4% United States
iPjif' Environmental Protectio
m »Agency
EPA/690/R-06/001F
Final
10-23-2006
Provisional Peer Reviewed Toxicity Values for
Aluminum
(CASRN 7429-90-5)
Superfund Health Risk Technical Support Center
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
Acronyms and Abbreviations
bw body weight
cc cubic centimeters
CD Caesarean Delivered
CERCLA Comprehensive Environmental Response, Compensation and
Liability Act of 1980
CNS central nervous system
cu.m cubic meter
DWEL Drinking Water Equivalent Level
FEL frank-effect level
FIFRA Federal Insecticide, Fungicide, and Rodenticide Act
g grams
GI gastrointestinal
HEC human equivalent concentration
Hgb hemoglobin
i.m. intramuscular
i.p. intraperitoneal
IRIS Integrated Risk Information System
i.v. intravenous
IUR inhalation unit risk
kg kilogram
L liter
LEL lowest-effect level
LOAEL lowest-observed-adverse-effect level
LOAEL(ADJ) LOAEL adjusted to continuous exposure duration
LOAEL(HEC) LOAEL adjusted for dosimetric differences across species to a human
m meter
MCL maximum contaminant level
MCLG maximum contaminant level goal
MF modifying factor
mg milligram
mg/kg milligrams per kilogram
mg/L milligrams per liter
MRL minimal risk level
MTD maximum tolerated dose
MTL median threshold limit
NAAQS National Ambient Air Quality Standards
NOAEL no-ob served-adverse-effect level
NOAEL(ADJ) NOAEL adjusted to continuous exposure duration
NOAEL(HEC) NOAEL adjusted for dosimetric differences across species to a human
NOEL no-ob served-effect level
OSF oral slope factor
p-IUR provisional inhalation unit risk
p-OSF provisional oral slope factor
p-RfC provisional inhalation reference concentration
1
-------
p-RfD
provisional oral reference dose
PBPK
physiologically based pharmacokinetic
ppb
parts per billion
ppm
parts per million
PPRTV
Provisional Peer Reviewed Toxicity Value
RBC
red blood cell(s)
RCRA
Resource Conservation and Recovery Act
RDDR
Regional deposited dose ratio (for the indicated lung region)
REL
relative exposure level
RfC
inhalation reference concentration
RfD
oral reference dose
RGDR
Regional gas dose ratio (for the indicated lung region)
s.c.
subcutaneous
SCE
sister chromatid exchange
SDWA
Safe Drinking Water Act
sq.cm.
square centimeters
TSCA
Toxic Substances Control Act
UF
uncertainty factor
l^g
microgram
[j,mol
micromoles
voc
volatile organic compound
11
-------
10-23-2006
PROVISIONAL PEER REVIEWED TOXICITY VALUES FOR
ALUMINUM (CASRN 7429-90-5)
Background
On December 5, 2003, the U.S. Environmental Protection Agency's (EPA's) Office of
Superfund Remediation and Technology Innovation (OSRTI) revised its hierarchy of human
health toxicity values for Superfund risk assessments, establishing the following three tiers as the
new hierarchy:
1. EPA's Integrated Risk Information System (IRIS).
2. Provisional Peer-Reviewed Toxicity Values (PPRTV) used in EPA's Superfund
Program.
3. Other (peer-reviewed) toxicity values, including:
~ Minimal Risk Levels produced by the Agency for Toxic Substances and Disease
Registry (ATSDR),
~ California Environmental Protection Agency (CalEPA) values, and
~ EPA Health Effects Assessment Summary Table (HEAST) values.
A PPRTV is defined as a toxicity value derived for use in the Superfund Program when
such a value is not available in EPA's Integrated Risk Information System (IRIS). PPRTVs are
developed according to a Standard Operating Procedure (SOP) and are derived after a review of
the relevant scientific literature using the same methods, sources of data, and Agency guidance
for value derivation generally used by the EPA IRIS Program. All provisional toxicity values
receive internal review by two EPA scientists and external peer review by three independently
selected scientific experts. PPRTVs differ from IRIS values in that PPRTVs do not receive the
multi-program consensus review provided for IRIS values. This is because IRIS values are
generally intended to be used in all EPA programs, while PPRTVs are developed specifically for
the Superfund Program.
