Spacecraft Water Exposure Guidelines for Selected Contaminants: Volume 2 (2007)

Raghupathy Ramanathan, Ph.D. NASA-Johnson Space Center Toxicology Group Houston, Texas

Barium (Ba) salts have a range of solubilities in water. Ba metal is insoluble in water but soluble in alcohol. The three least soluble Ba salts are the sulfate, the carbonate, and the sulfide. The acetate, the cyanide, the chloride, and the nitrates and alkaline salts, such as the oxides and hydroxide, are highly soluble in water (see Table 3-1), some even at 0°C. In general, the acid-soluble Ba salts are very toxic when compared with the least soluble salts, such as Ba sulfate (BaSO₄).

Ba is one of the alkaline earth metals. It occurs in nature as a free metal and as salts. It is also produced for various industrial uses. The Ba salt most commonly found in the earth’s crust is BaSO₄, which is found in limestone (barite), shales, and rocky sediments. In a crushed form, it is the source for several other Ba compounds.

The major use of BaSO₄ is in the oil and gas industry to make lubricant muds for drilling. Ba in salt forms has been reported to be present in almost all surface waters (at 2-340 micrograms per liter [g/L]) (Kopp and Kroner 1967). The release of Ba compounds from Ba manufacturing and processing plants and from sedimentary rocks by leaching in certain areas might be the reason for its presence in surface waters. The finished water of public systems frequently contains Ba at 1-172 g/L (Tuovinen 1980). A mean concentration of 43 g/L and a maximum of 380 g/L were reported in the largest U.S. cities (NRC 1977, pp. 229-230)

McCabe et al. (1970) and Calabrese (1977) reported that Ba was present in about 2,600 analyzed drinking water samples. It was found at about 1.5 milligrams (mg) per L in samples from areas of northern Illinois and northern Iowa. Because the solubility of Ba depends on the concentrations of total sulfate in the medium and because sufficient concentrations of sulfates are in the natural waters, it is difficult to maintain more than 1.5 mg/L in water (EPA 1985). The World Health Organization (WHO) has reported that the range of daily dietary intake of Ba is 300-1,770 g per day (4-25 µg per kilogram body weight per day). WHO (1990) also reported that the concentrations of Ba measured in U.S. drinking waters are 1-20 g/L. Several other reports indicated that the concentrations are much higher (Kopp 1969; Calabrese 1977).

Ba compounds are used to make not only drilling lubricants but also paints and pigments, textile dyes, greases, bricks, tile, glass, and rubber. Ba nitrate (Ba(NO₃)₂) is used in pyrotechnics. Because of its high radiopacity, BaSO₄ has been used by doctors for taking x-rays of the stomach, intestines, and respiratory and urinary tract systems and in bronchography. A high-density suspension of Ba sulfate—usually 340 g suspended in 150 mL of water—is administered orally for radiologic evaluation of the small bowel. This represents a dose of 4.89 g per kilogram (kg).

Humans might be exposed to Ba at high concentrations in occupational settings or in areas proximate to Ba mining and processing.

Only very scanty human data are available on the absorption of Ba and its salts from the gastrointestinal (GI) tract. Lisk et al. (1988) studied the absorption and excretion of selenium and Ba in humans who consumed Brazil nuts, which contain both. The nuts they used contained Ba at 1,953 parts per million (ppm). As part of the study, one male subject (68 kg) ingested nut meat equivalent to Ba at 179.2 mg (2.64 mg/kg) in a single dose. Total urine and feces for each day was collected for 15 days (d) after the dose. From the total amount of Ba excreted in feces and urine, which was only 10% of the ingested dose, it appears that 90% of the dose was absorbed. This is the highest percentage ever reported for Ba absorption. It is not clear what form of Ba is contained in Brazil nuts or how other materials, including fat, would have facilitated such a high absorption.

The absorption rate was estimated to be 5% in the adult (ICRP 1973) on the basis of results from two human volunteers. It has also been suggested that infants might have significantly greater absorption rates than adults (Lengemann 1959). For radiologic examination of the small bowel, a solution of Baricon, a highly insoluble suspension of BaSO₄ at about 4.89 g/kg, or Ba at 2.88 g/kg, is usually administered to patients. In clinical practice, the amount given to adults is not based on body weight; the commercial preparation of preweighed BaSO₄ (340 g) is suspended in 150 mL of water. According to the Merck Index (Merck 1989), BaSO₄ is practically insoluble in water (1 g in 400,000 parts), dilute acid, and alcohol. It is soluble only in hot, concentrated sulfuric acid. In the above example, the 150 mL of water dissolves BaSO₄ at only 0.375 mg (Ba at 0.003 mg/kg). Neither the pH of stomach acid nor the alkalinity of the intestinal contents contributes to the solubility of ingested BaSO₄ used in the clinical setting. Dallas and Williams (2001) cited a study from a student thesis (Bligh 1960) in which isotopic Ba was given intravenously (iv) or given in orange juice as an oral dose to cancer patients. Feces and urine samples were collected for 7-10 d, and radioactivity was deter-

mined. Bligh 1960, as cited in Dallas and Williams (2001), also performed equivalent experiments in 15-month- (mo) old female rats to compare Ba metabolism in humans with that in animals. The study concluded that an average of 9% of radioactive Ba was absorbed in humans and about 10% in rats.

In laboratory animals, absorption varies significantly with species, age, and composition of the diet. For example, Taylor et al. (1962) reported that although older rats absorbed 7% of ingested Ba chloride (BaCl₂), the amount absorbed by rats 22 d old or younger was at least 10 times higher. Fasted adult rats absorbed about 20%. GI absorption in dogs has been calculated to be about 7% (Cuddihy and Griffith 1972).

Measurements of Ba in the serum of dogs indicate that peak absorption from the GI system occurs within 1 h (Chou and Chin 1943). McCauley and Washington (1983) examined the effect of oral intubation of various anions of the Ba salt on the relative rates of uptake and tissue deposition in rats. Male Sprague-Dawley rats weighing 250-300 g were maintained on a diet of Ba at less than 1 mg/kg of food for at least 1 mo before the experiment. They were orally administered ¹³¹Ba as sulfate (SO₄), chloride (Cl), or carbonate (CO₃) at pH 7.0 (0.5 mL/100 g body weight of a 10 mg/L solution). Animals sacrificed at 2, 5, 10, 20, 30, 60, and 120 minutes (min) and 24 hours (h) after intubation. Amount of radioactivity was measured in the blood and various tissues at various intervals. One group of animals was also intubated with ¹³¹Ba as the chloride anion (BaCl₂) after the 24-h fast and studied at various times thereafter.

With ¹³¹BaCl₂, in the nonfasted rats, the maximum concentrations of ¹³¹Ba in blood was reached in 60 min, and at 24 h, the activity was about 90% of the maximum value, indicating a very slow rate of elimination. However, in the fasted rats, the peak radioactivity was measured at 15 min, and the radioactivity after 4 h was only 50% of the peak activity, indicating that fasting can affect both the rate of absorption and elimination. In the nonfasted rats, when ¹³¹Ba was intubated as the sulfate of the carbonate anion, the peak radioactivity in the blood was reached at 60 min, similar to the nonfasted BaCl₂ group. The authors stated that when large oral doses of BaSo4 are used in radiopaque x-ray diagnosis, only a very small fraction is absorbed. No overt toxic effects have been reported from “Ba swallows” (BaSO₄ administered orally) during diagnostic procedures.

From the available studies on the distribution of Ba in the human body, it appears that Ba distributes mostly (over 93% of the body burden) in the bones and teeth and to a small extent, in the eye, lungs, skin, and adipose tissue in humans at less than 1% of total body weight (Schroeder et al. 1972; ATSDR 1992).

From autopsy data, Sowden and Stitch (1957) reported that the Ba concentrations in human bones ranged from 7 ppm in children (0-3 y) to 8.5 ppm in adults (33-74 y), indicating no age-related accumulation. In mice iv injected with ¹³¹BaCl₂, the radioactivity was localized primarily in the bones, although distribution into other tissues was also observed (Dencker et al. 1976). Twenty-four hours after ¹³¹Ba (as ¹³¹BaCl₂) intubation, the ¹³¹Ba concentration measured in five tissues (activity per g tissue weight) was in the order: heart > eye > skeletal > muscle > kidney > liver. The concentration in the cardiac muscle was at least threefold higher than that of the eye. In addition, the accumulation of radioactivity in the eye from the BaSO₄ and the BaCl₂ groups were similar in the BaCO₃-intubated rats; it was 50% of that of the BaCl₂-treated group (McCauley and Washington 1983). The radioactivity in the bones was not reported.

In several studies, radioactive Ba was iv injected for the evaluation of the kinetics profile in urine and feces. In humans, the main excretion route for ingested Ba is the feces (about 72%); about 3% is excreted in the urine (Tipton et al. 1966). Harrison et al. (1967) reported that when 13³Ba was iv injected to one healthy 60-y-old man, 9% of the injected dose was excreted in the urine and 84% was excreted in the feces. In another study by the same authors, cumulatively 20% of the injected dose was excreted in urine and feces in 24 h, 70% in 3 d, and about 85% after 10 d (Tipton et al. 1966). The excretion half-life was about 35 h. According to the National Research Council (NRC) (1977), 20% of ingested Ba is excreted in the feces and 7% in the urine within 24 h. In dogs that received Ba by gavage, most of the Ba was excreted within a few days (Cuddihy and Griffith 1972). Syed et al. (1981) reported that 10-15% of a radioactive dose of Ba iv injected was excreted in the feces within the first 24 h after dosing; the report noted very similar rates in rats and mice. In one study, it was reported that rats and humans excreted Ba in

comparable ways, and fecal excretion of Ba was found to be greater than urinary excretion (Bligh 1960, as cited in Dallas and Williams 2001). Approximately 0.5% of the Ba dose was excreted into bile within 2 h in bile-duct-cannulated Sprague-Dawley rats iv injected with ¹³³BaCl₂ at 1.8 µg. In parallel to plasma concentrations, biliary Ba concentrations reached their peak in the first 15 min after administration and rapidly declined thereafter; Ba was not concentrated in the liver. Biliary excretion is not important for elimination when Ba is ingested (Edel et al. 1981). The kinetics of Ba elimination appear to have three phases. Rundo et al. (1967) estimated that the half-times of ¹³³Ba for these phases were 3, 6, 34.2, and 1,033 d.

