorosis). In the workplace, exposure to hydrogen fluoride rarely occurs in the absence of exposure to other particu- lates (such as calcium fluoride [CaF2] or sodium aluminum fluoride [Na3AlF6]) or gaseous fluoride-containing materials (such as tetrafluorosilane [SiF4]); this confounds interpretation of the results with respect to the effects of hydrogen fluoride. That is particularly true of possible effects of systemic fluoride absorp- tion because the source of the fluoride is not known with certainty. Many studies of worker health in the aluminum industry have found an as- sociation between occupational exposure to fluoride in aluminum “potrooms” and respiratory disease or asthma (see, for example, Kaltreider et al. 1972; Soy- seth and Kongerud 1992; Kongerud et al. 1994). Most studies, however, did not reveal potential etiologic agents. In aluminum potrooms, workers are exposed to particulate fluoride, gaseous fluoride (presumably hydrogen fluoride), sulfur dioxide, and other irritants. A recent study of the health of workers in the alumi- num industry suggests that exposure to airborne fluorides is associated with an increased incidence of asthma (Taiwo et al. 2006). Analysis of records on 1 The calculation is as follows: (15 m3/day)(1 day/24 h)(3 ppm)(0.82 mg/m3 per ppm)(19/20 mg fluoride per mg hydrogen fluoride) = 1.5 mg of fluoride for 1-h exposure, assuming 100% absorption. 2 The light-adaptive reflex is defined as reflex changes in ocular sensitivity to light based on dark adaptation. It is measured as a marker of neurologic effects. 

80 Exposure Guidance Levels for Selected Submarine Contaminants 12,000 workers (about 10% of whom worked in potrooms) found an increased risk of asthma in workers exposed to gaseous fluoride at 0.27 ± 0.53 ppm (mean ± SD) for an average of 16 ± 10.8 years. The study included only people who had a new diagnosis of asthma after two or more asthma-free years in the work- place; thus, anyone who developed occupationally related asthma in the first 2 years of employment was excluded. The average age of the potroom workers was 43.7 ± 10.1 years. Using a multivariate generalized linear model relating the natural logarithm of predicted asthma rate, the authors concluded that asthma risk was significantly associated with exposure to hydrogen fluoride and current smoking but not to other contaminants in the workplace, such as particulate fluoride and sulfur dioxide. The relative risk for development of asthma was estimated by the model to be 1.18 per 0.1 mg/m3 change in hydrogen fluoride (95% confidence interval [CI], 1.09-1.3), which corresponds to a relative risk of 1.18 per 0.12 ppm. Although documentation of an association does not indicate cause and effect, a 1-h exposure to hydrogen fluoride does cause increased BAL lymphocytes (Lund et al. 1999), and persistent respiratory symptoms were re- ported in the general community after exposure to hydrogen fluoride in an in- dustrial accident (Wing et al. 1991). Those facts raise concern that the increased incidence of asthma in potroom workers may reflect a response to hydrogen fluoride. It is important to note that the presence of high dust concentrations and other irritants and occasional short-term (15-min) high-exposure excursions to hydrogen fluoride may have contributed to the response (Taiwo et al. 2006). Waldbott and Lee (1978) reported a case of systemic fluoride toxicity from repeated exposures to hydrogen fluoride gas in the alkylation unit of an oil company. Estimated exposures over the worker’s 10 years of employment were often above 3 ppm, on the basis of odor detection, and were thought to have been very high (25-200 ppm) during some procedures. Chronic symptoms in- cluded reduction in pulmonary function, gastrointestinal problems, and severe back and leg pains. Fluoride measured in bone 10 years after the maximal expo- sures was significantly above normal. Given a hydrogen fluoride concentration of 3 ppm, an 8-h workday, a ventilation rate of 15 m3/day, and complete absorp- tion, the worker’s minimum systemic fluoride dose was 8 mg/day, averaged over the entire week, or about 0.08 mg/kg-day for his reported weight of 230 lb (105 kg).3 Derryberry et al. (1963) reported a significantly higher frequency of albu- minuria in a group of workers exposed to airborne fluoride in a phosphate- fertilizer plant than in nonexposed controls (12.2% vs 4.5%) and suggested a relationship between fluoride excretion and renal function. Urinary fluoride ex- cretion averaged 4.6 mg/L (range, 2.1-14.7 mg/L) in the exposed group and 1.15 mg/L (range, 0.15-3.2 mg/L) in the controls. Two studies of aluminum potroom workers did not yield similar findings (reviewed by Hodge and Smith 1977). 3 The calculation is as follows: (15 m3/day)(1 day/24 h)(3 ppm)(0.82 mg/m3 per ppm)(19/20 mg fluoride per mg hydrogen fluoride)(8 h/day)(5/7) = 8 mg/day, assuming 100% absorption and dose averaged over the entire week. 

Hydrogen Fluoride 81 Ando et al. (2001) attributed decreased glomerular filtration rates to chronic exposure to fluoride from coal combustion; both inhalation of airborne fluoride and ingestion of contaminated food were involved. A few other reports have linked ingestion of fluoride to renal damage (for example, increased concentra- tions of the renal enzymes N-acetyl-β-glucosaminidase and γ-glutamyl transpep- tidase in children’s urine; Liu et al. 2005; Xiong et al. 2007) or urolithiasis (Singh et al. 2001). NRC (2006) has also reviewed human renal effects of fluo- ride exposure and made recommendations for further research. NRC (2006) concluded that fluoride interferes with normal endocrine function in humans. Reported effects include increased thyroid stimulating hor- mone; altered concentrations of thyroid hormones, calcitonin, or parathyroid hormone; secondary hyperparathyroidism; and impaired glucose tolerance (see Table 4-3). Thyroid effects were associated with estimated average or typical fluoride intakes as low as 0.05-0.1 mg/kg-day (0.01-0.03 mg/kg-day with iodine deficiency). Increased likelihood of impaired glucose tolerance was associated with intakes above 0.07 mg/kg-day, and increased parathyroid hormone concen- trations and secondary hyperparathyroidism were found at fluoride intakes of at least 0.15 mg/kg-day. Adequacy of nutrition seems to play a role in many end points; effects are less likely, or require higher fluoride intakes, with improved nutrition. Most of the studies reviewed in NRC (2006) were cross-sectional and TABLE 4-3 Summary of Selected Endocrine Effects Associated with Oral Fluoride Exposure in Humans Estimated Estimated Fluoride NOAEL Fluoride LOAEL Key End Point (mg/kg-day) (mg/kg-day) References Altered thyroid functiona 0.01-0.05 0.05-0.1 Bachinskii et al. 1985; Jooste et al. 1999; Susheela et al. 2005 Increased calcitonin 0.02-0.04 0.06 Teotia et al. concentrations 1978 Increased PTH concentrations 0.02-0.06 0.15 Teotia et al. or secondary 1978 hyperparathyroidism Impaired glucose tolerance 0.03 0.07 Trivedi et al. 1993 a Altered T4 or T3 concentrations, increased TSH concentrations, or increased goiter prevalence. Values shown are based on situations with adequate iodine intake. Iodine deficiency can decrease NOAEL and LOAEL. Abbreviations: LOAEL, lowest observed-adverse-effect level; NOAEL, no-observed- adverse-effect level; PTH, parathyroid hormone; TSH, thyroid-stimulating hormone. Source: NRC 2006. 

82 Exposure Guidance Levels for Selected Submarine Contaminants did not evaluate individual exposures; case reports, clinical studies, and experi- mental studies were also included. Most of the epidemiologic studies have in- volved long-term or lifelong exposures. On the basis of extensive review of epidemiologic, observational, and clinical studies, NRC (2006) concluded that lifetime exposure to fluoride at drinking-water concentrations of 4 mg/L and higher is likely to result in higher bone-fracture rates in the population than exposure at 1 mg/L (estimated average fluoride intakes from all sources, around 0.08 mg/kg-day vs 0.03 mg/kg-day). The evidence suggested an increased risk of bone fracture at 2 mg/L (estimated average fluoride intake from all sources, around 0.05 mg/kg-day), but NRC (2006) did not consider the available information to be conclusive. In general, the risk of fractures (especially hip fractures) increases with the concentration of fluoride in the bones—in effect, the bones become more brittle. Skeletal fluorosis includes a variety of radiographic and clinical presenta- tions, from increased skeletal density (stage I) to chronic joint pain, arthritic symptoms, calcification of ligaments, and osteosclerosis of cancellous bones (stage II) to excessive calcification in joints, ligaments, and vertebral bodies, muscle wasting, and neurologic deficits due to spinal-cord compression (stage III, or "crippling" skeletal fluorosis; NRC 2006). A number of reports describe skeletal fluorosis of various degrees in workers exposed to gaseous or particu- late fluorides (see, for example, Roholm 1937; Franke and Auermann 1972; Schlegel 1974; Franke et al. 1975; Baud et al. 1978; Dominok et al. 1984). On the basis of data collected by Derryberry et al. (1963), the California Office of Environmental Health Hazard Assessment (OEHHA 2003) derived a lowest observed-adverse-effect level (LOAEL) and a no-observed-adverse-effect level (NOAEL) of fluoride of 1.89 and 1.07 mg/m3, respectively, for increased bone density as observed radiographically (corresponding to hydrogen fluoride at 2.4 and 1.4 ppm in an occupational setting). Given a ventilation rate of 15 m3/day and a 40-h workweek, those concentrations would correspond to average sys- temic fluoride intakes of 6.8 and 3.8 mg/day, or 0.1 and 0.05 mg/kg-day for a 70-kg man. The range of exposure durations at the time of examination was 7.1- 24.8 years for people who had minimally increased bone density and 4.5-25.9 for people who had normal bone density (Derryberry et al. 1963, cited in OEHHA 2003). Bone fluoride concentrations in the ranges reported for stage II and stage III skeletal fluorosis will probably be reached by long-term (approach- ing lifetime) fluoride intakes of around 0.05 mg/kg-day (estimated from NRC 2006), but bone fluoride concentrations appear to be a marker, rather than a de- terminant, of the risk of skeletal fluorosis (NRC 2006). Franke et al. (1975) re- ported a lack of clear correlation among bone fluoride concentrations, radiologic changes, and symptoms; some workers with slight radiologic changes reported intense pain in the spine and large joints, and some with radiologically distinct fluorosis reported few complaints. 