Because science and available information evolve, PPRTVs are initially derived with a
three-year life-cycle. However, EPA Regions or the EPA Headquarters Superfund Program
sometimes request that a frequently used PPRTV be reassessed. Once an IRIS value for a
specific chemical becomes available for Agency review, the analogous PPRTV for that same
chemical is retired. It should also be noted that some PPRTV manuscripts conclude that a
PPRTV cannot be derived based on inadequate data.
1
-------
10-23-2006
Disclaimers
Users of this document should first check to see if any IRIS values exist for the chemical
of concern before proceeding to use a PPRTV. If no IRIS value is available, staff in the regional
Superfund and RCRA program offices are advised to carefully review the information provided
in this document to ensure that the PPRTVs used are appropriate for the types of exposures and
circumstances at the Superfund site or RCRA facility in question. PPRTVs are periodically
updated; therefore, users should ensure that the values contained in the PPRTV are current at the
time of use.
It is important to remember that a provisional value alone tells very little about the
adverse effects of a chemical or the quality of evidence on which the value is based. Therefore,
users are strongly encouraged to read the entire PPRTV manuscript and understand the strengths
and limitations of the derived provisional values. PPRTVs are developed by the EPA Office of
Research and Development's National Center for Environmental Assessment, Superfund Health
Risk Technical Support Center for OSRTI. Other EPA programs or external parties who may
choose of their own initiative to use these PPRTVs are advised that Superfund resources will not
generally be used to respond to challenges of PPRTVs used in a context outside of the Superfund
Program.
Questions Regarding PPRTVs
Questions regarding the contents of the PPRTVs and their appropriate use (e.g., on
chemicals not covered, or whether chemicals have pending IRIS toxicity values) may be directed
to the EPA Office of Research and Development's National Center for Environmental
Assessment, Superfund Health Risk Technical Support Center (513-569-7300), or OSRTI.
This document has passed the STSC quality review and peer review evaluation indicating
that the quality is consistent with the SOPs and standards of the STSC and is suitable for use by
registered users of the PPRTV system.
INTRODUCTION
Verified toxicity values for aluminum (Al) and its compounds are unavailable on IRIS or
HEAST (U.S. EPA, 2006, 1997), except for a chronic oral RfD of 4E-4 mg/kg-day for aluminum
phosphide. However, occupational guidelines and standards have been established for a number
of chemical and physical forms of Al, including, from ACGIH, 8-hour TWA-TLVs of 10 mg/m3
for the compound as a metal dust or oxide, 5 mg/m3 as "pyro" powders or welding fumes, and 2
mg/m3 for soluble salts or organic forms of the metal (ACGIH, 1998). From NIOSH, 10-hour
TWA-RELs of 10 mg/m3 are specified for "total" Al dust versus 5 mg/m3 for the respirable
portion (NIOSH, 1994). NIOSH covers all other forms of the metal by identical values to those
specified by ACGIH (ACGIH, 1998). OSHA PELs for Al include an 8-hour TWA value of 15
mg/m3 for "total" metal dust, versus 5 mg/m3 for the respirable portion (NIOSH, 1994). The
U.S. EPA's CARA list (U.S. EPA, 1994) cites a HEA for Al (U.S. EPA, 1987), and AT SDR has
updated its toxicological profile of the element (ATSDR, 1998).
2
-------
10-23-2006
The U.S. FDA (2000) has specified a maximum aluminum concentration of 25 mcg/L in
large-volume parenterals (LVP) used in total parenteral nutrition (TPN). The FDA regulation
applies to all LVPs used in TPN, including but not limited to parenteral amino acid solutions,
highly concentrated dextrose solutions, parenteral lipid emulsions, sodium chloride and
electrolyte solutions, and sterile water for injection.
Research papers pertinent to the potential toxicological and carcinogenic effects of A1
were sought through computer searches of the HSDB, RTECS, MEDLINE and TOXLINE (and
its subfiles) databases, covering the time period 1995-1999. The literature searches were
conducted in June, 1999.