A review of the literature on pharmacokinetics (absorption, distribution, and elimination) and toxicity (described in the sections below) indicates that there are several complexities in those areas. For example, although BaSO₄ is considered extremely insoluble in water, some studies comparing uptake kinetics could not differentiate BaSO₄ from highly soluble Ba compounds. There are wide variations in the amounts reported for Ba absorption as a function of dose that have not been described clearly in the literature. Although several studies have described the cardiovascular effects of Ba, BaSO₄ is used extensively in everyday diagnostic radiology without any serious adverse consequences that might cause it to be discontinued.

Ba, like other alkaline earth metals, is a bone-seeking element (Bauer et al. 1957), but its function in bone is not known. Ba can replace calcium and can carry out several calcium-mediated effects, such as the release of adrenal hormones and of neurotransmitters from adrenergic synapses (Douglass and Rubin 1964). Ba can also release catecholamines in the absence of acetylcholine. The most studied effects of Ba toxicity are vasoconstriction, hypertension, and those that occur in the muscular system. These end points have been the focus of several investigations of chronic low-dose exposure.

Several studies conducted in the late 1960s and early 1970s suggest that electrocardiogram (ECG) abnormalities were encountered during Ba

enema x-rays of older patients and patients with heart diseases. ECG abnormalities were pronounced in patients over the age of 60. Minor ECG abnormalities occurred in the lower age range in patients over 60 y of age (Berman et al. 1965; Eastwood 1972). In 1975, Roeske et al. studied patients over 60 and reported potentially serious effects such as incidence of arrhythmias and ST-segment changes during Ba enema x-ray studies. In 58 unselected, consecutive patients (ages 60-98 y old) undergoing routine Ba enema examinations, 12-lead ECG and 100-cycle cardiac rhythm recordings were performed, in addition to systemic arterial pressure measurements in the supine and upright positions. These tests were done before, during, and after the enema study. Subsequent ST-segment changes were also analyzed from the ECG recordings. In 27 of the 58 patients, abnormal ECG recordings (positive alterations as defined by stated criteria) were noted. Twenty-three patients developed significant arrhythmias during the Ba enema, four exhibited new ST-segment depressions, and 31 patients had negative results. The significant observations were that patients who were taking digitalis (an anti-arrhythmia medication) did not exhibit positive ECG alterations, whereas patients who were not receiving digitalis exhibited positive results. Furthermore, randomly chosen patients who received an iv dose of glucagon to prevent intestinal spasms did not respond differently from those who did not receive glucagon. This indicated that atrial arrhythmias and ST-segment changes might not be because of Ba-induced GI spasms.

Blakeborough et al. (1997) reported the results of a 1992-1994 survey, including questionnaires completed by U.K. consultant radiologists, of complications of Ba enema examinations in 750,000 patients. An overall mortality of about 1 in 57,000 was reported. A total of 82 complications (contraindications) were reported. Of 16 patients observed to have cardiac arrhythmias, 9 died. Seven of the nine patients who died were over the age of 75. Only a few of the other patients had a previous history of cardiac problems. BaSO₄, the insoluble salt of Ba, was used. There were no reported cases of such effects from Ba swallows during radiologic examinations. Thus, Ba-induced cardiac arrhythmias seem to occur only during enemas in extremely few cases, and patients over age 70 appear to be more susceptible if they have a prior documented history of cardiac disease. Because these patients had a health condition that warranted such a diagnostic procedure, application of these results to normal healthy populations could be limited.

In general, the toxicity of soluble Ba salts is far greater than that of insoluble forms (NRC 1977) when administered orally. Profound changes in cardiac, skeletal, and smooth-muscle functions resulting from

acute and accidental ingestion of soluble Ba salts have been reported. Several investigators (Roza and Berman 1971; Talwar and Sharma 1979; Wetherill et al. 1981) have reported cases of hypertension, paralysis of skeletal muscle, and cardiac arrests due to ingested soluble Ba salts. Infusion by iv of BaCl₂ to anesthetized dogs or guinea pigs resulted in increased blood pressure and cardiac arrhythmia (Roza and Berman 1971; Hicks et al. 1986). The study also reported skeletal muscle flaccidity and paralysis in dogs (Roza and Berman 1971). Although extremely rare, anomalies of cardiac rhythm during the Apollo and Skylab missions (Leguay and Seigneuric 1981) and episodes of ventricular tachycardia during long-duration missions (Fritsch-Yelle et al. 1998, see also readers’ notes in Ellestad 1999) have been reported. On the basis of those findings, there is concern that ingestion of Ba could potentiate such effects during short- and long-duration missions.

There have been numerous case reports of humans exposed to Ba through accidental or intentional oral ingestion. Death occurred in six cases of accidental or intentional ingestion of Ba salts. Two deaths were attributed to cardiac arrest (one from severe GI hemorrhage), and in three cases, the specific cause was not determined (Ogen et al. 1967; Das and Singh 1970; Talwar and Sharma 1979). Several cases of food poisoning from Ba carbonate were reported by Lewi and Bar-Khavim (1964) and Diengott et al. (1964). Symptoms of gastroenteritis, a feeling of numbness, diarrhea, vomiting, muscular twitching, and paralysis resulted from ingesting meat contaminated with BaCO₃. Of all cases, 5-10% were fatal within the first 48 h. In two patients, flaccid paralysis was found to be because of hypokalemia caused by low-serum potassium, and ECG recordings showed typical hypokalemic changes. Administration of potassium changed the clinical course or accelerated recovery from Ba poisoning, indicating that at least some of the effects are the result of hypokalemia.

The acute oral LD₅₀ (dose lethal to 50% of subjects) for Ba varies with species, compound, and age. For example, in the rat and guinea pig, the LD₅₀ for BaCl₂ is 118 and 76 mg/kg, respectively (Sax 1984). In humans, the lowest dose that caused death (LD_LO) for BaCl₂ is 11.4 mg/kg, while it is 70, 170, and 90 mg/kg in the mouse, rabbit, and dog, respectively (Sax 1984). Mortality has been observed in experimental animals following acute and chronic oral exposures to BaCl₂ and Ba acetate (Ba(CH₃COO)₂) (Schroeder and Mitchener 1975; Tardiff et al. 1980; Borzelleca et al. 1988). The acute oral LD₅₀ values for female and male rats were determined to be 269 and 277 mg/kg, respectively (Borzelleca et al. 1988), after the rats were gavaged with BaCl₂ in water at doses

ranging from 60 to 960 mg/kg (Ba at 40-640 mg/kg). Primary necropsy indicated hemorrhagic areas in the stomach and inflammation of the intestines. Tardiff et al. (1980) initially conducted an acute oral toxicity study of BaCl₂ in adult (60-70 d old) and weanling (21-25 d old) male and female rats to determine the LD₅₀ for each group. The acute oral LD₅₀ values were calculated to be 132 mg/kg in the adult rats and 220 mg/kg in the weanling rats.

Syed and Hosain (1972) reported LD₅₀ values for different Ba salts administered via iv to two strains of mice (ICR and Swiss-Webster). For the chloride, nitrate, and acetate, the LD₅₀ values were in the ranges of 8.12-11.32 mg/kg for the Swiss-Webster mice and 19.2-23.31 mg/kg for the ICR mice. The values for BaCl₂, Ba(NO₃)₂, and Ba(CH₃COO)₂ did not vary appreciably, although BaCl₂ was the most toxic. Because of the wide variation in the absorption of Ba salts, these values cannot be extrapolated to oral doses.

In a 1-d exposure study, Borzelleca et al. (1988) gavaged rats with BaCl₂ at 30, 100, and 300 mg/kg after a 24-h fast. A decrease in body weight, liver-to-brain weight ratios, and increases in kidney weight as a percentage of body weight were noted only at the highest dose. It appears that 100 mg/kg is a no-observed-adverse-effect level (NOAEL) for 1 d. Changes that occurred in the clinical chemistry were not dose related. At necropsy, male rats that received 300 mg/kg showed ocular discharge, fluid in the trachea, and darkened liver. In addition, inflammation of both the small and large intestines was seen in both the male and female rats at 300 mg/kg. On the basis of these findings, BaCl₂ at 100 mg/kg (equivalent to elemental Ba at 66 mg/kg) was identified as a NOAEL.

Infusion via iv of BaCl₂ to anesthetized dogs (Roza and Berman 1971) or guinea pigs resulted in increased blood pressure and cardiac arrhythmia (Hicks et al. 1986). Roza and Berman (1971) conducted a study on intact, anesthetized mongrel male and female dogs to elucidate the mechanisms of Ba-induced hypokalemia and hypertension and the interaction of Ba and potassium in the hearts of dogs in vivo. The interesting observations in this study were hypertension, a decrease in plasma potassium and an increase in red-cell potassium (shift in the potassium from extracellular to intracellular water) that caused hypokalemia, and myocardial toxicity resulting from hypokalemia. Increased mean corpuscular-red-cell volume (23% increase) leading to a substantial increase in hematocrit was also observed. In part of the study, seven dogs were iv infused with BaCl₂ at two rates of infusion (one dog at 1 micromole [µmol]/kg/min and six more dogs at 2 µmol/kg/min). ECG changes were tracked. The cessation of infusion was determined by the appearance of

an abnormal ECG. Although an increase in blood pressure was invariably seen during the first 5-10 min of infusion, it subsided 30-40 min after the infusion was finished. Although this was an iv-infusion study, one can attempt to extrapolate to an oral dose using an oral-absorption factor. There is no NOAEL, but the appearance of an abnormal ECG is considered a lowest-observed-adverse-effect level (LOAEL).