Hydrogen Fluoride 83 Effects in Animals Acute Toxicity A rich dataset on the acute lethality and toxicity of hydrogen fluoride in laboratory animals exists. The data have been extensively reviewed by NRC (2004) and ATSDR (2003). In general, brief (1-h or less) exposure at more than 100 ppm results in severe respiratory tract lesions consisting of necrosis or in- flammation of the nasal passages and to a lesser extent, if at all, those lesions in the lower airways. Concentration-time relationships have been examined in short-term studies, and it has been uniformly concluded that the acute-lethality data on exposures of 30 min or less are best described by the relationship Cn × t = k, where C = concentration, t = time, k = constant, and n = 2 (Rosenholtz et al. 1963; ten Berge et al. 1986; Alexeef et al. 1993; NRC 2004), indicating that concentration is more important than time relative to acute lethality. Dalbey et al. (1998) noted that the relationship held for 2-min and 10-min exposures but commented that it might not hold for 60-min exposures. Thus, uncertainty exists relative to concentration-time relationships for exposures of 60 min or more. Although the data are not directly applicable to establishment of guidance on longer exposure because they are from short-term studies that focused on lethal- ity, they do strongly indicate that the respiratory tract is the primary target of hydrogen fluoride in brief high-concentration exposures. The preponderance of nasal lesions in short-term exposure studies is no doubt due to the extensive removal or extraction of inspired hydrogen fluoride in the nasal cavity of rodents (Morris and Smith 1982). Rodents are obligate nose- breathers, so rodent toxicity studies may underestimate the lower airway injury that might result in mouth-breathing humans exposed to hydrogen fluoride. Pseudo-mouth-breathing in rats can be obtained by insertion of an oral-tracheal cannula. Two studies of rats have used that technique. Nose-breathing and pseudo-mouth-breathing rats were exposed to hydrogen fluoride at 1,300 ppm for 30 min with a 24-h follow-up in the study of Stavert et al. (1991). In nose- breathing rats, the study authors found that hydrogen fluoride induced a marked reduction in minute ventilation (due to reduced breathing frequency), a response indicative of nasal trigeminal sensory nerve activation (Alarie 1973). In contrast, they found that pseudo-mouth-breathing rats exhibited an initial increase fol- lowed by a progressive decrease in ventilation, a response typical of lower air- way vagal sensory nerve activation (Alarie 1973). None of the nose-breathing rats died, but 25% of the pseudo-mouth-breathing animals died within 24 h after exposure. Severe nasal necrosis and inflammation occurred in the anterior por- tion of the nose in the nose-breathing animals, and no significant lower airway lesions were observed (Stavert et al. 1991). In contrast, moderate to severe ne- crosis and inflammation of the trachea and major bronchi, without epithelial damage to the smaller airways, occurred in the pseudo-mouth-breathing animals, and mild neutrophilic inflammation was observed in the alveoli. It is not known 

84 Exposure Guidance Levels for Selected Submarine Contaminants whether the alveolar neutrophils represented alveolar damage, aspiration of air- way neutrophils, or translocation during the airway fixation process. Using a similar oral-tracheal cannulation method, Dalbey et al. (1998) ex- amined the response of pseudo-mouth-breathing rats exposed to hydrogen fluo- ride for 2, 10, or 60 min with a 24-h follow-up. The aim of the study was to ex- amine concentration-time relationships. The 60-min exposures were at 20 and 48 ppm (1,200 and 2,800 ppm-min, respectively) for pseudo-mouth-breathing animals and 34 ppm (2,040 ppm-min) for nose-breathing animals. The 2-min and 10-min groups in the study were exposed at about 1,200, about 2,800, about 9,500 and about 17,000 ppm-min. Breathing frequencies decreased dramatically in the nose-breathing animals exposed at 1,000 ppm or higher for short periods, indicative of nasal trigeminal nerve activation, but were not markedly altered in the pseudo-mouth-breathing animals. In the 60-min low-concentration groups, breathing frequencies and minute ventilation were increased over baseline in nose-breathing but not pseudo-mouth-breathing animals. Pulmonary-function tests, BAL, and histopathology were used to characterize the injuries after the nonlethal exposures; all measures indicated the induction of airway injury by hydrogen fluoride in the 2-min or 10-min exposures at 9,500 and 17,000 ppm- min. As in the study of Stavert et al. (1991), after exposures of 10 min or less, nasal lesions predominated in the nose-breathing animals and large tracheobron- chial airway lesions in the pseudo-mouth-breathing animals. Hydrogen fluoride- induced effects were much less pronounced in the 60-min exposure groups. BAL protein and glucose-6-phosphate dehydrogenase were slightly increased in the 60-min 48-ppm group but not in the 60-min 20-ppm group. Pulmonary func- tion (as assessed by FVC, FEV1, and diffusing capacity) was unaffected, but a marginal increase in total lung capacity was noted. Histologic changes were not observed in the lower airways of pseudo-mouth-breathing animals after expo- sure at 20 or 48 ppm for 60 min. No nasal lesions were observed in nose- breathing rats after 60-min exposure at 34 ppm. In summary, short-term animal experiments indicate that the respiratory tract is the primary target of hydrogen fluoride in these exposure scenarios. Na- sal sensory nerve activation occurs during exposure in nasal-breathing animals. Injury occurs in the first airway that hydrogen fluoride comes into contact with: the nose in nose-breathing animals and the trachea in pseudo-mouth-breathing animals. At the same concentration-time product, injury is more severe in groups exposed at higher concentrations for shorter periods, indicating the pre- dominant influence of exposure concentration. A NOAEL for a 60-min exposure was 34 ppm (the only concentration tested) in nose-breathing rats and 20 ppm in mouth-breathing rats. Machle et al. (1934) exposed a small number of rabbits and guinea pigs to hydrogen fluoride at 29-9,784 ppm for 5 min to 41 h. Signs of irritation (cough- ing, sneezing, and mucoid conjunctival and nasal discharges) were observed at all exposures but were mild at 61 ppm and less. “Erosion of areas of the cornea and necrosis of the turbinates” were also observed frequently. Pulmonary hem- orrhage, edema, and bronchitis were common in animals that died within 48 h of 

Hydrogen Fluoride 85 exposure. Lesions were also observed in the liver and kidneys, but the extent to which they reflected changes that were secondary to the pulmonary effects or underlying disease in the control animals used is not clear. NOAEL or LOAEL values could not be determined from this study. The data in this study indicate that a Cn × t relationship exists for exposures of 4 min to 8 h with lethality as the end point, but the presentation of the data precludes precise estimation of n in the concentration-time relationship. Morris (1979) performed a concentration-response study in which rats (six per group) were subjected to whole-body exposure to hydrogen fluoride at 13, 33, 88, 142, 181, or 218 ppm for 6 h and animals were killed 6 h after exposure (Table 4-4). Exposures at 181 or 218 ppm resulted in 100% mortality. Mucoid nasal discharges were apparent in the animals, and hemorrhagic lungs were ob- served at necropsy. Histopathologic examination of the lungs and kidneys was performed on surviving but not dying animals. An underlying degree of chronic peribronchial and perivascular lymphocytic inflammation was observed in con- trol rats and rats exposed to hydrogen fluoride and was not related to exposure. No lung lesions related to exposure were observed in the animals exposed at 142 ppm or lower. The nasal cavity was not examined. Renal proximal tubular in- jury, as evidenced by nuclear pyknosis, was increased in a concentration- dependent manner; 88 ppm was the lowest concentration associated with injury. Blood urea nitrogen concentrations were also increased in a concentration- response manner; concentrations were significantly increased in animals ex- posed at 33 ppm or higher compared with control animals. Plasma ionic fluoride concentrations averaged 0.032, 0.57, 1.03, 2.72, and 5.73 μg/mL in rats exposed at 0, 13, 33, 88, and 142 ppm, respectively, and were significantly increased over control values in all exposed groups. Lung fluoride exceeded plasma fluo- ride by 2-3 fold, and this suggests that inspired hydrogen fluoride penetrated to the lungs in this exposure regimen. (That was not the case in animals exposed for 1 h.) The study documents increased systemic burdens of fluoride after hy- drogen fluoride exposure and the presence of renal injury in animals exposed at 33 ppm or higher. Acute lethality, probably of respiratory tract origin, was ob- served, but the failure to include examination of nasal lesions precludes deter- mination of a useful NOAEL. Repeated Exposures and Subchronic Toxicity Table 4-4 summarizes the results of studies of repeated exposure of ani- mals. Stokinger (1949) exposed dogs, rabbits, guinea pigs, rats, and mice to hydrogen fluoride at 8.6 or 33 ppm 6 h/day, 6 days/week for 5 weeks. Subcuta- neous hemorrhages around the eyes and feet were apparent in rats and mice, primarily at 33 ppm. Subcutaneous hemorrhages also occurred in the feet of rats at 8.6 ppm, although the lesions were less severe. At 33 ppm, there was 100% mortality in rats and mice and no deaths in the other species. No deaths were 

86 TABLE 4-4 Effects of Hydrogen Fluoride in 6-Hour or Longer Animal Studies Concentration (ppm) Time Species and Effects Reference 13, 33, 88, 142, 6h Rats. 100% mortality at 181 and 218 ppm, presumably due to nasal obstruction; Morris 1979 181, 218 histologic analysis not performed on dying animals; no histopathologic pulmonary lesions observed at 142 ppm or lower, renal tubular cell pyknosis observed at 88 ppm and higher, and blood urea nitrogen increased at 33 ppm and higher; nasal tissues not examined 18.6 6-7 h/day, Rabbit, guinea pig, monkey. Lethal to two of three guinea pigs; necropsy Machle and 5 days/week, performed 7-9 months after exposure protocol; bronchial (guinea pigs) and Kitzmiller 1935 10 weeks alveolar (guinea pigs, rabbits) lung lesions observed; no lung damage in monkeys, but renal injury observed in monkeys and rabbits; nasal tissues not examined 8.6, 33 6 h/day, Rat, mouse, guinea pig, rabbit, dog. 100% mortality at 33 pm in rats and mice; no Stokinger 1949 5 days/week, deaths in other species; histopathologic analysis on dog, rabbit, and rat at 33 ppm, total of 166 and moderate lung hemorrhage or edema observed in all three species; only gross exposure-hours necropsy performed on 8.6-ppm group, and no effect reported in rat and rabbit, but focal lung hemorrhage observed in one of five dogs; nasal tissues were not examined 1, 10, 25, 65, 100 6 h/day, Male and female rat. 100% mortality in males at 65 ppm or greater, 100% Placke et al. 5 days/week, mortality in females at over 25 ppm, no deaths at other concentrations; ocular 1990 2 weeks opacity, skin lesions, nasal and ocular discharges observed at 25 ppm or greater; body weights were decreased at 10 ppm or higher 0.1, 1, 10 6 h/day, Male and female rat. 25% mortality in males, 5% mortality in females at 10 ppm; Placke and 5 days/week, no observed histopathologic lesions in any exposure group Griffin 1991 90 days 

Hydrogen Fluoride 87 observed at 8.6 ppm. In the animals exposed at 33 ppm, pulmonary hemorrhage and edema of varied degree were noted in the dogs, rabbits, and rats (the only three species in which pathologic findings were examined). Renal cortical de- generation was observed in the rats. Only gross examination was performed on animals at 8.6 ppm. Localized hemorrhagic areas were noted in the lungs of one of five dogs but not the rats or rabbits. The lack of histopathologic examination of the nose and lungs in the 8.6-ppm groups makes it difficult to determine whether this concentration is a NOAEL or LOAEL; however, the observation by gross examination of localized hemorrhage in one of five dogs suggests that the LOAEL was exceeded. Machle and Kitzmiller (1935) exposed four rabbits, three guinea pigs, and two monkeys to hydrogen fluoride at 18.6 ppm 6-7 h/day, 5 days/week for 10 weeks. Surviving animals were allowed to recover for 7-9 months before termi- nation. Lesions were found in the lungs, liver, and kidneys of exposed animals. Two guinea pigs died during the exposure phase of the study; pathologic exami- nation revealed pulmonary damage in both; pulmonary injury was also observed in the guinea pig that survived the recovery period. All four rabbits survived until termination, and all had alveolar damage (edema and cellular infiltration) and renal injury (tubular necrosis or degeneration). Both monkeys survived ex- posure and exhibited similar renal injury but not lung damage. The exposure concentration of 18.6 ppm clearly represented a frank-effect level as indicated by the deaths and by the tissue damage that was observed in all species. Battelle Laboratories (Placke et al. 1990) performed a repeated-exposure inhalation study with five male and five female rats per group exposed to hydro- gen fluoride 6 h/day, 5 days/week for a total of 10 exposures. Exposure concen- trations were 0, 1, 10, 25, 65, and 100 ppm. Exposure at 65 or 100 ppm caused 100% mortality; 100% mortality also occurred in female rats in the 25-ppm group, but no male rats exposed at this concentration died. No deaths occurred in the 1-ppm and 10-ppm groups. Corneal opacity, skin lesions, and ocular and nasal discharges were observed in the animals exposed at 25 ppm or higher; these effects probably reflected the irritating properties of hydrogen fluoride. Histopathologic examination of tissues was not performed. Body weights were reduced in animals exposed at 10 ppm or higher. The authors concluded that 10 ppm was a LOAEL and 1 ppm a NOAEL. Battelle Laboratories (Placke and Griffin 1991) performed a 90-day re- peated-exposure inhalation study with groups of 20 male and 20 female rats exposed 6 h/day, 5 days/week for a total of 65 exposures (over 90 days). Expo- sure concentrations were 0, 0.1, 1, and 10 ppm. Mortality (25% in males and 5% in females) occurred in the 10-ppm group but no others. Complete histology, including four nasal sections per animal was performed on all animals with the protocol of Young (1981). No lesions were observed, even in the dying animals; a cause of death was not determined in the study. Rosenholtz et al. (1963) reported that nasal lesions induced by hydrogen fluoride are localized to the external nares and nasal vestibule. The most anterior nasal section in Young’s protocol does not include the most anterior portions of the nose (squamous epi- 