REVIEW OF PERTINENT DATA
The review by Stokinger (1981) gives an account of A1 as an all-pervasive component of
products that are central to the daily lives of most Americans. For example, the metal is a crucial
part of manufactured products for the building, automobile and container industries, while A1 as
powder or flake is a component in a number of consumer products, such as paints, fireworks, etc.
A1 complexes and minerals are used in the brewing and paper industries, and as coagulants for
water purification. Aluminum oxide finds application in abrasives, as a catalyst or absorbent,
and as a component in fillers. Aluminum chloride is included in cosmetic formulations such as
deodorants.
Human exposure to A1 arises principally from food and water, through its widespread use
in food additives, packaging and cooking utensils and Al-containing medications, particularly
antacid, buffered aspirin, anti-ulcer and anti-diarrheal formulations (Marquis, 1989; Li one,
1985). Pennington and Schoen (1995) estimated daily A1 intakes of 0.1-0.3 mg/kg-day for
infants and children 6 months-6 years of age and 0.1-0.18 mg/kg-day for older children and
adults, based on the FDA Total Diet Study (1993) and the U.S. Department of Agriculture
Nationwide Food Consumption Survey (1987-1988). These data are in broad agreement with
those of Wilhelm et al. (1995) who reported the dietary intake of A1 in German children (living
in the Duisberg area) as ranging from 0.008 to 0.11 mg Al/kg-day. In addition, these values are
consistent with a range of 1-20 mg/day (0.014-0.3 mg/kg -day) for normal oral daily Al intake
from food and water reported by other investigators (Ganrot, 1986; Iyengar et al., 1987; Wilhelm
et al., 1990). However, users of Al-containing medications can ingest much larger amounts of
the element, possibly as high as 840-5000 mg/day (12-71 mg/kg-day) from antacids, 126-728
mg/day (1.8-10.4 mg/kg-day) from buffered aspirins and 828 mg/day (11.8 mg/kg-day) from
anti-ulcer compounds when taken at recommended dosages (Lione, 1985).
Toxicokinetics of Aluminum
There is a large amount of information available on the absorption, transfer from tissue to
tissue and elimination of Al from the body, including data that have been amassed from studies
on either human volunteers or laboratory animals. In general, the chemical appears to be poorly
absorbed from the gastrointestinal tract, though the portion of the load that is retained will vary
depending on the concentration, the chemical species administered, the fasting or fed state of the
3
-------
10-23-2006
host, gastrointestinal pH, animal model, etc. For example, Yokel and McNamara (1988)
administered single oral doses of a number of A1 compounds (both water soluble and insoluble)
to New Zealand white rabbits and obtained absorbed proportions of the load ranging from 0.27%
to 27%. Fractional uptake of A1 in humans under normal conditions (i.e., with no intake of large
quantities of A1 from medicine) was estimated to be 0.1-0.3% assuming an intake of 20 mg
Al/day (0.3 mg Al/kg-day) and urinary excretion of 20-50 jag Al/day (0.3-0.7 |ig Al/kg-day)
(Ganrot, 1986). However, little information is available on the actual mechanism by which the
element and its compounds are transported across the brush border. (Wilhelm et al., 1990; Lione,
1985).
Although the overall extent of Al absorption is poor following oral exposure, there may
be significant intake of the compound by those taking large amounts of Al compounds in
patented remedies. As stated, absorption of Al is influenced by gastrointestinal conditions and
content because Al can form various complexes with different solubilities and oxidation states
depending on pH and interactions with dietary constituents. At low pH (3-5) in aqueous
solutions, the soluble (ionic) forms of the Al prevail (Al3+); at high pH (>8), Al in the form of
soluble aluminum oxide is present; and at pH 5-8, the element is predominantly in the form of
aluminum hydroxide, which is insoluble (van der Voet and de Wolff, 1986; Wilhelm et al.,
1990). Ingested constituents that can influence absorption by forming complexes with Al
include phosphate, fluoride, calcium, citrate and lactate. For example, Al is used to bind dietary
phosphorus and decrease its absorption as a control for hyperphosphatemia, and citrate and
lactate are complexing agents that can significantly increase Al absorption (Slanina et al., 1984,
1985, 1986; Partridge et al., 1989; Domingo et al., 1991; Ittel et al., 1991; Lione, 1985; Wihelm
et al., 1990).