In a separate study by Borzelleca et al. (1988), male and female Sprague-Dawley rats (22-30 d old, 10 per group) were gavaged with BaCl₂ in deionized water at 100, 145, 209, or 300 mg/kg for 10 d (doses of Ba at 66, 96,138, or 198 mg/kg/d). Mortality of the female rats in the 198 mg/kg group increased, and one male rat from the 209 mg/kg group died. No consistent pathologic findings were noted. Decreased body weights and decreased ovary-to-brain weight ratios were noted in female rats. Decreased blood urea nitrogen (BUN) was observed in females in all treated groups but was observed in males only at the highest dose. Male rats showed a decrease in leukocytes at 209 mg/kg but not at the higher dose. The significant differences in BUN at all doses in female rats indicate that female rats are more sensitive to the short-term toxic effects of BaCl₂. A dose of Ba at 66 mg/kg appears to be a LOAEL for changes in BUN. The clinical significance of the extent of the change observed in this study is questionable.

In a human study conducted by Wones et al. (1990), 11 male volunteers (ranging in age from 27 to 61 y of age) with no history of hypertension, diabetes, or any cardiovascular disease participated in a 10-week (wk) protocol. The subjects consumed plain drinking water at 1.5 L/d for the first 2 wk. That was followed by 4 wk of consuming 1.5 L of water containing Ba as BaCl₂ at 5 ppm/d; and that was followed by 4 wk of consuming 1.5 L/d of water containing Ba as BaCl₂ at 10 ppm. The subjects were on a controlled basal diet (dietary contribution of Ba was assumed to be 0.75 mg/d). ECGs and 24-h continuous ECG monitoring were performed. The treatments did not result in any apparent changes in blood pressure, cholesterol, triglyceride, glucose, or potassium concentrations. According to the findings of the Wones et al. (1990) study, a

dose of 10 mg/L (equivalent to a dose of Ba at 0.21 mg/kg/d and water consumption at 1.5 L/d) can be considered a NOAEL for adverse effects of Ba in humans exposed for 4 wk.

Tardiff et al. (1980) exposed male and female Charles River rats to Ba (as BaCl₂) at 0, 10, 50, or 250 ppm in drinking water for 4, 8, or 13 wk. The estimated doses ranged from 1.7-45.7 mg/kg/d. For all dose levels, the intake of Ba decreased because of reduced water intake, and at termination of the study, the dosage rate was at half of the initial dose. No changes related to Ba ingestion were observed in clinical signs, hematologic parameters, or serum chemistry. The only change noted was a statistically significant decrease in water consumption in the highest-dose group. A slight decrease of adrenal weights in treated animals was noted. The increase in dose, not the increase in duration, produced increased concentrations of tissue Ba; the highest concentration was found in bone. Blood pressure was not measured in this study. A concentration of 50 ppm was identified as a NOAEL for change in water consumption.

The National Toxicology Program (NTP 1994) conducted a 13-wk study in which male and female F344/N rats were exposed to BaCl₂ dihydrate at 0, 125, 500, 1,000, 2,000, and 4,000 ppm (Ba at 0, 10, 65, 110, and 200 mg/kg/d for males and 0, 10, 35, 65, 115, and 180 mg/kg/d for females). In the highest-dose group, three males and one female died. The cause of death could not be determined. Significant decreases in water consumption were noted in the 4,000 ppm group. Relative and absolute organ-weight changes were also noted in the 2,000 and 4,000 ppm groups. Mild focal and multifocal areas of dilation were seen in the renal proximal tubules of the 4,000 ppm group. In addition to these effects, significant decreases in the magnitude of undifferentiated motor activity were observed in both sexes of rats receiving the 4,000 ppm dose, whereas the decreases were marginal at other doses. No other changes pertaining to neurobehavioral assessments were noted. Also, no changes were seen in cardiovascular measurements such as heart rate, systolic blood pressure, or ECG. The three notable adverse effects in the 13-wk NTP drinking water study are the effects on the renal proximal convoluted tubules (accompanied by significantly increased relative kidney weights), the effects on motor activity, and the significant decreases in water consumption. For renal effects, a LOAEL of 2,000 ppm (Ba at 180-200 mg/kg) and a NOAEL of 1,000 ppm (Ba at 110-115 mg/kg) were identified.

In another subchronic-to-chronic exposure study by Perry et al. (1989), female weanling Long Evans rats (45 g body weight) were pro-

vided drinking water containing Ba (salt not specified) at 1, 10, or 100 ppm. Water intake, systolic pressure, hematocrit, plasma catecholamine, and plasma concentrations of inorganic ions and others were measured at 1, 2, 4, 5, 12, and 16 mo. The average systolic pressure increased significantly after exposure at 100 ppm for 1 mo or longer. Thus, a dose of 10 ppm can be identified as a NOAEL using changes in systolic pressure as the toxicity end point. It is not known if weanling rats are more sensitive than adult rats to BaCl₂-induced hypertension. The study used a ryebased diet deficient in some required minerals, which limits the findings.

A health survey of workers at a Sherwin-Williams plant concluded that workers exposed by inhalation to Ba ores and BaCO₃ during grinding and mixing operations for at least 5 y had a significantly higher incidence of hypertension (7/12 or 58%) compared with workers who never worked in Ba processes (5/25 or 20%) (NIOSH 1982).

Data on chronic toxicity of oral ingestion of Ba⁺² come from human epidemiologic studies and animal studies. In a retrospective human epidemiologic study, Brenniman and Levy (1985) collected data from two communities in northern Illinois that had markedly different concentrations of Ba⁺² in their drinking water. In one community, the mean concentration of Ba⁺² in the drinking water was 0.1 mg/L (0-0.2 mg/L); in the other community, the mean concentration was 7.3 mg/L (2-10 mg/L). The study report had two parts: the first compared the mortality in these two communities (Brenniman et al. 1979), and the second (Brenniman et al. 1981) reported the results of blood pressure and health questionnaire data from a randomly selected pool of subjects. The pool included residents 18-75 y old who had lived in the community for more than 10 y. Data analysis was also performed on a subpopulation that did not have water softeners in their homes. The authors reported that the water concentration of other minerals, socioeconomic conditions, and demographic characteristics were comparable in the two communities.

In the first part, age- and gender-adjusted cardiovascular mortalities in the two communities were compared. Male and female mortalities from all “cardiovascular diseases” and “heart disease (arteriosclerosis)” in the high-exposure community were significantly higher than those in the low-exposure community. Generally, the group of people age 65 and older accounted for the largest clinical differences (Brenniman et al.

1979). Population mobility was greater in the high-exposure community, and confounding factors, such as the use of water softeners, were not accounted for.

In the second part of the study, three blood pressure measurements were taken within a span of 20 min, and responses to a questionnaire were collected from test subjects. The health questionnaire included questions about subjects’ medications and documented heart diseases, stroke and renal diseases, age, gender, smoking habits, and family history. No significant differences in mean systolic or diastolic blood pressures or in rates of hypertension, heart disease, stroke, or kidney disease were found between the men and women of the community with increased Ba concentrations and those of the community with low Ba concentrations. In a later analysis of the data, criteria were used for the number of years of residency, use of high blood pressure medications, and use of water softeners. No significant differences in mean systolic and diastolic pressure were found between the two communities in either males or females (Brenniman et al. 1981). When males and females 18-75 y of age who did not use water softeners or take high blood pressure medications and who had lived in the community more than 10 y were considered, data from only 85 males and 116 females from one community and 71 males and 93 females from the other community could be used. Study results were not affected when these criteria were included in the analysis. In addition, the prevalence of hypertension, stroke, heart disease, and kidney disease was not significantly different in males and females in these two communities. These results have been summarized by Brenniman and Levy (1985).

Perry et al. (1989) administered BaCl₂ to female weanling Long Evans rats in drinking water at 0, 1, 10, or 100 ppm (estimated doses of Ba at 0, 0.07, 0.7, or 7.1 mg/kg/d) for 1, 4, or 16 mo (the 1-mo study was described above). Ba doses in the 16-mo study were 0, 0.054, 0.54, or 5.4 mg/kg/d. Over the 16 mo of the experiment period, the average intake of water varied from 16 to 28 mL/d, and food intake ranged from 15 to 23 g/d per rat. The rats were fed a diet low in trace metals. Systolic blood pressure was measured at 1, 2, 4, 8, 12, and 16 mo. In the 4-mo study, significant increases in blood pressure were observed in rats in the highest-dose group; no change in blood pressure was observed in rats treated with either of the lower doses. Significant increases in blood pressure also were noted in the 16-mo study in rats treated with 10 ppm and 100 ppm doses (0.54 and 5.4 mg/kg/d); no changes were observed in the 1-ppm group. Thus, the systolic blood pressure progressively increased with time in the 100 ppm group; it was 17 millimeters of mercury (mm

Hg) higher than in the corresponding controls. Functional and biochemical studies of the heart after 16 mo of exposure to Ba at 100 ppm indicated significantly decreased contractile element-shortening velocity (decreased cardiac-contraction rates and depressed excitability). These studies identified a NOAEL of 1 ppm (0.17 mg/kg/d) for hypertension.

In another chronic exposure study (McCauley et al. 1985), groups of male and female rats received drinking water containing Ba (as BaCl₂) at 1, 10, 100, or 250 ppm—female rats for 36 wk and male rats for 46 and 68 wk. Histologic examinations of many tissues did not reveal any significant changes. It was reported that there were no significant changes in food or water consumption. In an extension of the study during which ECG recordings were made, Sprague-Dawley-derived CD rats were administered BaCl₂ in drinking water for 5 mo at 0 or 250 ppm (estimated intakes of 1 and 38.5 mg/kg/d due to the 1 mg/kg/d diet contribution). In this study, Ba induced a significant enhancement of 1-norepinephrine-induced bradycardia compared with controls 4 min after norepinephrine administration. The Ba-treated rats showed normal heart rates by 60 min, whereas controls had depressed heart rates.

In the chronic portion of the 1994 NTP study, male and female F344/N rats were exposed to BaCl₂ in drinking water for 2 y at 0, 500, 1,250, and 2,500 ppm (calculated doses of Ba at 0, 15, 30, and 60 mg/kg/d for males and 0, 15, 45, and 75 mg/kg/d for females). No significant increases in mortality occurred in the Ba-exposed groups. Only small reductions were noted in body weights in the 2,500 ppm group. Dose-related decreases in water consumption of about 23% were observed in male and female rats. The only chemical-related sign of kidney toxicity was an increase in the relative and absolute kidney weights of females at 2,500 ppm, an effect seen even at the 15-mo interim evaluation.