88 Exposure Guidance Levels for Selected Submarine Contaminants thelium-lined vestibule); thus, it is possible that the most likely sites of lesions in the noses of the rats in the Battelle study were not examined. In summary, mortality and respiratory tract injury commonly result from single and repeated 6-h inhalations. Nasal lesions are probably present, but his- tologic examination of nasal tissues after repeated exposures has not been uni- formly performed or, if performed, may have missed the affected area. Further- more, renal damage is observed after single or repeated 6-h exposures, most likely because of fluoride absorption. The published studies fail to establish a clear NOAEL because effects were observed at all exposure concentrations. Repeated exposure at 8.6 ppm may yield a LOAEL based on pulmonary hemor- rhage in one of five dogs, but the absence of histopathologic analysis makes this conclusion tenuous. Additional experiments have shown subchronic effects of inhalation expo- sure to hydrogen fluoride. They involved continuous exposure (24 h/day) of albino rats to hydrogen fluoride at 0.01, 0.04, or 0.1 ppm for 5 months (Sadilova et al. 1965). The authors reported disturbances in conditioned reflexes and lengthened motor chronaxie of lower hind leg flexors in rats exposed at 0.04 ppm or higher and morphologic changes in nerve cells in the motor and sensory areas of the brains of animals exposed at 0.1 ppm. Others have concluded that it is not possible to evaluate the underlying experimental work fully (Smith and Hodge 1979). More important, the results seem highly unlikely to be related to the exposures. At a ventilation rate of 0.2 m3/day (0.14 L/min), rats exposed at 0.04 and 0.1 ppm inhale a total of 6 and 20 μg of fluoride. Given that a rat con- sumes rodent chow at 10 g/100 g of body weight per day and that it typically contains fluoride at at least about 30 μg/g or more (de Lopez et al. 1976; Morris 1979), it is unlikely that the systemic fluoride burdens were significantly altered by the inhaled hydrogen fluoride. NRC (2006) reviewed several subchronic animal studies of oral fluoride exposure. Bobek et al. (1976) found disturbed thyroid function in male rats given fluoride at 0.4-0.6 and 4-6 mg/kg-day in drinking water for 60 days; ATSDR (2003) derived a LOAEL of 0.5 mg/kg-day from the study. Hara (1980) described effects on thyroid function in rats at fluoride doses as low as 0.1 mg/kg-day for 54-58 days. Zhao et al. (1998) found increased thyroxine concen- trations and reduced radioiodine uptake in mice with normal iodine intake at fluoride doses of 3 mg/kg-day for 100 days.4 Rigalli et al. (1992, 1995) reported disturbed glucose tolerance in female rats at fluoride intakes of about 10 mg/kg- day for 90 or 100 days. It is important to note that rats and mice tend to require 4 Estimated fluoride doses (here and later in the report) were calculated from informa- tion provided in the cited papers. In the absence of reported consumption rates, the com- mittee used a water consumption rate of 0.1 L/kg of body weight or a food consumption rate of 0.1 kg of feed per kilogram of body weight, as appropriate. 

Hydrogen Fluoride 89 fluoride intakes at least 5-20 times higher than humans do to yield similar physiologic concentrations or health effects (NRC 2006). Chronic Toxicity NRC (2006) reviewed a number of chronic animal studies for several end points, including endocrine, bone, and neurologic effects. Altered glucose me- tabolism occurred in rabbits at fluoride exposures of 7-10.5 mg/kg-day for 6 months (Turner et al. 1997), and altered thyroid metabolism occurred in rats at 3 mg/kg-day for 7 months (Guan et al. 1988). Several studies of Turner et al. (reviewed by NRC 2006) found decreased bone strength in rats and rabbits with long-term fluoride intake, corresponding to bone fluoride concentrations of 6,000-8,000 mg/kg. ATSDR (2003) derived a NOAEL of 0.15 mg/kg-day and a LOAEL of 0.5 mg/kg-day for decreased vertebral strength and bone mineraliza- tion in male rats given fluoride in drinking water for 16 or 48 weeks (Turner et al. 2001). Reproductive Toxicity in Males Ortiz-Pérez et al. (2003) reported altered serum hormone concentrations with normal semen measures in men occupationally exposed to fluoride in a factory that was producing hydrofluoric acid and aluminum fluoride. The ob- served effects included increased follicle-stimulating hormone (FSH) and de- creased testosterone, inhibin B, and prolactin. Total fluoride exposures of the occupationally exposed men were estimated to be 3-27 mg/day from drinking water and occupational exposure compared with 2-13 mg/day in a group ex- posed to fluoride only from drinking water. The intakes correspond to fluoride doses of 0.03-0.2 mg/kg-day and 0.05-0.4 mg/kg-day for the low and high expo- sure groups, respectively, on the basis of 70-kg body weight (NRC 2006). The fluoride exposures of the two groups overlapped, and the occupational group had exposure to other chemicals besides fluoride. Tokar and Savchenko (1977) also reported increased FSH and decreased testosterone in a study of 41 men (33-45 years old) who had fluorosis (appar- ently of occupational origin) compared with 19 control men who had no occupa- tional contact with fluorine compounds. In men with at least 15 years of expo- sure to fluorine compounds, there was also an increased concentration of blood luteinizing hormone (LH). Full details of the study, including the magnitude of fluoride exposure and information about concurrent exposures, are not available in English. Susheela and Jethanandani (1996) found a significant reduction in serum testosterone concentrations in skeletal-fluorosis patients (exposed to fluoride at 1.5-14.5 mg/L in drinking water) compared with controls (no skeletal fluorosis and exposed to fluoride at 0.1-0.9 mg/L in drinking water). Male relatives of the 

90 Exposure Guidance Levels for Selected Submarine Contaminants skeletal-fluorosis patients who did not have skeletal fluorosis but drank the same water as the skeletal-fluorosis patients had intermediate concentrations of testos- terone. If one assumed a 2-L/day water intake for 60-kg men,5 fluoride doses would be about 0.05-0.5 mg/kg-day for the skeletal-fluorosis patients and their male relatives without fluorosis, and 0.003-0.03 mg/kg-day for the controls. Thus, with or without skeletal fluorosis, men with high fluoride intakes had re- duced serum testosterone. The mean serum fluoride concentration associated with reduced testosterone was 0.24 mg/L in the men with skeletal fluorosis and 0.19 mg/L in the relatives without fluorosis but only 0.05 mg/L in the control group. The studies by Susheela and Jethanandani (1996) and Ortiz-Pérez et al. (2003) suggest, but do not confirm, that fluoride intake can alter the reproduc- tive-hormone environment. The dosage necessary to produce the effects cannot be established with certainty on the basis of those studies but would probably be at least 0.05 mg/kg-day, whereas exposure at below 0.03 mg/kg-day appears to be associated with a low risk of effects on male reproductive hormone status. Several animal studies have also reported effects of fluoride exposure on male reproduction. Das Sarkar et al. (2006) reported decreases in testicular en- zymes and low plasma concentrations of testosterone, FSH, and LH in rats ex- posed to sodium fluoride at 20 mg/kg-day (fluoride at 9 mg/kg-day) for 28 days. Administration of calcium and vitamin E reversed the effects to control values in that study. Araibi et al. (1989) reported significant decreases in serum testoster- one, a 50% reduction in fertility, and a decrease in the percentage of seminifer- ous tubules containing spermatozoa in rats exposed to sodium fluoride (fluoride intake of 9 mg/kg-day) in their feed for 60 days. A decrease in diameter of the seminiferous tubules was observed at a fluoride intake of 4.5 mg/kg-day. Abnormalities of spermatogenesis and decreased ability of epididymal spermatozoa to capacitate in vitro have been reported in mice given fluoride at 10 or 100 mg/L in drinking water (tap water with or without aluminum at 10 mg/L) for 3 months (Dvoráková-Hortová et al. 2008). Effects were not seen in mice that received fluoride alone at 1 mg/L but were seen when fluoride (1 mg/L) and aluminum (10 mg/L) were jointly administered. That finding is con- sistent with other reports that fluoride toxicity increases in the presence of aluminum (NRC 2006). The estimated fluoride intakes were 0.1, 1, and 10 mg/kg-day. Immunotoxicity NRC (2006) reviewed the available information on immune-system effects of fluoride. Both stimulatory and inhibitory effects have been reported. Mice exposed to fluoride at 5 or 10 mg/m3 in air (4 h/day for 14 days, aerosols less 5 Because the study was conducted in India, a smaller body size was assumed (NRC 2006). 

Hydrogen Fluoride 91 than 10 μm in diameter) showed concentration-dependent suppression of pul- monary bactericidal activity against Staphylococcus aureus (Yamamoto et al. 2001). At 10 mg/m3, there was a significant decrease in the number of alveolar macrophages in BAL fluid in mice not bacterially challenged. There was also a significant increase in polymorphonuclear leukocytes and lymphocytes at 10 mg/m3 with a significant decrease in body weight and an increase in lung weight (attributed to microscopically observed edema). No significant pulmonary effect was seen at 2 mg/m3 in comparison with the controls. The authors concluded that fluoride inhalation might reduce the ability to cope with bacterial infections. Unlike inspired hydrogen fluoride, aerosols penetrate to the alveoli, and this makes direct extrapolation of the study results difficult. A study of the relationship between asthma and occupational fluoride ex- posure was discussed earlier in the chapter (see “Occupational and Epidemi- ologic Studies”). Genotoxicity NRC (2006) concluded that the evidence of genotoxicity of fluoride was inconsistent. Fluoride does not appear to be a direct mutagen, but a number of mammalian in vitro systems have shown dose-dependent cytogenetic or cell- transformational effects of fluoride exposure. Several papers suggest an indirect mechanism, such as interaction with DNA synthesis or repair enzymes, rather than a direct interaction with DNA (Aardema et al. 1989; Aardema and Tsutsui 1995; Meng and Zhang 1997). Lasne et al. (1988) suggested a mechanism of promotion without excluding a genetic mechanism. Human cells also seem to be much more susceptible to chromosomal damage by fluoride than rodent cells (Kishi and Ishida 1993). Meng et al. (1995) and Meng and Zhang (1997) re- ported increased cytogenetic effects (sister-chromatid exchanges, chromosomal aberrations, and micronuclei in peripheral blood lymphocytes) in workers ex- posed to airborne fluoride, mostly as hydrogen fluoride and tetrafluorosilane, at concentrations of 0.5-0.8 mg/m3. Genotoxic effects in vitro have been reported at fluoride concentrations at or above about 5 mg/L, depending on the experimental system examined (Table 4-5). In nonfatal cases of acute hydrofluoric acid poisoning, urinary and serum fluoride concentrations as high as 110 and 42 mg/L, respectively, have been reported (Björnhagen et al. 2003; Vohra et al. 2008). Urinary fluoride concentra- tions as high as 44 mg/L have been reported after occupational fluoride expo- sures in excess of 3 mg/m3 (Derryberry et al. 1963; summarized by OEHHA 2003). Thus, it is possible that renal and bladder epithelium could experience fluoride concentrations, from exposure to inhaled hydrogen fluoride, in a range at which genotoxic effects have been reported. 