A number of recent reports of studies on the gastrointestinal absorption of Al have
examined the influence of organic anions such as citrate. In general, the presence of such
components appears to enhance the absorption of Al, within narrow limits. For example, Deng
et al. (1998) administered a single oral dose of either distilled water, 2 mmoles/L aluminum
chloride or 2 mmoles/L aluminum chloride plus 2 mmoles/L sodium citrate to six male Wistar
rats/group. Animals were bled at 1, 2 and 4 hours after dosing, then terminated after 6 hours.
Inductively coupled plasma (ICP) was used to measure Al concentrations in blood, bone (tibia),
kidney, liver and the intestinal wall. Irrespective of treatment, the appearance of Al in the blood
of dosed groups peaked after 1 hour, with the concentration of the element at higher levels in
those animals receiving citrate in addition to aluminum chloride. In those animals receiving
aluminum chloride alone, significant tissue concentrations of the element were restricted to the
gastrointestinal wall. Those receiving citrate displayed measurable quantities of the element in
several of the other monitored tissues, including bone.
Sutherland and Greger (1998a) used a similar dosing regimen to examine the kinetics of
absorption and elimination of Al in male Sprague-Dawley rats that had received a single oral
dose of 0, 0.25, 0.5 or 1 mmoles/L/kg body weight aluminum lactate in 1 mL of 16% citrate.
Concentrations of Al in serum, liver, kidney or bone (tibia) were measured at various post-dosing
time intervals up to 6 hours. Depending on the dose, absorption factors for Al of up to 4.2% of
the administered dose were observed, with the greater proportion retained in bone. The authors
reported a slower rate of absorption in those animals receiving Al at the higher doses, an
4
-------
10-23-2006
observation potentially indicating reduced gut motility and/or saturation of the transcellular
absorption processes at the higher concentrations. Aluminum deposited in kidney and bone
appeared to turn-over at a slower rate than in the liver.
The influence of citrate on the gastrointestinal absorption of A1 in man was examined
directly by Taylor et al. (1998) who administered a drink containing A1 and citrate to three
volunteers. Aluminum and citrate concentrations were monitored in serial blood and urine
samples for up to 24 hours. The kinetics of citrate and Al differed markedly, the former peaking
in plasma after 32 minutes, versus 87 minutes for Al. This suggests that Al probably does not
cross the gastrointestinal barrier as the citrate. Furthermore, the authors reported that the overall
extent of Al absorption had probably not exceeded 1% in their experiment, a finding that
contrasts with the higher values reported by Sutherland and Greger (1998a) in Sprague-Dawley
rats and by Deng et al. (1998) in Wistar rats.
As discussed in a report by Glynn et al. (1999), gastrointestinal absorption of Al from
aqueous media will be almost impossible to predict, because of the likelihood that the element
will become absorbed to food particles in the intestinal lumen. Accordingly, depending on the
dose, mode of delivery and caloric state of the experimental animal (fed/fasted), significant
amounts of aqueous forms of Al will be absorbed only when available binding sites on food have
become saturated. This presents an inherently complex overall picture of the element's
absorption since, additionally, the normal dietary content of Al will be substantial. Thus, it may
be assumed that some sequestered Al will be absorbed along with non-sequestered water soluble
forms of the element, while the rest will be retained within the gastrointestinal tract.
Sutherland and Greger (1998b) used their aluminum lactate in 16% citrate dosing
regimen to examine the comparative importance of biliary versus urinary excretion of Al. Five
to seven male Sprague-Dawley rats/group who had previously received an implanted bile
cannula were treated by gavage. Another similarly-treated cohort of five animals/group were
housed in metabolic cages immediately after dosing to provide 0- to 3-hour and 3- to 6-hour
urine specimens. At termination, all animals were sacrificed and exsanguinated, and tissue, bile
and urine samples were measured by graphite furnace atomic absorption spectroscopy. Among
the key findings to emerge from this study was the incremental appearance of Al in bile as early
as 15 minutes after dosing. However, overall amounts of Al were greater in the 3-hour urine
samples than those that had accumulated in bile samples collected within a similar time frame.