NTP (1994) also examined the subchronic and chronic toxicity of Ba in mice. In these studies, groups of male and female B6C3F₁ mice received Ba as BaCl₂ dihydrate in drinking water. They received 0, 125, 500, 1,000, 2,000, or 4,000 ppm for 13 wk or 0, 500, 1,250, or 2,500 ppm for 2 y. The animals were fed an NIH-07 diet; Ba content was not reported. Increased mortality was observed in the subchronic and chronic toxicity studies at the highest doses tested (4,000 and 2,500 ppm, respectively). Tubule dilatation, renal tubule atrophy, tubule cell regeneration, and the presence of crystals primarily in the lumen of the renal tubules were observed, indicating renal toxicity. The mice also exhibited elevated BUN, another indicator of renal toxicity. For renal effects, NOAELs were identified at 2,000 ppm for the subchronic exposure dura-

tion (205 and 200 mg/kg/d for male mice and female mice, respectively) and 1,250 ppm for the chronic exposure duration (75 and 90 mg/kg/d for male and female mice, respectively).

Schroeder and Mitchner (1975a, b) exposed groups of male and female Long Evans rats and Charles River CD mice to Ba (as Ba(CH₃COO)₂) at 0 or 5 ppm in drinking water. They reported proteinuria—significant compared with that in controls—in male rats exposed to Ba in water for 152 d. An increase in serum cholesterol in females and an alteration in serum glucose concentrations in males were also observed. No adverse alterations in life span, growth, or histopathology of the heart, lungs, kidneys, liver, or spleen were observed in either species. Thus, these studies identify a LOAEL of 5 ppm (Ba at 0.61 mg/kg/d) for renal glomerular damage, evidenced as proteinuria, in male rats maintained on low-mineral diets. A NOAEL of 5 ppm (Ba at 1.2 mg/kg/d) was identified for similarly exposed mice.

No data on in vivo genotoxic effects of Ba were available. Most in vitro studies in prokaryotic test systems revealed that Ba was not mutagenic. For example, BaCl₂ and Ba(NO₃)₂ were negative in the Ames assays with Salmonella typhimurium strains TA1535, TA1538, TA1537, TA97, TA98, and TA100 with or without metabolic activation (Monaco et al. 1991). BaCl₂ did not inhibit growth in wild and recdeficient Bacillus subtilis strains (Nishioka 1975). Negative results have also been observed for Ba(NO₃)₂ in the rec assay using B. subtilis strains H17 and H45 (Kanematsu et al. 1980). The NTP (1994) report describes a detailed battery of in vitro genetic toxicologic assays conducted on BaCl₂ (NTP 1994). BaCl₂ induced gene mutations in L5178Y mouse lymphoma cells with metabolic activation but not in the absence of metabolic activation (NTP 1994). Tests on the fidelity of DNA replication by avian myeloblastosis virus DNA polymerase did not indicate any effects of Ba(CH₃COO)₂ or BaCl₂ (Sirover and Loeb 1976). In mammalian cells, BaCl₂ did not induce sister chromatid exchanges or chromosomal aberrations in cultured Chinese hamster ovary cells, with or without activation (NTP 1994).

None of the reported epidemiologic studies indicated any Barelated cancer in humans. Under the U.S. Environmental Protection Agency’s (EPA’s) proposed Guidelines for Carcinogenic Risk Assessment (EPA 1996), Ba would be classified as a Group D (“not classifiable as a carcinogen to humans”). Rats and mice were exposed to Ba at 5 ppm (equivalent to 0.7 mg/kg/d for rats and 0.95 mg/kg/d for mice) as Ba(CH₃COO)₂ in drinking water for their lifetime (about 540 d) (Schroeder and Mitchener 1975a, b). Gross and microscopic examinations of heart, lungs, liver, kidneys, and spleen did not reveal differences in the incidence of tumors between Ba-acetate-treated animals and vehicle controls in either the Long-Evans study or in the Swiss-Webster mice study. Only one exposure dose was used in this study. In the Tardiff et al. (1980) study, no histopathologic abnormalities of the liver, kidneys, spleen, heart, brain, skeletal muscle, femur, or adrenals were found in rats exposed to Ba at up to 250 ppm in drinking water for 90 d. In the McCauley et al. (1985) BaCl₂ drinking water study, neoplasms were observed in several tissues, but they were not chemical related.

In a chronic exposure study conducted by the NTP (1994), male and female B6C3F₁ mice (60 animals of each gender per dose group) received BaCl₂ dihydrate in drinking water at concentrations of 0, 500, 1,250, or 2,500 ppm for 2 y. From water consumption and body weight data, the authors estimated the daily doses for the treated groups to be 30, 75, and 160 mg/kg/d for males and 40, 90, and 200 mg/kg/d for females. The animals continued in the study until they were moribund or died naturally, or they were sacrificed at the end of the study. Necropsy and complete histopathologic examinations were performed on all animals. The incidence of neoplasms in the Ba-exposed mice was not significantly higher than in control mice.

In the same chronic exposure study (NTP 1994), male and female F344/N rats (60 animals per gender per dose group) received drinking water containing BaCl₂ dihydrate at 0, 500, 1,250, or 2,500 ppm for 2 y. Using measured water consumption and body weights, the authors estimated daily Ba doses for the treated groups at 15, 30, and 60 mg/kg/d for males and 15, 45, and 75 mg/kg/d for females. No chemical-related, noncarcinogenic histologic changes were observed in any organs or tissues. No statistically significant increases in the incidence of neoplasms were observed in the Ba-exposed rats. Although benign and malignant pheochromocytoma of the adrenal medulla (combined) and mononuclear cell leukemia were noted in male rats, mammary gland neoplasms (fibroade-

noma, adenoma, or carcinoma) were observed in female rats, these changes appeared to decrease with increasing dose. These studies strongly indicate that Ba (as Ba salts) administered orally is noncarcinogenic to animals (see Table 3-2 for a summary of the studies).

No data were located regarding reproductive effects of orally ingested Ba in humans. However, some data from animal studies have been reported. Borzelleca et al. (1988) reported that in rats gavaged with BaCl₂ at 98 mg/kg/d, no changes in testicular weight and no gross lesions of the ovaries or testes were observed; however, at that dose, decreased ovary weight and decreased ovary-to-brain-weight ratio were noted. Intermediate and chronic oral exposure of rats to nominal concentrations of Ba (37.5 or 15 mg/kg/d, respectively) in drinking water was not associated with any gross or histopathologic lesions of the uterus, ovaries, or testes (McCauley et al. 1985). There are no reports on the effects of Ba (orally ingested) on reproductive function, even though inhalation of BaCO₃ dust has been reported to induce disturbances in spermatogenesis in male animals and shortened estrous cycle and morphologic changes of ovaries in females (Tarasenko et al. 1977).

Little information is available on the developmental effects of Ba after oral exposure. Morton et al. (1976) reported that when Ba concentrations in drinking water increased, the rate of congenital central nervous system malformations decreased. Increased mortality, increased leukocyte count, disturbances in liver function, and increased urinary excretion of hippuric acid were observed in offspring of female rats administered BaCO₃ at 18.3 mg/kg/d orally during pregnancy (Tarasenko et al. 1977).

Acceptable concentration (AC) values were determined following the guidelines of the NRC (2000). For each exposure duration, the

spacecraft water exposure guideline (SWEG) value (see Table 3-3) was set on the basis of the lowest value among the ACs for all the significant adverse effects at that exposure duration. ACs were calculated assuming a nominal potable water use of 2.8 L/d (including 0.8 L of water used for the reconstitution of food), in contrast to EPA’s reference volume of 2 L/d of water. A value of 70 kg was used as the nominal adult body weight. Changes in body weights and organ weights alone were not considered biologically relevant adverse effects.

Spaceflight is known to reduce the volume of blood and the number of peripheral blood cells; therefore, spaceflight might have an additive effect to those of chemicals that produce adverse hematologic effects. Due to shifts in body fluids, astronauts tend to feel less thirsty and may become dehydrated; therefore, chemicals that affect water consumption could be very detrimental to spaceflight crew. Hence, a factor of 3 was applied during the AC calculations to reduce the risk of these effects. Astronauts are physically challenged, resulting in potentially increased sensitization to cardiac arrhythmia. The National Aeronautics and Space Association (NASA) uses a factor of 5 to modify the ACs of chemicals that can increase sensitization to cardiac effects. In the case of Ba, such a factor has been applied when the end point is cardiac toxicity (see 1-d AC and 10-d AC).

Inorganic chemical analysis of humidity condensate samples and the recycled-processed water samples from several missions in Mir indicated that in the humidity condensates, the ionic Ba concentrations occurred at a maximum of 113.92 mg/L. The average concentration in all humidity condensate samples was 0.64 mg/L. Analysis of Ba in the recycled-processed water from humidity condensates indicated the presence

of Ba⁺² at a maximum of 2.44 mg/L and an average of 0.19 mg/L. It was found in 18 of 22 samples analyzed. The maximum concentration exceeded the NASA-Russian interim water quality specifications of 1 mg/L and the EPA maximum contaminant level (MCL) of 2.0 mg/L. (See Table 3-4 for summary of guidelines set by government organizations.) Hence, it was decided to derive SWEGs for different durations of exposure though drinking water (Pierre et al. 1999). The NRC (1982) calculated a recommended concentration of 4.7 mg/L. It was originally derived from the American Conference of Governmental Industrial Hygienists (ACGIH) 8-h time-weighted average (TWA) using an absorption factor of 20% and a safety factor of 2.

The Agency for Toxic Substances and Disease Registry (ATSDR) (1992) did not derive minimal risk levels (MRLs) for Ba ingested orally for acute, intermediate, or chronic durations, because human case studies did not provide adequate dose characterization related to adverse effects, and the animal studies did not provide sufficient data to identify a target organ. Even though a LOAEL and a NOAEL can be derived from the blood pressure studies of Perry et al. (1985, 1989), the resulting MRL would have been 20 times lower than the WHO estimated intake of Ba from dietary and other sources.