92 Exposure Guidance Levels for Selected Submarine Contaminants TABLE 4-5 Summary of Results of Positive Genotoxicity Studies of Fluoride End Point Reference In vitro Systems Dose-dependent transformation of SHE cells after NaF at 75 and 100 Lasne et al. µg/mL (F- at 34 and 45 mg/L) 1988 Cell transformation (or promotion of cell transformation) after NaF at Lasne et al. 25 µg/mL (F- at 11.3 mg/L) in BALB/3T3 mouse embryo cells 1988 Dose-dependent increase in chromosomal aberrations in CHO cells Aardema et al. exposed during G2 to NaF at 25-100 µg/mL (F- at 11-45 mg/L); no 1989 significant increases after NaF at or below 10 µg/mL (F- at 4.5 mg/L) Chromosomal aberrations (clastogenicity) in cell lines from Kishi and chimpanzees and men but not in other primates or in rodents; NaF at 1- Ishida 1993 6 mM (F- at 19-114 mg/L) Chromosomal aberrations in CHO cells after NaF at or above 50 mg/L Aardema and (F- at 22.6 mg/L) Tsutsui 1995 Clastogenicity (increase in chromosomal aberrations) in cultured Oguro et al. human diploid cells after F- at 5 or 10 ppm (5 or 10 mg/L) for 2.5 h or 1995 continuously for 5-6 days Dose-dependent increases in chromosomal aberrations in rat vertebral Mihashi and body-derived cells after NaF at 0.5-1.0 mM (F- at 9.5-19 mg/L) for 24- Tsutsui 1996 48 h Small dose-dependent increase in chromosomal aberrations in cultured Gadhia and human lymphocytes after NaF at 10, 20, or 30 µg/mL (F- at 4.5-13.5 Joseph 1997 mg/L); no increase in SCEs DNA damage (ascertained with Comet assay) to human embryo Wang et al. hepatocyte cells after NaF at 40, 80, 160 µg/mL (F- at 18, 36, 72 mg/L) 2004 Cytoxicity in human primary cell cultures (skin fibroblasts) and Lestari et al. permanent cell lines; IC10 of HF of 0.6-1.0 mM (F- at 11.4-19 mg/L) or 2005 of NaF of 1.3-3.3 mM (F- at 25-63 mg/L); NOAEC of HF of 0.08-0.24 mM (F- at 1.5-4.6 mg/L) or of NaF of 0.09-0.32 mM (F- at 1.7-6.1 mg/L) In vivo Studies in Humans Increased SCEs, micronuclei in 53 fluorosis patients with F- at 4-15 Wu and Wu mg/L in drinking water 1995 Increased SCEs, chromosomal aberrations, and micronuclei in Meng et al. peripheral blood lymphocytes from occupational exposures to F- at 1995; Meng 0.50-0.80 mg/m3 (mostly as HF and SiF4) and Zhang 1997 − Abbreviations: CHO, Chinese hamster ovary; F , fluoride ion; G2, phase G2 of cell cycle; HF, hydrogen fluoride; IC10, inhibitory concentration at 10%; NaF, sodium fluoride; NOAEC, no observable adverse effect concentration; SCE, sister-chromatid exchange; SHE, Syrian hamster embryo; SiF4, tetrafluorosilane. 

Hydrogen Fluoride 93 Carcinogenicity NRC (2006) concluded that fluoride appears to have the potential to initi- ate or promote cancers, but the overall evidence from human and animal studies is mixed. Several occupational studies (Grandjean et al. 1992; Romundstad et al. 2000; Grandjean and Olsen 2004) are consistent with an association between exposure to inhaled fluoride and bladder cancer (reviewed by NRC 2006). Esti- mated fluoride concentrations at which the cohorts were exposed were 15-20 mg/m3 (Grandjean et al. 1992; Grandjean and Olsen 2004) and 2.5 mg/m3 or less (Romundstad et al. 2000). The U.S. Environmental Protection Agency has not classified fluoride with respect to carcinogenicity (EPA 1989). The International Agency for Research on Cancer lists inorganic fluorides in group 3, "not classi- fiable as to its carcinogenicity to humans," on the basis of reviews in 1982 and 1987 (IARC 1987, p. 208). TOXICOKINETIC AND MECHANISTIC CONSIDERATIONS Several lines of reasoning suggest that respiratory tract irritation by hy- drogen fluoride results from local effects due to its acidic nature. The respiratory tract is not a target for fluoride ion (NTP 1990) and pulmonary immunotoxic effects of fluoride-containing dust typically require higher concentrations than that needed for local irritation. In contrast, the respiratory tract is known to be sensitive to acidic vapors. Aqueous soluble weak acids, such as acetic acid, are known sensory irritants in animals and humans (Morris et al. 2003; Shusterman et al. 2005), and the injury pattern in the nose after acute exposure to hydrogen fluoride (Rosenholtz et al. 1963; Dalbey et al. 1998) is identical with that after exposure to weak acids (Buckley et al. 1984). Thus, the acidic nature of hydro- gen fluoride probably contributes substantially to its irritant properties. It should be noted, however, that biologically significant concentrations of hydrogen ion are not absorbed systemically during exposure to hydrogen fluoride at 3 ppm or lower—at a ventilation rate of 15 m3/day, the total absorbed acid burden in the human is 1.3 μmol/min at that exposure concentration. Although that may be sufficient to induce acidification locally in the respiratory tract, it is not suffi- cient to result in significant systemic acidification. Regional deposition patterns are critical in influencing respiratory tract in- jury due to inspired irritants (EPA 1994). Much evidence suggests that hydrogen fluoride is scrubbed efficiently in the first airways with which it comes into con- tact: the nose during nose-breathing and the trachea and bronchi during mouth- breathing. In humans, nasal irritation predominates over lower airway irritation, and this suggests a nasal site of action. In rats, short-term (1-h or less) exposure caused predominantly nasal injury with little if any lower airway injury in nose- breathing animals. In mouth-breathing animals, tracheal injury and bronchial injury occur indicating that these tissues are sensitive to injury when a sufficient amount of hydrogen fluoride reaches them (Stavert et al. 1991; Dalbey et al. 

94 Exposure Guidance Levels for Selected Submarine Contaminants 1998). That injury does not occur in distal airways of mouth-breathing animals suggests substantial scrubbing in the large airways. Studies on nasal uptake efficiency during a 35-min protocol revealed greater than 99% uptake in the upper respiratory tract of the rat (Morris and Smith 1982). Complete scrubbing of water-soluble ionizable acids in the nasal passages is consistent with theoretical understanding of vapor uptake processes (Morris and Smith 1982; Morris 2006). Lung fluoride concentrations in rats ex- posed to hydrogen fluoride for 1 h are no higher than plasma fluoride concentra- tions (Morris 1979; Morris and Smith 1982); this suggests that substantial amounts of hydrogen fluoride do not penetrate to the lungs via the airstream during short-term exposure. In contrast, after 6-h exposure in rats, lung fluoride concentrations exceed plasma concentrations by 3-fold, and this suggests that penetration to the lower airways does occur in prolonged exposures (Morris 1979). That may reflect the gradual accumulation of fluoride in the lungs due to the persistent delivery of even a small percentage of the inspired hydrogen fluo- ride or enhanced penetration of the nose as exposure times lengthen and nasal lesions occur. Future experiments would be needed to resolve the issue, but the results suggest that the potential for lung injury may increase with exposure time. Inhaled hydrogen fluoride is rapidly absorbed into the bloodstream, as in- dicated by increased plasma and tissue fluoride concentrations after exposure. Blood fluoride concentrations were increased in a concentration-dependent manner after 35-min isolated upper respiratory tract exposure to hydrogen fluo- ride and indicated that rapid and efficient absorption occurs in the nasal cavity (Morris and Smith 1982). Such absorption patterns in the nose are common (Black and Hounam 1968; Yokoyama et al. 1971). Increased systemic fluoride concentrations after exposure to hydrogen fluoride have been documented in the rat and dog (Morris and Smith 1982; Largent 1961). Plasma fluoride concentra- tions are significantly increased after 1-h exposure to hydrogen fluoride in the human; the peak plasma concentration occurs at the end or within 30 min of the end of exposure (Lund et al. 1997). In their entirety, the data indicate that in- haled hydrogen fluoride is rapidly absorbed in the respiratory tract. Particularly with respect to short-term exposures (35 min in rats and 60 min in humans), it is unlikely that significantly increased plasma concentrations would have been observed if fluoride deposition and absorption were not highly efficient. Al- though precise determinations are not possible, it is appropriate to assume that inhaled hydrogen fluoride is completely deposited and that the fluoride is com- pletely and rapidly absorbed into the bloodstream during inhalation exposure (Morris and Smith 1982). A number of human studies have documented in- creased urinary fluoride concentrations or systemic fluoride effects due to inha- lation of airborne hydrogen fluoride or particulate fluorides (see, for example, Roholm 1937; Collings et al. 1951; 1952; Largent et al. 1951; Derryberry et al. 1963; Franke and Auermann 1972; Schlegel 1974; Franke et al. 1975; Hodge and Smith 1977; Baud et al. 1978; Waldbott and Lee 1978; Grandjean and Thomsen 1983; Dominok et al. 1984; Grandjean et al. 1990; Rees et al. 1990). 