EPA derived an oral reference dose (RfD) of 0.07 mg/kg/d on the basis of a NOAEL of 7.3 mg/L obtained from the epidemiologic study of Ba exposure from Illinois drinking water supplies (Brenniman et al. 1984). An uncertainty factor (UF) of 3 for intra-individual sensitivities was used, and the human reference body weight of 70 kg was assumed, as well as a water intake rate of 2 L/d. The values were substantiated by the Wones et al. (1990) experimental study in humans. The study did not report any changes, including in ECGs, and thus, may serve as a good lower-bound risk estimate for adverse effects on the cardiovascular system.

Dallas and Williams (2001) reviewed the principal and supporting studies in which hypertension was identified as the critical effect or end point that were used by EPA in 1998 for the oral RfD (McCauley and Washington 1983; Brenniman and Levy 1985; McCauley et al. 1985; Wones et al. 1990; NTP 1994). EPA also considered increases in kidney weights reported in the NTP (1994) subchronic and chronic exposure rat and mice studies to be an adverse renal effect. According to Dallas and Williams, the NTP (1994) study should be used as the principal study, and renal effects should be considered the most appropriate end point for deriving an RfD. Hence, these authors assert that EPA should revise the

RfD for Ba on the basis of renal effects, such as the renal tubular dilatation, multifocal to diffuse nephropathy, and kidney lesions observed in the NTP (1994) study. This is also supported by the McCauley et al. (1985) study in which signs of inefficient glomerular filtration and hypertension were reported. Dallas and Williams (2001) identified NOAELs for renal effects from the NTP studies (60 mg/kg/d for male rats, 75 mg/kg/d for female rats, 75 mg/kg/d for male mice, and 90 mg/kg/d for female mice) and, using the most sensitive NOAEL (male rats appeared to be more sensitive), calculated an RfD. Modifying factors proposed by Doursen (1994) were used in the derivation of the proposed RfD:

where

60 mg/kg/d = NOAEL;

10 = human variability (intraspecies variability);²

3 = interspecies extrapolation factor;

1 = subchronic to chronic factor;

3 = insufficiency in the database (uncertainty whether this dose will protect against effects such as developmental or reproductive toxicity);

1 = LOAEL to NOAEL (because NOAEL is already known) extrapolation factor; and

1 = outstanding uncertainties adequately addressed.

No human acute study data were available to derive a 1-d AC. The available data are from accidental oral poisoning with Ba salts at uncertain doses yielding very serious adverse effects. For example, the case report from Talwar and Sharma (1979) describes the cardiotoxic effects

of BaCl₂ following an accidental ingestion of rat poison. Although the changes were suggestive of acute myocardial injury, there were no data on the dose. There are a number of such reports.

Using mongrel dogs, Roza and Berman (1971) studied the hypokalemic and cardiovascular effects, such as hypertension, resulting from iv infused BaCl₂. In these experiments, an increase in blood pressure invariably was observed, but that toxicity end point cannot be used for humans, because the association between blood pressure and Ba intake has not been well established. The iv infusion of BaCl₂ was done for various periods of time at various infusion concentrations until the dogs showed evidence of abnormal ECGs. In addition, the investigators observed generalized muscle twitching and skeletal muscle contractions. In another part of the investigation, the authors observed changes in the mean corpuscular volume (23% increase) and a decrease in plasma potassium with a concomitant increase in red-cell potassium. A 1-d AC was calculated on the basis of cardiotoxicity.

The dogs were infused with 1 or 2 µmol/kg/min. The cessation of infusion was determined by the appearance of abnormal ECG. The total dose was calculated on the basis of that time and the dose of Ba was converted to µg/kg/min. Multiplying the dose and time yields the cumulative dose. That was set as the LOAEL. It was calculated for each dog, and a mean of the doses was taken. One value was rejected as an outlier. An absorption factor to reflect differences between the iv and oral routes was used, and a value of 10% was used in this case. The iv LOAEL was multiplied by a factor of 10 to get the oral-equivalent LOAEL dose. Even though the infusion was carried out for a duration shorter than 100 min, the dose is considered to be a 1-d dose and was not extrapolated for time. To obtain a NOAEL, a factor of only 2 was applied because this NOAEL can be considered sufficiently lower than the lower-bound estimate of the LOAEL. The detailed 1-d AC calculations are as follows, as recommended by the committee (the oral-equivalent LOAEL in this study for abnormal ECG was 84 mg/kg/d):

where

2 = LOAEL-to-NOAEL extrapolation factor;

10 = species extrapolation factor; and

5 = spaceflight factor and for cardiotoxicity and other effects on muscle contraction.

The following factors were considered when using iv data to derive exposure concentrations for the oral route. Split oral doses over the period of a day would not result in a consistent blood level (steady-state concentration) that would be sustained over as long a period of time as an infusion and cause such abnormal responses in cardiograms. The built-in safety factor is that the total amount of Ba will not be consumed at one time but will be ingested in several boluses over active crew time (about 16 h).

Two 1-d ingestion studies on Ba salts were initially designed to derive an LD₅₀ in animals (Tardiff et al. 1980; Borzelleca et al. 1988). Borzelleca et al. (1988) also studied the adverse effects of ingestion of Ba salts at doses below the LD₅₀. Rats were gavaged with BaCl₂ at 30, 100, or 300 mg/kg (equivalent to Ba at 20, 66, and 198 mg/kg) in water. Twenty-four hours later, decreases in body weight and liver-to-brain-weight ratios, increases in kidney weight as a percentage of body weight (at the highest dose), and observed changes in hematocrit and clinical chemistry were not dose related. At necropsy, male rats receiving BaCl₂ at 300 mg/kg showed ocular discharge, fluid in the trachea, and darkened liver. In addition, inflammation of the small and large intestines was observed in both male and female rats at 300 mg/kg. The rats were fasted for 24 h prior to the gavage dose. There was a very high mortality at this dose, but there were no mortalities at 100 mg/kg, which is only three times lower. The mortality dose-response curve appears to be steep. In the subsequent 10-d study, when the same high dose of 300 mg/kg was administered by gavage to nonfasted rats, only 3 of 10 females died even after 10 doses. Subsequently, a dose of BaCl₂ at 100 mg/kg (equivalent to Ba at 66 mg/kg) can be identified as a NOAEL for these effects on the GI tract and the lungs. From these data for GI inflammatory effects, a 1-d AC can be calculated as follows:

where

70 kg = nominal body weight;

10 = species extrapolation factor; and

2.8 L/d = nominal water consumption.

Wones et al. (1990) administered BaCl₂ in drinking water to 11 human male volunteers in a 10-wk protocol. In the first 2 wk, no Ba was

administered. In the next 4 wk, the subjects were exposed to Ba at 5 ppm (0.11 mg/kg/d). That was followed by 4 wk of exposure to Ba at 10 ppm (0.21 mg/kg/d). This study was evaluated for deriving a 10-d AC. No significant changes in systolic blood pressure or in any variables related to hypertension (plasma total cholesterol, triglycerides, low-density lipoprotein [LDL] or high-density lipoprotein [HDL], serum albumin, and potassium concentrations) were noted. This study indicated that Ba at 0.21 mg/kg/d did not induce any abnormal ECG. If an AC were calculated on the basis of that value, it would be overly conservative. These data were from a human study, but it is interesting to note that in the 2-y NTP (1994) BaCl₂ subchronic and chronic exposure study, there was no evidence of changes in cardiovascular measurements, which included systolic blood pressure, heart rate, and ECGs from each of the rats at the end of 15 mo of exposure to BaCl₂ through drinking water.

In an NTP (1994) study, male and female F344/N rats were administered concentrations of BaCl₂ dihydrate in drinking water at 0, 125, 250, 500, 1,000, or 2,000 ppm for 15 d (estimated delivered doses of Ba at 10, 15, 35, 60, or 110 mg/kg). Water consumption by male and female rats that received Ba at 110 mg/kg was about 16% lower than that of the controls during week 2. A NOAEL for Ba of 60 mg/kg/d was identified.

A 10-day AC for Ba for decreased water consumption can be calculated as follows:

where

60 mg/kg/d = NOAEL;

70 kg = nominal body weight;

10 = species extrapolation factor;

2.8 L/d = nominal water consumption; and

3 = spaceflight factor for dehydration.

The 10-d short-term toxicity study by Borzelleca et al. (1988) was also considered for deriving an AC for 10 d. In this protocol, male and female Sprague-Dawley-derived CD rats were gavaged a water solution of BaCl₂ at doses of 100, 145, 209, and 300 mg/kg (equivalent to Ba at doses of 66, 96, 138, and 198 mg/kg/d). Death of female rats occurred at the 138 and 198 mg/kg doses of Ba. Although no compound-related body-weight or relative-organ-weight changes were noted, decreases in BUN, a variable usually of clinical importance in the diagnosis of renal toxicity, decreased with all four doses in the female rats and at 198 mg/kg doses in the male rats. The changes, although statistically signifi-

cant, were small and may not be of clinical significance. In addition, the decrease did not appear to be dose related. Hence, the data were not used for AC derivation. In the NTP (1994) BaCl₂ drinking water study, BUN was not altered in male or female rats.

The study by Perry and associates (Kopp et al. 1985; Perry et al. 1989) was also evaluated for deriving a 10-d AC. Weanling female Long-Evans hooded rats were provided Ba as BaCl₂ in drinking water at 1, 10, and 100 ppm (Ba at 0.071, 0.71, and 7.1 mg/kg/d) and studied for 1-16 mo. Indirect measurement of systolic pressure of unanesthetized rats indicated that after exposure to Ba at 100 ppm (7.1 mg/kg/d), the average systolic pressure increased significantly (12 mm Hg higher than the mean of controls) for 1 mo or longer. From this study, Ba at 100 ppm (7.1 mg/kg/d) appears to be a LOAEL for blood pressure changes. The study was not used for AC calculations because the rats received a rye-based diet, which is low in trace metal content compared with standard lab chow, which includes calcium and potassium. It was noted that the calcium content of the diet is below minimum requirements, and because calcium and Ba act as agonists, the animals may have been more sensitive to Ba administration. Hence, the data were not used because of the two key confounding factors of calcium and potassium content in the diet.

A 10-d AC was also obtained from the 1-d acute-cardiotoxicity study carried out in dogs by Roza and Berman (1971) described in the 1-d AC section. Once Ba is absorbed into the systemic circulation, it is rapidly removed from blood and deposited in the bone. Because cardiotoxicity and effects on skeletal muscle activity depend on circulating Ba concentrations, it is doubtful whether the blood concentrations of Ba will accumulate if ingested daily at this dose (21 mg/L). Furthermore, in the NTP (1994) studies in rats that received Ba at up to 200 mg/kg/d (as BaCl₂·2H₂O) in drinking water for 13 wk, there were no changes in heart rate or blood pressure, nor did that result in abnormal ECGs. So, the 1-d AC was extended for use also as a 10-d AC, all safety factors still being applicable:

To derive a 100-d AC, the NTP 13-wk subchronic drinking water exposure study (NTP 1994), and studies by Tardiff et al. (1980) and Perry et al. (1989) were considered.

In the Tardiff study, groups of young adult rats (4 wk old) of both sexes were provided BaCl₂ in drinking water for 4, 8, and 13 wk at concentrations of 0, 10, 50, and 250 mg/L. (The average Ba dose range was 1.7-45.7 mg/kg/d, based on the cumulative 13 wk dose given by the authors.) There were no compound-related adverse effects on clinical signs, hematologic measurements, serum enzyme activities, serum ions, gross pathology, or histopathology. The only significant observation was the reduction in water consumption in the high-dose group in male and female rats. Since dehydration is an important factor for spaceflight crews, a 100-d AC was derived for this parameter. A LOAEL for Ba of 45.7 mg/kg/d and a NOAEL of 9.7 mg/kg/d were identified.

Thus, a 100-d AC for Ba for reduction in water consumption can be calculated as follows:

where

9.7 mg/kg /d = NOAEL;

70 kg = nominal body weight;

10 = species extrapolation factor;

2.8 L/d = nominal water consumption;

3 = spaceflight factor for dehydration; and

100 d/91 d = time extrapolation factor.

A benchmark dose (BMD) calculation could not be carried out, because standard deviations for individual doses and corresponding water consumption data were not available.

The Perry et al. (1989) study could not be used because of reasons mentioned under the 10-d AC derivation section: the rats were on a low-mineral diet, which may have increased the sensitivity of rats to Bainduced hypertension.

In an NTP (1994) 13-wk subchronic exposure study, groups of male and female F344/N rats were exposed to BaCl₂ dihydrate in drinking water at 0, 125, 500, 1,000, 2,000, or 4,000 ppm (0, 10, 65, 110, and 200 mg/kg/d for males and 0, 10, 35, 65, 115, and 180 mg/kg/d for females). In at least 30% of both male and female rats exposed to the highest dose, minimal-to-mild focal and multifocal areas of dilation of the renal proximal convoluted tubules were observed. These renal histopathologic lesions were not severe. Similar effects were observed in the glomerulus by McCauley et al. (1985) in rats given Ba⁺² at 1,000 ppm in

their drinking water; thus, the NTP observation is considered consistent and significant. Because this effect was also accompanied by an increase in absolute and relative kidney weights at the 2,000 ppm dose, a LOAEL of 2,000 ppm and a NOAEL of 1,000 ppm (Ba at 65 mg/kg/d for both sexes) for renal effects were identified. Another important observation in this NTP study (1994) is that there was a substantial reduction in the calcium concentrations in the upper levels of femoral bone in male and female rats receiving BaCl₂ in drinking water at 65 mg/k/d measured at 15 mo after the start of the dose. Ba is a bone-seeking element and is known to replace calcium from the bone. Bone demineralization and kidney stone formation are important concerns for short- and long-duration spaceflights; this renal effect may be exacerbated by Ba exposure. Hence, a factor of 3 for space-related effects will also be applied.

A 100-d AC for renal effects can thus be calculated as follows:

where

65 mg/kg/d = NOAEL;

10 = species extrapolation factor;

70 kg = nominal body weight;

2.8 L/d = nominal water consumption;

100 d/91 d = time factor for extrapolation; and

3 = spaceflight factor to further protect against renal effects and bone demineralization.

In the 13-wk subchronic toxicity NTP study (1994), the other toxicologic end point observed was the neurobehavioral decrement measured as the magnitude of undifferentiated motor activity in both sexes of rats and mice receiving BaCl₂ dihydrate administered in drinking water at 4,000 ppm for 13 wk (Dietz et al. 1992; NTP 1994). In the female mice, the forelimb grip strength was also significantly lower than that of controls at 13 wk. In the above NTP subchronic exposure rat study, a LOAEL for a decrease in motor activity was identified at a concentration of 4,000 ppm (Ba at 200 mg/kg/d). A dose of 110 mg/kg/d was identified as a NOAEL.

Thus, the 100-d AC for neurobehavioral effects can be derived as follows:

where

110 mg/kg/d = NOAEL;

10 = species extrapolation factor;

70 kg = nominal body weight;

2.8 L/d = nominal water consumption; and

100 d/91 d = time extrapolation factor.

A BMD calculation was carried out for the above data. A BMDL₀₁ was determined for male and female rats using the data in Tables 3-5 and 3-6: Ba at 41 mg/kg/d for male rats and 32.5 mg/kg/d for female rats, where BMDL₀₁ is the 95% lower-confidence limit of a BMD corresponding to a 1% effect. Female rats appear to be more sensitive than males.

A 100-d AC for neurobehavioral effects can be calculated as follows:

where

32.5 mg/kg/d = BMDL₀₁;

70 kg = nominal body weight;

10 = species extrapolation factor;

2.8 L/d = nominal water consumption; and

100 d/91 d = time extrapolation factor.

Four chronic toxicity studies were considered to derive a 1,000-d AC for Ba in drinking water. The human retrospective epidemiologic study conducted by Brenniman and coworkers in 1979, 1981, and 1985 and was considered for the 1,000-d AC derivation. In short, the data consisted of results from a study on two communities with widely differing concentrations of Ba in the drinking water (mean concentrations of Ba at 0.1 mg/L in one community and 7.3 mg/L in the other community). A high correlation between age-adjusted mortality from cardiovascular diseases and concentrations of Ba was noted.

The analysis did not take into consideration a few confounding factors applicable to the high-Ba community, such as change in the population dynamics (an increase of 70% in the population of this community), use of water softeners (because the water source for this community was hard), and use of blood pressure medications. In addition, the noted increase in the mortality was higher in subjects who lived in the community 10 y or less, meaning that the increase was concentrated in those new populations who migrated into this community. No prior medical history of these individuals before they migrated into this region is known.

In a follow-up study, no significant differences were observed in the mean systolic and diastolic blood pressures of men or women between the two communities. When the above-mentioned additional criteria were incorporated into the data analysis, the prevalence rates for hypertension, stroke, heart disease, and kidney disease in males and females were not significantly different between these two communities.

Hypertension as the toxicity end point and mortality from cardiovascular diseases in the Brenniman and Levy (1985) study were not used for deriving the 1,000-d AC for the following reasons: (1) blood pressure measurements were taken only once during the data-collection session, and (2) the mortality was not robust enough to be used to set an AC.

EPA used that study to derive an oral RfD for soluble Ba salts. The value of 7.3 mg/L (the average Ba concentration in the high-Ba community) was used to derive an adjusted NOAEL by using a UF of 3 because the study had a large number of test subjects of both sexes in a broad age range.

Thus, the EPA oral RfD of 0.07 mg/kg/d was calculated as follows:

where 2 L/d is the nominal water consumption, 70 kg is the nominal body weight, and 3 is the factor to reduce the NOAEL.

Dallas and Williams (2001) considered the Brenniman and Levy (1985) study on two Illinois communities with 70-fold differences in Ba concentrations in water in which it was concluded that there were no significant differences in blood pressure, heart diseases, or kidney diseases associated with Ba and from which EPA chose a NOAEL without a LOAEL. According to Dallas and Williams (2001), cardiovascular effects occur at doses well above those reported for renal effects, and therefore, renal effects should be the end points used for oral RfD. Using the data from the NTP (1994) BaCl₂ drinking water study discussed in the earlier sections, these authors proposed an RfD of 0.6 mg/kg/d using a NOAEL for Ba of 60 mg/kg/d and a modifying factor of 90 (as described earlier in this section).

Three animal studies were considered for AC calculations. In the Schroeder and Mitchener (1975a, b) studies, rats and mice exposed for life to Ba as Ba acetate at 5 ppm in drinking water showed protein-urea after 5 mo, indicating renal toxicity. Unfortunately, only one Ba concentration was used, and proteinuria was tested only semi-quantitatively. Therefore, the observation was not used for AC calculations, despite data from the NTP 2-y study and the Dietz et al. (1992) study on mice using BaCl₂ in drinking water, which demonstrated renal toxicity. In the second study, Perry et al (1989) reported Ba-induced hypertension when weanling rats were orally exposed to BaCl₂ at 0, 1, 10, and 100 ppm in water for 1, 4, and 16 mo (Ba at 0, 0.071, 0.71, or 7.1 mg/kg/d). Increased systolic pressure was noted in rats exposed to 7.1 mg/kg/d for 16

mo. The study was not used because of reasons outlined in the 10-d AC section.

In the third study, a chronic toxicity study by NTP (1994), both male and female rats were exposed to BaCl₂ in their drinking water for 2 y at concentrations of 0, 500, 1,250, and 2,500 ppm (Ba at 0, 15, 30, and 60 mg/kg/d for males and 0, 15, 45, and 75 mg/kg/d for females). Three toxicologically significant observations were noted in this report. Water consumption decreased in a dose-dependent manner; an approximate 23% reduction compared with control values occurred at 2,500 ppm. Evidence of renal injury was noted in female mice: BUN increased, and the kidneys had abnormal pigmentation. Renal nephropathy and crystal formation in the renal tubules were observed in mice that received BaCl₂ at 2,500 ppm in the diet for 2 y. The chemical-related renal toxicity was not seen in rats. These effects were used to derive 1,000-d ACs as follows.