Hydrogen Fluoride 95 The absorption, distribution, and excretion of fluoride have been reviewed by ATSDR (2003). In brief, fluoride is rapidly and efficiently absorbed from the gastrointestinal tract via passive diffusion. Absorption efficiencies can approach 99-100%. In that regard, gastrointestinal absorption appears to be similar to res- piratory tract absorption. Absorbed fluoride ion does not accumulate in most soft tissues but does accumulate in bone. Fluoride is incorporated into bone by re- placing the hydroxyl ion to form hydroxyfluroapatite. It has been estimated that about 60% of an intravenous dose of fluoride is sequestered in bone and the re- mainder eliminated in urine. Fluoride exerts systemic effects on the body in a number of ways: enzyme inhibition, alteration of normal physiologic signaling mechanisms, interference with endocrine function, disruption of calcium balance, and increased brittleness of bones due to incorporation of fluoride into the apatite lattice structure (NRC 2006). NRC (2006) has described the pharmacokinetics of fluoride and various possible mechanisms of specific toxic effects. The presence of aluminum fluo- ride or beryllium fluoride complexes may cause increased toxicity or additional toxic effects. Some effects of fluoride are more likely in the presence of low calcium intake, low iodine intake, or generalized poor nutrition. Fluoride reten- tion and hence toxicity may be increased in the presence of renal impairment (NRC 2006). INHALATION EXPOSURE LEVELS FROM THE NATIONAL RESEARCH COUNCIL AND OTHER ORGANIZATIONS A number of organizations have established or proposed acceptable expo- sure limits or guidelines for inhaled hydrogen fluoride. Selected values are summarized in Table 4-6. COMMITTEE RECOMMENDATIONS The committee’s recommendations for EEGL and CEGL values for hy- drogen fluoride are summarized in Table 4-7. The current U.S. Navy values are provided for comparison. 1-Hour EEGL Nasal irritation appears to be the most sensitive response to hydrogen fluoride. Nasal sensory nerve activation has been shown to occur in the rat (as evidenced by decreased breathing frequencies) and the human (as evidenced by nasal tickling, soreness, and other symptoms). The controlled human 

96 Exposure Guidance Levels for Selected Submarine Contaminants TABLE 4-6 Selected Inhalation Exposure Levels for Hydrogen Fluoride from the National Research Council and Other Organizationsa Exposure Level Organization Type of Level (ppm) Reference Occupational ACGIH TLV-TWA (measured as 0.5 ACGIH 2005 fluoride) TLV-Ceiling (measured 2 as fluoride) NIOSH REL-TWA 3 NIOSH 2005 REL-ceiling (15-min) 6 OSHA PEL-TWA 3 29 CFR 1910.1000 General Public ATSDR Acute MRL 0.02 ATSDR 2008 NAC/NRC AEGL-1 (1-h) 1 NRC 2004 AEGL-2 (1-h) 24 AEGL-1 (8-h) 1 AEGL-2 (8-h) 12 a Comparability of EEGLs and CEGLs with occupational-exposure and public-health standards or guidance levels is discussed in Chapter 1 (“Comparison with Other Regula- tory Standards or Guidance Levels”). Abbreviations: ACGIH, American Conference of Governmental Industrial Hygienists; AEGL, acute exposure guideline level; ATSDR, Agency for Toxic Substances and Dis- ease Registry; MRL, minimal risk level; NAC, National Advisory Committee; NIOSH, National Institute for Occupational Safety and Health; NRC, National Research Council; OSHA, Occupational Safety and Health Administration; PEL, permissible exposure limit; REL, recommended exposure limit; TLV, Threshold Limit Value; TWA, time-weighted average. TABLE 4-7 Emergency and Continuous Exposure Guidance Levels for Hydrogen Fluoride Current U.S. Navy Values Committee Recommended Exposure Level (ppm) Values (ppm) EEGL 1-h 2 3 24-h 1 1 CEGL 90-day 0.1 0.04 Abbreviations: CEGL, continuous exposure guidance level; EEGL, emergency exposure guidance level. 

Hydrogen Fluoride 97 experimental studies of Lund et al. (1997, 1999, 2002, 2005) provide the most reliable data for establishing a 1-h EEGL. The studies incorporated exercise in the exposure regimen, and the study groups included people with allergic rhini- tis. In those studies, several subjects experienced moderate to marked irritation in response to hydrogen fluoride at exposure concentrations of 3.1 ppm or higher, whereas no subjects exposed to concentrations of 2.9 ppm or lower re- ported this degree of irritation. Therefore, the data suggest that marked irritation occurs at 3.0 ppm or higher. That conclusion is supported by the work of Largent (1961), who reported on one subject with a rhinovirus infection who experienced considerable discomfort at 3.4 ppm. Nasal inflammation as assessed by nasal lavage is induced by hydrogen fluoride at 4.0 ppm but is unlikely to have long-term health significance. Lower airway lymphocytic inflammation was observed after 1-h exposures to hydrogen fluoride at over 0.9 ppm; how- ever, no significant alterations in pulmonary function (FEV1 and FVC) were observed at this or higher exposure concentrations. No significant short-term or long-term health effects would be expected in connection with the lower airway lymphocytic inflammation. Animal studies indicate that the lower respiratory tract may be more sensitive to the effects of hydrogen fluoride than nasal tissues. However, Lund et al. included exercise and anticipated mouth-breathing, thus alleviating the concern that lower respiratory tract effects may not have been observed because of nose-breathing. On the basis of the above data, 3 ppm was selected as the appropriate point of departure for derivation of a 1-h EEGL to minimize the possibility of marked sensory irritation. The level of irritation that may be experienced is not considered sufficient to impair the ability to perform essential tasks. An uncer- tainty factor for interindividual variability does not appear to be required inas- much as the study group included potentially sensitive people, such as those with rhinitis resulting from hay fever. Thus, a 1-h EEGL of 3 ppm is recom- mended. That exposure level is probably above the odor threshold and may cause mild irritation but is unlikely to produce more severe irritation. 24-Hour EEGL An extensive data base for establishment of a 24-h EEGL does not exist, so determining an appropriate point of departure for a 24-h EEGL is not straightforward. Considerable data exist on 1-h exposures of animals and hu- mans, but extrapolation of 1-h data to 24 h is problematic. The rodent single-6-h exposure study of Morris (1979) did not include examination of the most sensi- tive target, the nasal passages. Animal-toxicity data from repeated exposures might provide insights into an appropriate 24-h EEGL, but most of those studies used lethal concentrations, so their results are unsuitable for this purpose. One study reported no deaths and focal pulmonary hemorrhage (by gross examina- tion) in one of five dogs after multiple 6-h exposures at 8.6 ppm (Stokinger 1949). Although those data are not sufficient for risk-assessment purposes, they 

98 Exposure Guidance Levels for Selected Submarine Contaminants suggest that severe life-threatening injury did not occur in the dog even with multiple exposures at that concentration. The human subject study of Largent (1961) incorporated multiple daily 6- h exposures to hydrogen fluoride at 2.6-4.7 ppm. Subjects reported only slight nasal irritation, so the 1-h threshold value of 3.0 ppm for marked irritation might be appropriate for a 6-h exposure. Certainly, the multiple-exposure data suggest that increased sensitivity to the irritating effects of hydrogen fluoride does not occur. Although far from strong, the few animal data suggest that substantial respiratory tract damage is unlikely to result from a single 6-h exposure at that concentration. The most appropriate Cn × t function for extrapolation from 6 h to 24 h is not known inasmuch as there is considerable uncertainty regarding the Cn × t function. Therefore, rather than using a concentration-time adjustment based on a Cn × t function, the committee recommended that a database uncer- tainty factor of 3 be used. Application of the uncertainty factor results in a 24-h EEGL of 1.0 ppm. Given a total ventilation of 15 m3 in 24 h, an absorbed fluoride dose of 12 mg for a 70-kg man can be derived. That dose would most likely not produce sys- temic injury in a single exposure. Some people may experience nasal irritation at the 24-h EEGL. Some degree of a respiratory inflammatory response might also result, but it would not be expected to impair the ability to perform essential tasks nor would it be expected to be of long-term health consequence. 90-Day CEGL Few data are available for establishment of a 90-day CEGL. Epidemi- ologic data on occupational exposure suggest that occupational exposure to hy- drogen fluoride at an average of 0.27 ppm may be associated with an increased risk of asthma (Taiwo et al. 2006). Multivariate analysis provided a relative risk for hydrogen fluoride of 1.18 per 0.12 ppm (95% CL 1.09-1.3). It should be noted that the workers were exposed to multiple fluorides (particulate and gase- ous) for many years, so precise evaluation of the effects of hydrogen fluoride is difficult. Nonetheless, acute exposure to hydrogen fluoride results in mild lower respiratory tract (for example, tracheobronchial) lymphocytic inflammation (at 0.9 ppm or higher for 1 h; Lund et al. 1999), a case report indicates the devel- opment of RADS after acute exposure to hydrogen fluoride (Franzblau and Sa- hakian 2003), and prolonged respiratory symptoms occurred in the general pub- lic after an accidental release of hydrogen fluoride (Dayal et al. 1992). Thus, on the basis of the weight of the evidence, it is difficult to discount the potential of a causal relationship between occupational exposure to hydrogen fluoride and asthma. There are few animal data on subchronic or chronic respiratory effects of hydrogen fluoride. Severe effects, including ocular opacity and ocular and nasal fluid discharge, occur in rats exposed at 25 ppm for multiple days (Placke et al. 1990). In the Battelle subchronic 90-day inhalation study (Placke and Griffin 

Hydrogen Fluoride 99 1991), 25% mortality was observed in male rats exposed to hydrogen fluoride at 10 ppm. Neither mortality nor histologic lesions were observed at 0.1 or 1 ppm for 90 days, but it is not clear that the critical regions of the nose were examined (Placke and Griffin 1991; Rosenholtz et al. 1963). On the basis of body-weight reductions, the study authors considered 1 ppm to represent a NOAEL. The sub- chronic animal exposure study of Stokinger (1949) included only a small num- ber of animals, and complete histopathologic evaluation was not performed, so use of the results is problematic. In that study, one of five dogs exposed to hy- drogen fluoride at 8.6 ppm 6 h/day, 5 days/week for 10 weeks exhibited focal pulmonary hemorrhage that was observable during necropsy, but no histologic examinations of the nose or lungs were included. Thus, the committee consid- ered the animal studies as inadequate for establishing a CEGL. If a human NOAEL were derived from the animal NOAEL of 1 ppm by applying an interspecies uncertainty factor of 3-10, the resulting value would not differ markedly from a NOAEL derived from occupational epidemiology in which a relative risk of asthma for hydrogen fluoride of 1.18 per 0.12 ppm was estimated. Given the weaknesses of the animal database and the weight that hu- man data should receive for risk assessment, it is appropriate to base the 90-day CEGL on the human occupational epidemiologic data. Because the other agents in the workplace (such as particulate fluorides and sulfur dioxide) and the short- term high-exposure excursions may have contributed to the asthma risk in the workers, the relative risk for hydrogen fluoride of 1.18 per 0.12 ppm may over- estimate the health risk. Therefore, 0.12 ppm is a reasonable point of departure for CEGL derivation. A concentration-time extrapolation of the data is problem- atic; workers in the study of Taiwo et al. (2006) were exposed 8 h/day, 5 days/week for at least 2 years compared with submariners’ exposure 24 h/day for 90-day durations. The two scenarios may represent similar exposure concen- trations on an annualized basis; therefore, a concentration-time extrapolation is not proposed. The worker population in the epidemiologic study most likely included people who had rhinitis or were otherwise sensitive, so an interindi- vidual uncertainty factor is not suggested here. Although the occupational epi- demiologic study of Taiwo et al. (2006) was well performed, uncertainties are associated with using this study, particularly the uncertainty introduced by ex- cluding people who had a new diagnosis of asthma within 2 years of beginning work. On the basis of that uncertainty and the difficulty of extrapolating a typi- cal 5 days/week occupational exposure to the submarine setting, a database un- certainty factor of 3 was applied. The resulting 90-d CEGL is 0.04 ppm. Sensory irritation is not expected at this exposure level on the basis of the 1-h and multi- ple 6-h human studies by Lund et al. (1997) and Largent (1961). The proposed 90-d CEGL of 0.04 ppm most likely protects against sys- temic fluoride-induced toxicity if airborne hydrogen fluoride is the only impor- tant source of fluoride exposure. At a ventilation rate of 15 m3/day, it corre- sponds to a total absorbed dose of 0.5 mg/day (0.007 mg/kg-day for a 70-kg person). However, for systemic effects, it is necessary to consider fluoride expo- sure from all sources (EPA 1988; NRC 2006). Furthermore, it is necessary to 