A dose-related decrease in water consumption was seen as early as 5 wk in the 60 mg/kg/d group. First, a 1,000-d AC was derived based on reduction of water consumption in rats (not in mice). Although at the highest dose (60 mg/kg/d) the decreased water consumption was pronounced, water consumption was also found to be lower in the 30 mg/kg/d group. A NOAEL was identified as 15 mg/kg/d for Ba. A 1,000-d AC for a decrease in water consumption is calculated as follows:

where

15 mg/kg/d = LOAEL;

70 kg = nominal body weight;

10 = species extrapolation factor;

2.8 L/d = nominal water consumption;

1,000 d/730 d = time factor; and

3 = spaceflight factor for dehydration.

The 1,000-d AC for reduction in water consumption is 9 mg/L based on the NTP study (1994).

A 1,000-d AC was also calculated using data on the renal toxicity end points (nephropathy and renal crystal formation) from male and female mice after 2 y of exposure to BaCl₂ in their drinking water). These data have been summarized in Table 3-7. A summary of the NOAEL and BMDL₀₁ values are shown in Table 3-8.

Using both the NOAEL and the BMDL₀₁ values listed in Table 3-8, ACs were calculated for both renal toxicity end points as follows:

where

70 kg = nominal body weight;

10 = species extrapolation factor;

2.8 L/d = nominal water consumption;

1,000 d/730 d = time extrapolation factor; and

3 = spaceflight safety factor for renal effects risk and bone demineralization risks.

Male mice seem to be more susceptible to renal crystal formation as well as nephropathy than the female mice, and the values based on the BMD approach are conservatively lower than the NOAEL method (see Table 3-9). The most conservative of AC values for renal crystals (24 mg/L) and nephropathy (19 mg/L) in male mice, derived using the BMD method, were chosen as the 1,000-d AC. These values are entered in the AC summary table (Table 3-10).

ACGIH (American Conference of Governmental Industrial Hygienists). 1986. Documentation of the Threshold Limit Values and Biological Exposure Indices, 5th Ed. Cincinnati, OH: ACGIH.

ATSDR (Agency for Toxic Substances and Disease Registry). 1992. Toxicological Profile for Barium and Compounds. Agency for Toxic Substances and Disease Registry, U.S. Public Health and Service, Washington, DC.

Bauer, G.C.H., A. Carlsson, and B. Lindquist. 1956. A comparative study on the metabolism of ¹⁴⁰Ba and ⁴⁵Ca in rats. Biochem. J. 63:535-542.

Bauer, G.C.H., A. Carlsson, and B. Lindquist. 1957. Metabolism of ¹⁴⁰Ba in man. Acta. Orthop. Scand. 26:241-254.

Berman, C.Z., M.G. Jacobs, and A. Bernstein. 1965. Hazards of the barium enema examination as studied by electrocardiographic telemetry. Preliminary report. J. Am. Geriat. Soc. 13: 672-686.

Blakeborough, A., M.B. Seridan, and A.H. Chapman. 1977. Complications of barium enema examinations: a survey of UK Consultant Radiologists 1992 to 1994. Clin. Radiol. 52(2):142-148.

Bligh, P.H. 1960. Metabolism of barium in rat and man. PhD Thesis, London University. London.

Bligh, P.H., and D.M. Taylor. 1963. Comparative studies of the metabolism of strontium and barium in the rat. Biochem. J. 87:612-618.

Borzelleca, J.F., L.W. Condie, Jr., and J.L Egle, Jr. 1988. Short-term toxicity (one- and ten-day gavage) of barium chloride in male and female rats. J. Am. Coll. Toxicol. 7:675-685.

Brenniman, G.R., and P.S. Levy. 1985. Epidemiological study of barium in Illinois drinking water supplies. Pp. 231-240 in Inorganics in Water and Cardiovascular Disease, E.J. Calabrese, R.W. Tuthill, and L. Condie, eds. Princeton, NJ: Princeton Scientific Publishing.

Brenniman, G.R., T. Namekata, W.H. Kojola, B.W. Carnow, and P.S. Levy. 1979. Cardiovascular disease death rates in communities with elevated concentrations of barium in drinking water. Environ. Res. 20:318-324.

Brenniman, G.R., W.H. Kojala, P.S. Levy, B.W. Carnow, and T. Namekata. 1981. High barium levels in public drinking water and its association with elevated blood pressure. Arch. Environ. Health 36(1):28-32.

Calabrese, E.J. 1977. Excessive barium and radium-226 in Illinois drinking water. J. Environ. Health 39:366-369.

Chou, C., and Y.C. Chin. 1943. The absorption, fate, and concentration in serum of barium in acute experimental poisoning. Chin. Med. J. 61:313-322.

Clary, J.J., and R.G. Tardiff. 1974. The absorption distribution and excretion of orally administered ¹³³BaC1₂ in weanling male rats [abstract]. Toxicol. Appl. Pharmacol. 29:139.

Cuddihy, R.G., and W.C. Griffith. 1972. A biological model describing tissue distribution and whole-body retention of barium and lanthanum in beagle dogs after inhalation and gavage. Health Phys. 23:621-633.

Dallas, C.E., and P.L. Williams. 2001. Barium: rationale for a new oral reference dose. J. Toxicol. Environ. Health B Crit. Rev. 4(4):395-429.

Das, N.C., and V. Singh. 1970. Unusual type of cardiac arrest: Case report. Armed Forces Med. J. India 26:344-352.

Dencker, L. A. Nilsson, C. Ronnback, and G. Walinder. 1976. Uptake and retention of ¹³³Ba and ¹⁴⁰Ba^140La in mouse tissues. Acta Radiol. Ther. Phys. Biol. 15(4):273-287.

Diengott, D., O Rozsa, N. Levy, S. Muammar. 1964. Hypokalemia in barium poisoning. Lancet 14:343-344.

Dietz, D.D., M.R. Elwell, W.E. Davis, and E.F. Meirhenry. 1992. Subchronic toxicity of barium chloride dihydrate administered to rats and mice in the drinking water. Fund. Appl. Toxicol. 19:527-537.

Douglass, W.W., and R.P. Rubin. 1964. Stimulant action of barium on the adrenal medulla. Nature 203:305-307.

Dourson, M.L. 1994. Methods for establishing oral reference doses (RfDs). Pp. 51-61 in Risk Assessment of Essential Elements, W. Mertz, C.O. Abernathy, and S.S. Olin, eds. Washington, DC: ILSI Press.

Durfor, C.N., and E. Becker. 1964. Public water supplies of the 100 largest cities in the United States, 1962. Water-Supply Paper 1812. United States Geological Survey, Washington, DC.

Durfor, C.N., and E. Becker. 1968. Public water supplies of the 100 largest cities in the United States. Water-Supply Paper 1812. United States Geological Survey, Washington, DC.

Eastwood, G.L. 1972. ECG abnormalities associated with the barium enema. JAMA 219(6):719-721.

Edel, J., A. Di Nucci, E. Sabbioni, L. Manzo, M. Tonini, C. Minnoia, and S. Candedoli. 1981. Biliary excretion of barium in the rat. Biol. Trace Elem. Res. 30(3):267-276.

Ellestad, M.H. 1999. Ventricular tachycardia during spaceflight. Am. J. Cardiol. 83(8):1300.

EPA (U.S. Environmental Protection Agency). 1983. Barium Occurrence in Drinking Water, Food, and Air. Office of Drinking Water, U.S. Environmental Protection Agency, Washington, DC.

EPA (U.S. Environmental Protection Agency). 1985. Draft Health Effects Criteria Document for Barium. Office of Drinking Water, U.S. Environmental Protection Agency, Washington, DC.

EPA (U.S. Environmental Protection Agency). 1988. Integrated Risk Information System. Barium and Compounds. Minor revisions were made to Oral RfD 1-21-1999. U.S. Environmental Protection Agency, Washington, DC.

EPA (U.S. Environmental Protection Agency). 1989. National primary and secondary drinking water regulations. Fed. Regist. 54:22062.

EPA (U.S. Environmental Protection Agency). 2002. 2002 Edition of the Drinking Water Standards and Health Advisories. U.S. Environmental Protection Agency, Washington, DC.

Erre, N., F. Manca, and A. Parodo. 1980. The short-term retention of barium in man. Health Phys. 38:225-227.

Fritsch-Yelle, J.M., U.A. Leuenberger, D.S. D'Aunno, A.C. Rossum, T.E. Brown, M.L. Wood, M.E. Josephson, and A.L. Goldberger. 1998. An episode of ventricular tachycardia during long-duration spaceflight. Am. J. Cardiol. 81(11):1391-1392)

Gould, D.B., M.R. Sorrell, and A.D. Lupariello. 1973. Barium sulfide poisoning. Some factors contributing to survival. Arch. Intern. Med. 132:891-894.

Hardcastle, J., P.T. Hardcastle, and J.M. Noble. 1985. The secretory action of barium chloride in rat colon. J. Physiol. 361:19-33.

Harrison, G.E., T.E.F. Carr, A. Sutton, and J. Rundo. 1966. Plasma concentration and excretion of calcium-47, strontium-85, barium-133 and radium223 following successive intravenous doses to a healthy man. Nature 209(22):526-527.

Harrison, G.E., T.E.F. Carr, and A. Sutton. 1967. Distribution of radioactive calcium, strontium, barium and radium following intravenous injection into a healthy man. Int. J. Radiat. Biol. 13:235-247.

Hicks, R., L.Q. Caldas, P.R. Dare, P.J. Hewitt. 1986. Cardiotoxic and bronchoconstrictor effects of industrial metal fumes containing barium. Arch. Toxicol. Suppl. 9:416-420.

ICPR (International Commission on Radiological Protection). 1973. Alkaline earth metabolism in adult man. ICPR Publication 20. Oxford: Pergamon Press.

IRIS (Integrated Risk Information System). 2005. Barium and Compounds (CASRN 7440-39-3). U.S. Environmental Protection Agency. [Online] Available: http://www.epa.gov/iris/subst/0010.htm. [accessed July 11, 2005].

Kanematsu, N., M. Hara, and T. Kada. 1980. Rec assay and mutagenicity studies on metal compounds. Mutat. Res. 77:109-116.

Kojola, W.H., G.R. Brenniman, and B.W. Carnow. 1978. A review of environmental characteristics and health effects of barium in public water supplies. Rev. Environ. Health 3:79-95.

Kopp, J.F. 1969. The occurrence of trace elements in water. Pp. 59-73 in Proceedings of the Third Annual Conference on Trace Substances in Environmental Health, D.D. Hemphill, ed. Columbia, MO: University of Missouri.

Kopp, J.F., and R.C. Kroner. 1967. Trace metals in waters of the United States. A five-year summary of trace metals in rivers and lakes on the United States (Oct 1, 1962-Sept 30, 1967). Federal Water Pollution Control Administration, U.S. Department of Interior, Cincinatti, OH.

Kopp, S.J., H.M. Perry, Jr., J.M. Feliksik, M. Erlanger, and E.F. Perry. 1985. Cardiovascular dysfunction and hypersensitivity to sodium pento-barbital

induced by chronic barium chloride ingestion. Toxicol. Appl. Pharmacol. 77(2):303-314.

Leguay, G., and A. Seigneuric. 1981. Cardiac arrhythmias in space. Role of vagotonia. Acta Astronaut. 8(7):795-801.

Lengemann, F.W., R.H. Wasserman, and C.L. Comar. 1959. Studies on the enhancement of radiocalcium and radiostrontium absorption by lactose in the rat. J. Nutr. 68(3):443-456.

Lewi, Z., and Y. Bar-Khayim. 1964. Food poisoning from barium carbonate. Lancet 14:342-343.

Lisk, D.J., C.A. Bache, and L.A. Essick. 1988. Absorption and excretion of selenium and barium in humans from consumption of Brazil nuts. Nutr. Rep. Int. 38:183-191.

McCabe, L.J., J.M. Symons, R.D. Lee, G.G. Robeck. 1970. Survey of community water supply systems. J. Am. Water Works Assoc. 62(11):670-687.

McCauley, P.T., and I.S. Washington. 1983. Barium bioavailability as the chloride, sulfate or carbonate salt in the rat. Drug. Chem. Toxicol. 6:209-217.

McCauley, P.T., B.H. Douglas, R.D. Laurie, and R.J. Bull. 1985. Investigations into the effect of drinking water barium on rats. Pp.197-210 in Inorganics in Drinking Water and Cardiovascular Disease, E.J. Calabrese, R.W. Tuthill, and L. Condie, eds. Princeton, NJ: Princeton Scientific Publications.

Merck Index. 1989. The Merck Index: An encyclopedia of chemicals, drugs, and biologicals, 11th Ed., S. Budavari, M.J. O’Neil, and A. Smith, eds. Rahway, NJ: Merck & Co.

Monaco, M., R. Dominici, R. Barisano, and G. Di Palermo. 1991. The evalustion of the presumed mutagenic activity of barium nitrate [in Italian]. Med. Lav. 82(5):439-445.

Morton, M.S., P.C. Elwood, and M. Abernathy. 1976. Trace elements in water and congenital malformation of the central nervous system in South Wales. Br. J. Prev. Soc. Med. 30:36-39.

Morton, W. 1945. Poisoning by barium carbonate. Lancet 2:738-739.

Newton, D., J. Rundo, and G.E. Harrison. 1977. The retention of alkaline earth elements in man, with special reference to barium. Health Phys. 33:45-53.

NIOSH (National Institute for Occupational Safety and Health). 1982. Health Hazard Evaluation Report; Sherwin-Williams Company, Coffeyville, Kansas. Report HETA/81-356-1183. National Institute for Occupational Safety and Health, Centers for Disease Control, Cincinatti, OH.

Nishioka, H. 1975. Mutagenic activities of metal compounds in bacteria. Mutat. Res. 31:185-190.

NRC (National Research Council). 1977. Drinking Water and Health. Washington, DC: National Academy Press. .

NRC (National Research Council). 1982. Drinking Water and Health, Vol. 4. Washington, DC: National Academy Press.

NTP (National Toxicology Program). 1994. Technical report on the toxicology and carcinogenesis studies of barium chloride dihydrate (CAS No.10326-

27-9) in F344/N rats and B6C3FI mice (Drinking Water Studies). NTP TR 432 (NIH Publication No. 94-3163. NTIS Publication No. PB94-214178). National Toxicology Program, U.S. Department of Health and Human Services, Research Triangle Park, NC.

Ogen, S., S. Rosenbluth, and A. Eisenberg. 1967. Food poisoning due to barium carbonate in sausage. Israel J. Med. Sci. 3:565-568.

Perry, Jr., H.M., E.F. Perry, M.W. Erlanger, and S.J. Kopp. 1985. Barium-induced hypertension. Pp.221-279 in Inorganics in Drinking Water and Cardiovascular Disease, E.J. Calabrese, ed. Princeton, NJ: Princeton Scientific Publications.

Perry, Jr., H.M., S.J. Kopp, E.F. Perry, and M.W. Erlanger. 1989. Hypertension and associated cardiovascular abnormalities induced by chronic barium feeding. J. Toxicol. Environ. Health 28(3):373-388.

Phelan, D.M., S.R. Hagley, and M.D. Guerin. 1984. Is hypokalemia the cause of paralysis in barium poisoning? Br. Med. J. 289:882.

Pierre, L.M., J.R. Schultz, R.L. Sauer, Y.E. Sinyak, V.M. Skuratov, N.N. Pratasov, and L.S. Bobe. 1999. Chemical Analysis of Potable Water and Humidity Condensate: Phase One Final Results and Lessons Learned. SAE Technical Paper Series no. 1999-01-2028. 29th International Conference on Environmental systems, Denver, CO, July 12-15.

Roeske, W.R., C. Higgins, J.S. Karliner, R.N. Berk, and R.A. O'Rourke. 1975. Incidence of arrhythmias and ST-segment changes in elderly patients during barium enema studies. Am. Heart J. 90(6):688-694.

Rossman, T.G., M. Molina, L. Meyer, P. Boone, C.B Klein, Z. Wang, F. Li, W.C. Lin, and P.L. Kinney. 1991. Performance of 133 compounds in the lambda prophage induction end point of the Microscreen assay and a comparison with S. typhimurium mutagenicity and rodent carcinogenicity assays. Mutat. Res. 260(4):349-367.

Roza, O., L.B. Berman. 1971. The pathophysiology of barium: hypokalemic and cardiovascular effects. J. Pharmacol. Exp. Ther. 177(2):433-439.

RTECS (Registry of Toxic Effects of Chemical Substances). 1999. Barium chloride, dihydrate (CAS # 103610-37-2). RTECS # CQ8751000. National Institute for Occupational Safety and Health, U.S. Department of Health and Human Services, Cincinnati, OH.

Rundo, J. 1967. The retention of barium-133 in man. Int. J. Radiat. Biol. Relat. Stud. Phys. Chem. Med. 13(3):301-302

Sax, N.I. 1984. Dangerous Properties of Industrial Chemicals, 6th Ed. New York: Van Nostrand Reinhold Co.

Schroeder, H., and M. Mitchener. 1975a. Life-term studies in rats: Effects of aluminum, barium, beryllium and tungsten. J. Nutr. 105:421-427.

Schroeder, H., and M. Mitchener. 1975b. Life-term effects of mercury, methyl mercury and nine other trace metals on mice. J. Nutr. 105:452-458.

Schroeder, H.A., I.H. Tipton, and A.P. Nason. 1972. Trace metals in man: Strontium and barium. J. Chronic. Dis. 25(9):491-517.

Sirover, M.A., and L.A. Loeb. 1976. Infidelity of DNA synthesis in vitro: Screening for potential metal mutagens or carcinogens. Science 194(4272):1434-1436.

Sowden, E.M., and S.R. Stitch. 1957. Trace elements in human tissue: 2. Estimation of the concentration of stable strontium and barium in human bone. Biochem. J. 67:104-109.

Syed, I.B., and F. Hosain. 1972. Determination of LD₅₀ of barium chloride and allied agents. Toxicol. Appl. Pharm. 22:150-152.

Syed, I.B., F. Hosain, and N.S. Mann. 1981. GI tract excretion of barium. Am. J. Protocol. Gastro. 32:16-20.

Talwar, K.K., and B.K. Sharma. 1979. Myocardial damage due to barium chloride poisoning. Ind. Heart J. 31:244-245.

Tarasenko, N.Y., O.A. Pronin, and A.A. Silayev. 1977. Barium compounds as industrial poisons (an experimental study). J. Hyg. Epidemiol. Microbiol. Immunol. 21(4):361-373.

Tardiff, R.G., M. Robinson, and N.S. Ulmer. 1980. Subchronic oral toxicity of BaCl₂ in rats. J. Environ. Pathol. Toxicol. 4:267-275.

Taylor, D.M., P.H. Bligh, and M.H. Duggan. 1962. The absorption of calcium, strontium, barium and radium from the gastrointestinal tract of the rat. Biochem. J. 83:25-29.

Tipton, I.H., P.L. Stewart, and P.G. Martin. 1966. Trace elements in the diet and excreta. Health Phys. 12:1683-1689.

Tuovinen, O.H., K.S. Button, and A. Vuorinen. 1980. Bacterial, chemical, and mineralogical characteristics of tubercles in distribution pipelines. J. Am. Water Works Assoc. 72:626-635.

Wetherill, S.F., M.J. Guarino, and R.W. Cox. 1981. Acute renal failure associated with barium chloride poisoning. Ann. Intern. Med. 95:187-188.

WHO (World Health Organization). 1990. Environmental Health Criteria 107: Barium. Sponsored by United Nations Environment Programme, International Labour Organisation, and World Health Organization, Geneva, Switzerland.

Wones, R.G., B.L. Stadler, and L.A. Frohman. 1990. Lack of effect of drinking water barium on cardiovascular risk factor. Environ. Health Perspect. 85:355-359.