100 Exposure Guidance Levels for Selected Submarine Contaminants consider effects that could occur within a 90-day exposure and long-term effects due to cumulative exposures, of which 90-day exposures on a submarine would be a part. Health end points that require long-term exposure or accumulation of fluoride include effects on bones (increased risk of fracture and of skeletal fluorosis). Other effects (such as endocrine effects) do not necessarily require accumulation or long-term exposure but may depend on current physiologic fluoride concentrations. In particular, subchronic thyroid effects have been re- ported in animals—an indication that altered concentrations of thyroid hormones do not necessarily require long exposure. Impaired glucose tolerance and male reproductive effects have also been reported in animals exposed to fluoride for 28-100 days. Table 4-8 summarizes systemic fluoride effects and corresponding esti- mated intakes in humans. For most end points, systemic effects in generally healthy people have been reported in situations corresponding to estimated aver- age chronic intakes of around 0.05 mg/kg-day or higher. Long-term fluoride exposure would include exposure to airborne and in- gested fluoride on board a submarine and on shore. Table 4-9 summarizes the estimated average fluoride intake (by source and total from all sources) based on information provided to the committee by the U.S. Navy (LCDR D. Martin, U.S. Navy, personal commun., July 8, 2008). The proposed 90-d CEGL of 0.04 ppm based on chronic respiratory effects would lead to a systemic fluoride intake of 0.5 mg/day (0.007 mg/kg-day) at an inhalation rate of 15 m3/day (EPA 1997) and an average body weight of 70 kg. That value would correspond to an esti- mated average total systemic fluoride intake (from all sources) of 0.023 or 0.026 mg/kg-day for normal or high activity levels, respectively (Table 4-9). That es- timated total daily fluoride intake is less than 0.05 mg/kg-day, the lowest dose associated with the potential for fluoride-induced systemic toxicity (Table 4-8). Persons who have fluoridated water on shore will have total systemic fluoride intakes on board a submarine that will be lower than those on shore. TABLE 4-8 Summary of Systemic Effects in Humans Associated with Chronic Intake of Fluoride from All Sources Typical Fluoride Intake Estimated NOAEL Associated with Effects Effects (mg/kg-day) (mg/kg-day) Endocrine effects 0.01-0.03 0.05 Increased risk of bone fracture NA ~0.05 Skeletal fluorosis (stage II) NA 0.05 Reduced testosterone concentrations <0.03 0.05 NA, not available. 

Hydrogen Fluoride 101 TABLE 4-9 Estimated Fluoride Intakes (mg/kg-day) for Specified Exposure Situationsa Source of Fluoride Exposure On Board Submarine On Shore b Drinking water (normal 0.0024 Fluoridated source:c 0.0173 activity) Nonfluoridated source:c 0.0024 Drinking water (high 0.005 Fluoridated source: 0.05 activity)d Nonfluoridated source: 0.005 Food and beveragese 0.0114 0.0114 f Pesticides 0.0007 0.0007 Toothpasteg 0.0014 0.0014 h Air 0.007 0.0006 Totals: Normal activity 0.023 Fluoridated source: 0.031 Nonfluoridated source: 0.017 High activity 0.026 Fluoridated source: 0.064 Nonfluoridated source: 0.019 a Based on NRC (2006) estimates for U.S. adults 20-49 years old unless otherwise indi- cated. b Assumes that drinking water on submarine is primarily from reverse-osmosis unit. Puri- fied water is expected to have low fluoride concentrations (<0.15 mg/L; NRC 2006). c Drinking water on shore could be from fluoridated sources (around 1 mg/L) or non- fluoridated sources (defined as <0.7 mg/L; assumed here to be 0.5 mg/L). Two types of sources are considered separately. d Assumes a drinking-water intake of 50 mL/kg of body weight per day (Table 2-4, NRC 2006) and fluoride concentrations of 1 mg/L (fluoridated source) or 0.1 mg/L (on board submarine or nonfluoridated source on shore). e Food on board submarine includes fresh and frozen ingredients, canned soups and vege- tables, and canned fruits and fruit juices. Commercial beverages (such as soft drinks and bottled tea) are available. Tea, coffee, and Kool-Aid are available. Tea is prepared from tea bags. This diet is considered comparable with average diet of adults in United States with respect to fluoride intake. f Exposure to fluoride from pesticides is considered typical for adults in United States. g Toothpaste use and inadvertent ingestion of toothpaste are considered typical for adults in United States. h Based on proposed 90-d CEGL of 0.04 ppm, inhalation rate of 15 m3/day, and average body weight of 70 kg. DATA ADEQUACY AND RESEARCH NEEDS NRC (2006) identified a number of research needs regarding fluoride toxi- cology for various health end points. In particular, nearly all human studies re- quire improved characterization of fluoride exposure, including individual fluo- ride intake. There are very few subchronic or short-term studies of humans with any route of exposure. There have been occupational studies and studies of ex- 

102 Exposure Guidance Levels for Selected Submarine Contaminants posures to airborne fluoride from coal combustion in China, but most of the lit- erature on effects in humans comes from ingestion exposures, primarily to fluo- ride in drinking water. Although many health end points (such as effects on bones) require long-term exposure or accumulation of fluoride, others (such as endocrine effects, low tolerance or hypersensitivity, and asthma) do not. A criti- cal research need for animal studies would be a 90-day continuous-inhalation bioassay for hydrogen fluoride. Studies that use multiple 90-day exposures that mimic the exposure of the submarine crew may be needed to examine fully the potential for the induction of asthma. A species appropriate for examination of induction of allergic airway disease and measures of airway function would be critical for such a study. REFERENCES Aardema, M.J., and T. Tsutsui. 1995. Sodium fluoride-induced chromosome aberrations in different cell cycle stages. Mutat. Res. 331(1):171-172. Aardema, M.J., D.P. Gibson, and R.A. LeBoeuf. 1989. Sodium fluoride-induced chromo- some aberrations in different stages of the cell cycle: A proposed mechanism. Mu- tat. Res. 223(2):191-203. ACGIH (American Conference of Governmental Industrial Hygienists). 2005. Threshold Limit Values (TLVs) for Chemical Substances and Physical Agents and Biological Exposure Indices (BEIs) for 2005. American Conference of Governmental Hy- gienists, Cincinnati, OH. Alarie, Y. 1973. Sensory irritation by airborne chemicals. CRC Crit. Rev. Toxicol. 2(3):299-363. Alexeeff, G.V., D.C. Lewis, and N.L. Ragle. 1993. Estimation of potential health effects from acute exposure to hydrogen fluoride using a “benchmark dose” approach. Risk Anal. 13(1):63-69. Amoore, J.E., and E. Hautala. 1983. Odor as an aid to chemical safety: Odor thresholds compared with threshold limit values and volatilities for 214 industrial chemicals in air and water dilution. J. Appl. Toxicol. 3(6):272-290. Ando, M., M. Tadano, S. Yamamoto, K. Tamura, S. Asanuma, T. Watanabe, T. Kondo, S. Sakurai, R. Ji, C. Liang, X. Chen, Z. Hong, and S. Cao. 2001. Health effects of fluoride pollution caused by coal burning. Sci. Total Environ. 271(1-3):107-116. Araibi, A.A., W.H. Yousif, and O.S. Al-Dewachi. 1989. Effect of high fluoride on the reproductive performance of the male rat. J. Biol. Sci. Res. 20(1):19-30. ATSDR (Agency for Toxic Substances and Disease Registry). 2003. Toxicological Pro- file for Fluorides, Hydrogen Fluoride, and Fluorine. U.S. Department of Health and Human Services, Public Health Service, Agency for Toxic Substances and Disease Registry, Atlanta, GA. September 2003 [online]. Available: http:// www.atsdr.cdc.gov/toxprofiles/tp11-p.pdf [accessed Mar. 27, 2009]. ATSDR (Agency for Toxic Substances and Disease Registry). 2008. ATSDR Minimal Risk Levels (MRLs). U.S. Department of Health and Human Services, Agency for Toxic Substances and Disease Registry, Atlanta, GA. December 2008 [online]. Available: http://www.atsdr.cdc.gov/mrls/pdfs/atsdr_mrls_december_2008.pdf [accessed Mar. 27, 2009]. 

Hydrogen Fluoride 103 Bachinskii, P.P., O.A. Gutsalenko, N.D. Naryzhniuk, V.D. Sidora, and A.I. Shliakhta. 1985. Action of the body fluorine of healthy persons and thyroidopathy patients on the function of hypophyseal-thyroid the system [in Russian]. Probl. Endokrinol. 31(6):25-29. Baud, C.A., R. Lagier, G. Boivin, and M.A. Boillat. 1978. Value of bone biopsy in the diagnosis of industrial fluorosis. Virchows Arch. A Pathol. Anat. Histol. 380(4):283-297. Bennion, J.R., and A. Franzblau. 1997. Chemical pneumonitis following household expo- sure to hydrofluoric acid. Am. J. Ind. Med. 31(4):474-478. Björnhagen, V., J. Höjer, C. Karlson-Stiber, A.I. Seldén, and M. Sundbom. 2003. Hydro- fluoric acid-induced burns and life-threatening systemic poisoning: Favorable out- come after hemodialysis. J. Toxicol. Clin. Toxicol. 41(6):855-860. Black, A., and R.F. Hounam. 1968. Penetration of iodine vapour through the nose and mouth and the clearance and metabolism of deposited iodine. Ann. Occup. Hug. 11(3):209-225. Bobek, S., S. Kahl, and Z. Ewy. 1976. Effect of long-term fluoride administration on thyroid hormones level in blood in rats. Endocrinol. Exp. 10(4):289-295. 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. Budavari, S., M.J. O’Neil, A. Smith, and P.E. Heckelman, eds. 1989. Hydrogen fluoride. P. 760 in The Merck Index: An Encyclopedia of Chemicals, Drugs, and Biologi- cals, 11th Ed. Rahway, NJ: Merck. Collings, G.H., Jr., R.B. Fleming, and R. May. 1951. Absorption and excretion of inhaled fluorides. AMA Arch. Ind. Hyg. Occup. Med. 4(6):585-590. Collings, G.H., Jr., R.B. Fleming, R. May, and W.O. Bianconi. 1952. Absorption and excretion of inhaled fluorides; further observations. AMA Arch. Ind. Hyg. Occup. Med. 6(4):368-373. Dalbey, W., B. Dunn, R. Bannister, W. Daughtrey, C. Kirwin, F. Reitman, M. Wells, and J. Bruce. 1998. Short-term exposures of rats to airborne hydrogen fluoride. J. Toxicol. Environ. Health Part A 55(4):241-275. Das (Sarkar), S., R. Maiti, and D. Ghosh. 2006. Management of fluoride induced testicu- lar disorders by calcium and vitamin-E co-administration in the albino rat. Reprod. Toxicol. 22(4):606-612. Dayal, H.H., M. Brokwick, R. Morris, T. Baranowski, N. Trieff, J.A. Harrison, J.R. Lisse, and G.A. Ansari. 1992. A community-based epidemiologic study of health sequelae of exposure to hydrofluoric acid. Ann. Epidemiol. 2(3):213-230 (as cited in ATSDR 2003). Dayal, H.H., T. Baranowski, Y.H. Li, and R. Morris. 1994. Hazardous chemicals: Psy- chological dimensions of the health sequelae of a community exposure in Texas. J. Epidemiol. Community Health 48(6):560-568. de Lopez, O.H., F.A. Smith, and H.G. Hodge. 1976. Plasma fluoride concentrations in rats acutely poisoned with sodium fluoride. Toxicol. Appl. Pharmacol. 37(1):75- 83. Derryberry, O.M., M.D. Bartholomew, and R.B. Fleming. 1963. Fluoride exposure and worker health. The health status of workers in a fertilizer manufacturing plant in relation to fluoride exposure. Arch. Environ. Health 6:503-514. Dominok, G., K. Siefert, J. Frege, and B. Dominok. 1984. Fluoride content of bones of retired fluoride workers. Fluoride 17(1):23-26. 

104 Exposure Guidance Levels for Selected Submarine Contaminants Dvoráková-Hortová, K., M. Sandera, M. Jursová, J. Vasinová, and J. Pecknicová. 2008. The influence of fluorides on mouse sperm capacitation. Anim. Reprod. Sci. 108(1-2):157-170. EPA (U.S. Environmental Protection Agency). 1988. Summary Review of Health Effects Associated with Hydrogen Fluoride and Related Compounds: Health Issue As- sessment. EPA/600/8-89/002F. Environmental Criteria and Assessment Office, Of- fice of Health and Environmental Assessment, Office of Research and Develop- ment, U.S. Environmental Protection Agency, Research Triangle Park, NC. December 1988 [online]. Available: http://cfpub.epa.gov/ncea/cfm/recordisplay.. cfm?deid=47539 [accessed Mar. 27, 2009]. EPA (U.S. Environmental Protection Agency). 1989. Fluorine (Soluble Fluoride) (CASRN 7782-41-4). Integrated Risk Information System, U.S. Environmental Protection Agency [online]. Available: http://www.epa.gov/ncea/iris/subst/0053. htm [accessed June 22, 2009]. EPA (U.S. Environmental Protection Agency). 1994. Methods for Derivation of Inhala- tion Reference Concentrations and Application of Inhalation Dosimetry. EPA/600/8-90/066F. Environmental Criteria and Assessment Office, Office of Health and Environmental Assessment, Office of Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC. October 1994 [online]. Available: http://www.epa.gov/raf/publications/pdfs/RFC METHODOLOGY.PDF [accessed Mar. 27, 2009]. EPA (U.S. Environmental Protection Agency). 1997. Exposure Factors Handbook, Vol. 1. General Factors. EPA/600/P-95/002Fa. National Center for Environmental As- sessment, Office of Research and Development, U.S. Environmental Protection Agency, Washington DC. August 1997 [online]. Available: http://www.epa. gov/ncea/efh/report.html [accessed Mar. 27, 2009]. Franke, J., and E. Auermann. 1972. Significance of iliac crest puncture with histological and microanalytical examination of the obtained bone material in the diagnosis of fluorosis [in German]. Int. Arch. Arbeitsmed. 29(2):85-94. Franke, J., F. Rath, H. Runge, F. Fengler, E. Auermann, and G.L. Lenart. 1975. Industrial fluorosis. Fluoride 8(2):61-85. Franzblau, A., and N. Sahakian. 2003. Asthma following household exposure to hydro- fluoric acid. Am. J. Ind. Med. 44(3):321-324. Gadhia, P.K., and S. Joseph. 1997. Sodium fluoride induced chromosome aberrations and sister chromatid exchange in cultured human lymphocytes. Fluoride 30(3):153- 156. Galletti, P.M., and G. Joyet. 1958. Effect of fluorine on thyroidal iodine metabolism in hyperthyroidism. J. Clin. Endocrinol. Metab. 18(10):1102-1110. Grandjean, P., and J.H. Olsen. 2004. Extended follow-up of cancer incidence in fluoride- exposed workers. J. Natl. Cancer Inst. 96(10):802-803 (as cited in NRC 2006). Grandjean, P., and G. Thomsen. 1983. Reversibility of skeletal fluorosis. Br. J. Ind. Med. 40(4):456-461. Grandjean, P., M. Hørder, and Y. Thomassen. 1990. Fluoride, aluminum, and phosphate kinetics in cryolite workers. J. Occup. Med. 32(1):58-63. Grandjean, P., J.H. Olsen, O.M. Jensen, and K. Juel. 1992. Cancer incidence and mortal- ity in workers exposed to fluoride. J. Natl. Cancer Inst. 84(24):1903-1909 (as cited in NRC 2006). Grimbergen, G.W. 1974. A double blind test for determination of intolerance to fluori- dated water. (Preliminary Report). Fluoride 7(3):146-152. Guan, Z.Z., Z.J. Zhuang, P.S. Yang, and S. Pan. 1988. Synergistic action of iodine- 

Hydrogen Fluoride 105 deficiency and fluorine-intoxication on rat thyroid. Chin. Med. J. 101(9):679-684. Hagar, R. 2008. Submarine Atmosphere Control and Monitoring Brief for the COT Committee. Presentation to the First Meeting on Emergency and Continuous Ex- posure Guidance Levels for Selected Submarine Contaminants, June 17, 2008, Washington, DC. Hara, K. 1980. Studies on fluorosis, especially effects of fluoride on thyroid metabolism [in Japanese]. Koku Eisei Gakkai Zasshi 30(1):42-57 (as cited in NRC 2006). Hodge, H.C., and F.A. Smith. 1977. Occupational fluoride exposure. J. Occup. Med. 19(1):12-39. Holstege, C., A. Baer, and W.J. Brady. 2005. The electrocardiographic toxidrome: The ECG presentation of hydrofluoric acid ingestion. Am. J. Emerg. Med. 23(2):171- 176. Horton, D.K., Z. Berkowitz, and W.E. Kaye. 2004. Hydrofluoric acid releases in 17 states and the acute health effects associated, 1993-2001. J. Occup. Environ. Med. 46(5):501-508. HSDB (Hazardous Substances Data Bank). 2008. Hydrogen Fluoride (CASRN: 7664-39- 3). TOXNET, Specialized Information Services, U.S. National Library of Medi- cine, Bethesda, MD [online]. Available: http://toxnet.nlm.nih.gov/ [accessed Mar. 27, 2009]. IARC (International Agency for Research on Cancer). 1987. Fluorides (Inorganic, Used in Drinking Water)/ Supplement 7, IARC [online]. http://www.inchem.org/ documents/iarc/suppl7/fluorides.html [accessed June 22, 2009]. Jooste, P.L., M.J. Weight, J.A. Kriek, and A.J. Louw. 1999. Endemic goitre in the ab- sence of iodine deficiency in schoolchildren of the Northern Cape Province of South Africa. Eur. J. Clin. Nutr. 53(1):8-12. Kaltreider, N.L., M.J. Elder, L.V. Cralley, and M.O. Colwell. 1972. Health survey of aluminum workers with special reference to fluoride exposure. J. Occup. Med. 14(7):531-541. Kongerud, J., J. Boe, V. Soyseth, A. Naalsund, and P. Magnus. 1994. Aluminum potroom asthma: The Norwegian experience. Eur. Respir. J. 7(1):165-172. Kishi, K., and T. Ishida. 1993. Clastogenic activity of sodium fluoride in great ape cells. Mutat. Res. 301(3):183-188. Largent, E.J. 1961. Pp. 34-39, 43-48 in Fluorosis: The Health Aspects of Fluorine Com- pounds. Columbus, OH: Ohio State University Press. Largent, E.J., P.G. Bovard, and F.F. Heyroth. 1951. Roentgenographic changes and uri- nary fluoride excretion among workmen engaged in the manufacture of inorganic fluorides. Am. J. Roentgenol. Radium Ther. 65(1):42-48. Lasne, C., Y.P. Lu, and I. Chouroulinkov. 1988. Transforming activities of sodium fluo- ride in cultured Syrian hamster embryo and BALB/3T3 cells. Cell Biol. Toxicol. 4(3):311-324. Lestari, F., A.J. Hayes, A.R. Green, and B. Markovic. 2005. In vitro cytotoxicity of se- lected chemicals commonly produced during fire combustion using human cell lines. Toxicol. In Vitro 19(5):653-663. Lindahl, C.B., and T. Mahmood. 2005. Fluorine compounds, inorganic. Pp. 852-858 in Kirk-Othmer Encyclopedia of Chemical Technology, Vol. 11, A. Seidel et al. eds. Hoboken, NJ: Wiley-Interscience. Liu, J.L., T. Xia, Y.Y. Yu, X.Z. Sun, Q. Zhu, W. He, M. Zhang, and A. Wang. 2005. The dose-effect relationship of water fluoride levels and renal damage in children [in Chinese]. Wei Sheng Yan Jiu 34(3):287-288. 

106 Exposure Guidance Levels for Selected Submarine Contaminants Lund, K., J. Ekstrand, J. Boe, P. Sostrand, and J. Kongerud. 1997. Exposure to hydrogen fluoride: An experimental study in humans of concentrations of fluoride in plasma, symptoms, and lung function. Occup. Environ. Med. 54(1):32-37. Lund, K., M. Refsnes, T. Sanstrom, P. Sostrand, P. Schwarze, J. Boe, and J. Kongerud. 1999. Increased CD3 positive cells in bronchoalveolar lavage fluid after hydrogen fluoride inhalation. Scand. J. Work Environ. Health 25(4):326-334. Lund, K., M. Refsnes, I. Ramis, C. Dunster, J. Boe, P.E. Schwarze, E. Skovlund, F.J. Kelly, and J. Kongerud. 2002. Human exposure to hydrogen fluoride induces acute neutrophilic, eicosanoid, and antioxidant changes in nasal lavage fluid. Inhal. Toxicol. 14(2):119-132. Lund, K., C. Dunster, I. Ramis, T. Sandström, F.J. Kelly, P. Sostrand, P. Schwarze, E. Skovlund, J. Boe, J. Kongerud and M. Refsnes. 2005. Inflammatory markers in bronchoalveolar lavage fluid from human volunteers 2 hours after hydrogen fluo- ride exposure. Hum. Exp. Toxicol. 24(3):101-108. Machle, W.F., and K. Kitzmiller. 1935. The effects of the inhalation of hydrogen fluo- ride. II. The response following exposure to low concentration. J. Ind. Hyg. 17:223-229. Machle, W.F., F. Thamann, K. Kitzmiller, and J. Cholak. 1934. The effects of the inhala- tion of hydrogen fluoride. I. Response following exposure to high concentrations. J. Ind. Health 16(2):129-145. Meng, Z., and B. Zhang. 1997. Chromosomal aberrations and micronuclei in lympho- cytes of workers at a phosphate fertilizer factory. Mutat. Res. 393(3):283-288. Meng, Z., H. Meng, and X. Cao. 1995. Sister-chromatid exchanges in lymphocytes of workers at a phosphate fertilizer factory. Mutat. Res. 334(2):243-246. Mihashi, M., and T. Tsutsui. 1996. Clastogenic activity of sodium fluoride to rat vertebral body-derived cells in culture. Mutat. Res. 368(1):7-13. Mitsui, G., T. Dote, K. Adachi, E. Dote, K. Fujimoto, Y. Shimbo, M. Fujihara, H. Shi- mizu, K. Usuda, and K. Kono. 2007. Harmful effects and acute lethal toxicity of intravenous administration of low concentrations of hydrofluoric acid in rats. Toxicol. Ind. Health 23(1):5-12. Morris, J.B. 1979. The Absorption, Distribution and Excretion of Inhaled Hydrogen Fluo- ride in the Rat. Ph.D. Thesis, University of Rochester, Rochester, NY. 293 pp. Morris, J.B. 2006. Nasal toxicology. Pp. 349-371 in Inhalation Toxicology, 2nd Ed., H. Salem, and S.A. Katz, eds. Boca Raton: CRC/Taylor and Francis. Morris, J.B., and F.A. Smith. 1982. Regional deposition and absorption of inhaled hydro- gen fluoride in the rat. Toxicol. Appl. Pharmacol. 62(1):81-89. Morris, J.B., P.T. Symanowicz, J.E. Olsen, R.S. Thrall, M.M. Cloutier, and A.K. Hub- bard. 2003. Immediate sensory-nerve mediated respiratory responses to irritants in healthy and allergic airway diseased mice. J. Appl. Physiol. 94(4):1563-1571. NIOSH (National Institute for Occupational Safety and Health). 2005. NIOSH Pocket Guide to Chemical Hazards. DHHS (NIOSH). No. 2005-149. National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention, U.S. Department of Health and Human Services, Cincinnati, OH [online]. Available: http://www.cdc.gov/niosh/npg/ [accessed Mar. 30, 2009]. NRC (National Research Council). 1988. Submarine Air Quality: Monitoring the Air in Submarines. Washington, DC: National Academy Press. NRC (National Research Council). 2004. Hydrogen fluoride. Pp. 123-197 in Acute Expo- sure Guideline Levels for Selected Airborne Chemicals, Vol. 4. Washington, DC: The National Academies Press. 

Hydrogen Fluoride 107 NRC (National Research Council). 2006. Fluoride in Drinking Water: A Scientific Re- view of EPA’s Standards. Washington, DC: The National Academies Press. NTP (National Toxicology Program). 1990. Toxicology and Carcinogenesis Studies of Sodium Fluoride (CAS No. 7681-49-4) in F344/N Rats and B6C3F1 Mice. NTP Technical Report No. 393. NIH Publication No. 91-2848. U.S. Department of Health and Human Services, National Toxicology Program, Research Triangle Park, NC. OEHHA (Office of Environmental Health Hazard Assessment). 2003. Chronic Toxicity Summary: Fluorides including Hydrogen Fluoride. Determination of Noncancer Chronic Reference Exposure Levels. August 2003 [online]. Available: http://www. oehha.ca.gov/air/chronic_rels/pdf/2ApnA_Fluoride_Final.pdf [accessed September 19, 2008]. Oguro, A., J. Cervenka, and K. Horii. 1995. Effect of sodium fluoride on chromosomal ploidy and breakage in cultured human diploid cells (IMR-90): An evaluation of continuous and short-time treatment. Pharmacol. Toxicol. 76(4):292-296. Ortiz-Pérez, D., M. Rodríguez-Martínez, F. Martínez, V.H. Borja-Aburto, J. Castelo, J.I. Grimaldo, E. de la Cruz, L. Carrizales, and F. Díaz-Barriga. 2003. Fluoride- induced disruption of reproductive hormones in men. Environ. Res. 93(1):20-30. Placke, M.E., and S. Griffin. 1991. Subchronic Inhalation Exposure Study of Hydrogen Fluoride in Rats. Battelle Memorial Institute, Columbus, OH. Submitted by Bat- telle Washington Environmental Program, Arlington VA. January 4, 1991. Placke, M.E., M. Brooker, R. Persing, J.Taylor, and M. Haterty. 1990. Final Report on Repeated-Exposure Inhalation Study of Hydrogen Fluoride in Rats. Battelle Me- morial Institute, Columbus, OH. Submitted by Battelle Washington Environmental Program, Arlington VA. September 1990. Rees, D., D.B. Rama, and V. Yousefi. 1990. Fluoride in workplace air and in urine of workers concentrating fluorspar. Am. J. Ind. Med. 17(3):311-320. Rigalli, A., J.C. Ballina, and R.C. Puche. 1992. Bone mass increase and glucose tolerance in rats chronically treated with sodium fluoride. Bone Miner. 16(2):101-108. Rigalli, A., R. Alloatti, I. Menoyo, and R.C. Puche. 1995. Comparative study of the effect of sodium fluoride and sodium monofluorophosphate on glucose homeostasis in the rat. Arzneimittel-Forschung. 45(3):289-292. Roholm, K. 1937. Fluorine Intoxication: A Clinical-Hygienic Study, with a Review of the Literature and Some Experimental Investigations. London: H.K. Lewis. Romundstad, P., A. Andersen, and T. Haldorsen. 2000. Cancer incidence among workers in six Norwegian aluminum plants. Scand. J. Work Environ. Health 26(6):461-469 (as cited in NRC 2006). Rosenholtz, M.J., T.R. Carson, M.H. Weeks, F. Wilinski, D.F. Ford, and F.W. Oberst. 1963. A toxicopathologic study in animals after brief single exposures to hydrogen fluoride. Am. Ind. Hyg. Assoc. J. 24:253-261. Sadilova, M.S., K.P. Selyankina, and O.K. Shturkina. 1965. Experimental studies on the effect of hydrogen fluoride on the central nervous system [in Russian]. Gig. Sanit. 30(5):155-160. Sanz-Gallén, P., S. Nogué, P. Munné, and A. Faraldo. 2001. Hypocalcaemia and hypo- magnesaemia due to hydrofluoric acid. Occup. Med. 51(4):294-295. Schlegel, H.H. 1974. Industrial skeletal fluorosis: Preliminary report on 61 cases from aluminum smelter [in German]. Soz. Praventiv. Med. 19:269-274. Shusterman, D., A. Tarun, M.A. Murphy, and J. Morris. 2005. Seasonal allergic rhinitic and normal subjects respond differentially to nasal provocation with acetic acid vapor. Inhal. Toxicol. 17(3):147-152. 

108 Exposure Guidance Levels for Selected Submarine Contaminants Singh, P.P., M.K. Barjatiya, S. Dhing, R. Bhatnagar, S. Kothari, and V. Dhar. 2001. Evi- dence suggesting that high intake of fluoride provokes nephrolithiasis in tribal populations. Urol. Res. 29(4):238-244 (as cited in NRC 2006). Smith, F.A., and H.C. Hodge. 1979. Airborne fluorides and man: Part II. Crit. Rev. Env. Contr. (1):1-25. Soyseth, V., and J. Kongerud. 1992. Prevalence of respiratory disorders among aluminum potroom workers in relation to exposure to fluoride. Br. J. Ind. Med. 49(2):125- 130. Stavert, D.M., D.C. Archuleta, M.J. Behr, and B.E. Lehnert. 1991. Relative acute toxici- ties of hydrogen fluoride, hydrogen chloride and hydrogen bromide in nose- and pseudo-mouth-breathing rats. Fundam. Appl. Toxicol. 16(4):636-655. Stokinger, H.E. 1949. Toxicity following inhalation of fluorine and hydrogen fluoride. Pp. 1021-1057 in Pharmacology and Toxicology of Uranium Compounds, Vol. 1, C. Voegtlin, and H.C. Hodge, eds. New York: McGraw-Hill. Susheela, A.K., and P. Jethanandani. 1996. Circulating testosterone levels in skeletal fluorosis patients. J. Toxicol. Clin. Toxicol. 34(2):183-189. Susheela, A.K., M. Bhatnagar, K. Vig, and N.K. Mondal. 2005. Excess fluoride ingestion and thyroid hormone derangements in children living in Delhi, India. Fluoride 38(2):98-108. Taiwo, O.A., K.D. Sircar, M.D. Slade, L.F. Cantley, S.J. Vegso, P.M. Rabinowitz, M.G. Fiellin, and M.R. Cullen. 2006. Incidence of asthma among aluminum workers. J. Occup. Environ. Med. 48(3):275-282. ten Berge, W.F., A. Zwart, and L.M. Appelman. 1986. Concentration-time mortality response relationship of irritant and systemically acting vapours and gases. J. Haz- ard. Mater. 13(3):151-162. Teotia, S.P., M. Teotia, R.K. Singh, D.R. Taves, and S. Heels. 1978. Endocrine aspects of endemic skeletal fluorosis. J. Assoc. Physicians India 26(11):995-1000. Tokar, V.I., and O.N. Savchenko. 1977. Effect of inorganic fluorine compounds on the functional state of the pituitary-testis system [in Russian]. Probl. Endokrinol. 23(4):104-107. Trivedi, N., A. Mithal, S.K. Gupta, and M.M. Godbole. 1993. Reversible impairment of glucose tolerance in patients with endemic fluorosis. Fluoride Collaborative Study Group. Diabetologia 36(9):826-828. Turner, C.H., L.P. Garetto, A.J. Dunipace, W. Zhang, M.E. Wilson, M.D. Grynpas, D. Chachra, R. McClintock, M. Peacock, and G.K. Stookey. 1997. Fluoride treatment increased serum IGF-1, bone turnover, and bone mass, but not bone strength, in rabbits. Calcif. Tissue Int. 61(1):77-83. Turner, C.H., W.R. Hinckley, M.E. Wilson, W. Zhang, and A.J. Dunipace. 2001. Com- bined effects of diets with reduced calcium and phosphate and increased fluoride intake on vertebral bone strength and histology in rats. Calcif. Tissue Int. 69(1):51- 57 (as cited in ATSDR 2003). Vohra, R., L.I. Velez, W. Rivera, F.L. Benitez, and K.A. Delaney. 2008. Recurrent life- threatening ventricular dysrhythmias associated with acute hydrofluoric acid inges- tion: Observations in one case and implications for mechanism of toxicity. Clin. Toxicol. 46(1):79-84. Waldbott, G.L. 1956. Incipient chronic fluoride intoxication from drinking water. II. Distinction between allergic reactions and drug intolerance. Int. Arch. Allergy Appl. Immunol. 9(5):241-249. Waldbott, G.L. 1958. Allergic reactions from fluorides. Int. Arch. Allergy Appl. Immu- nol. 12(6):347-355. 

Hydrogen Fluoride 109 Waldbott, G.L., and J.R. Lee. 1978. Toxicity from repeated low-grade exposure to hy- drogen fluoride—case report. Clin. Toxicol. 13(3):391-402. Wang, A.G., T. Xia, Q.L. Chu, M. Zhang, F. Liu, X.M. Chen, and K.D. Yang. 2004. Effects of fluoride on lipid peroxidation, DNA damage and apoptosis in human embryo hepatocytes. Biomed. Environ. Sci. 17(2):217-222. Wing, J.S., J.D. Brender, L.M. Sanderson, D.M. Perrotta, and R.A. Beauchamp. 1991. Acute health effects in a community after a release of hydrofluoric acid. Arch. En- viron. Health 46(3):155-160. Wu, D.Q., and Y. Wu. 1995. Micronucleus and sister chromatid exchange frequency in endemic fluorosis. Fluoride 28(3):125-127. Xiong, X., J. Liu, W. He, T. Xia, P. He, X. Chen, K. Yang, and A. Wang. 2007. Dose- effect relationship between drinking water fluoride levels and damage to liver and kidney functions in children. Environ. Res. 103(1):112-116. Yamamoto, S., K. Katagiri, M. Ando, and X.Q. Chen. 2001. Suppression of pulmonary antibacterial defenses mechanisms and lung damage in mice exposed to fluoride aerosol. J. Toxicol. Environ. Health A 62(6):485-494. Yokoyama, E., R. Yoder, and N.R. Frank. 1971. Distribution of 35S in blood and its ex- cretion in urine of dogs exposed to 35SO2. Arch. Environ. Health 22(3):389-395. Young, J.T. 1981. Histopathologic examination of the rat nasal cavity. Fundam. Appl. Toxicol. 1(4):309-312. Zhao, W., H. Zhu, Z. Yu, K. Aoki, J. Misumi, and X. Zhang. 1998. Long-term effects of various iodine and fluorine doses on the thyroid and fluorosis in mice. Endocr. Regul. 32(2):63-70 (as cited in NRC 2006).