The fact that control rats excreted 3 times more Al in bile than in urine during the first 3 hours
after dosing led the authors to conclude that, at low exposure to Al (in controls receiving Al
solely from food), the liver is capable of excreting the element to the bile, a mechanism that
becomes saturated as the level of Al administration becomes increased. Thereafter, urinary
excretion becomes the primary route of elimination in circumstances of Al overload.
Aluminum can also be absorbed by inhalation as indicated by age-related deposition in
the lungs of the general population and exposure-related increased blood and urine
concentrations in workers exposed to Al (Bast-Pettersen et al., 1994; Sjogren et al., 1996;
Hosovski et al., 1990; Wilhelm et al., 1990; U.S. EPA, 1987). Aluminum occurs primarily in
particulate form in the ambient atmosphere and as various dusts and fumes during its production
and use. Common forms of inhaled Al include aluminum oxide (alumina; AI2O3), pyro powders
5
-------
10-23-2006
(powder and flake Al-treated to reduce surface oxidation), A1 welding fume and soluble salts
(e.g., aluminum chloride and sulfate) (ACGIH, 1998).
Neurotoxicity as a Primary Toxicological Effect of Aluminum
One of the greatest health concerns regarding A1 is its neurological effects. The first
evidence for Al-induced neurotoxicity in humans was seen in patients who, as a result of
receiving long-term hemodialysis for chronic renal failure, developed a degenerative
neurological syndrome (dialysis dementia) characterized by the gradual loss of motor, speech
and cognitive functions (Alfrey, 1993). This dementia, attributable to A1 in the dialysate, is
usually fatal within 6-9 months after the first clinical signs appear. In addition, many patients
received high oral doses of A1 to act as phosphate binders. Autopsies of these patients revealed
increased concentrations of A1 in the gray matter and cerebral spinal fluid (CSF) but no evidence
of neurofibrillary degeneration (NFD) despite the elevated A1 levels. Once the connection
between A1 and dialysis dementia was established, A1 was removed from dialysis fluid and the
incidence of dementia rapidly declined, thereby strengthening the argument that A1 was a causal
agent in dialysis dementia (Ganrot, 1986).
Amyotrophic Lateral Sclerosis (ALS) and Parkinson's Disease (PD) are other
neurological diseases which have been associated with A1 exposure. ALS is a progressive
disease of the Central Nervous System (CNS) that is characterized by an accumulation of
neurofibrillary tangles. In Guam, southern West New Guinea and parts of Japan, there is an
unusually high prevalence of ALS and PD. This may be related to the natural abundance of A1
coupled with the virtual lack of magnesium and calcium in the drinking water supplies and soil
of these areas. In a study designed to evaluate effects of high A1 and low calcium levels in the
diet, much like the conditions associated with Guam and other similar areas, cynomolgus
monkeys were placed on a low calcium diet either with or without supplemental A1 and
manganese (Garruto et al., 1989). Chronic calcium deficiency alone produced neurodegenerative
effects, although neurofibrillary changes were most frequently seen in the monkeys on a low
calcium diet supplemented with Al and manganese.
Though a cause and effect relationship between Al and three forms of chronic
encephalopathy in humans: senile dementia of the Alzheimer type (SDAT, Alzheimer's Disease),
endemic Amyotrophic Lateral Sclerosis (ALS) and endemic Parkinsonism-dementia (PD, a
mixture of Parkinsonism and senile dementia) has been suggested, there is no firm evidence that
it plays a causal role in the development of these diseases (Ganrot, 1986; Lione, 1985). The
condition is degenerative and characterized by the progressive loss of speech, motor and
cognitive functions, with death typically occurring within 1-6 months. Autopsies of patients
revealed increased concentrations of Al in the gray matter and cerebral spinal fluid (CSF),
though with no conclusive evidence of NFD or other neuropathological changes despite the
elevated Al levels.
The neurotoxicity of Al is well documented in certain animal species. Aluminum induces
a spectrum of behavioral abnormalities and brain neurofibrillary degenerative changes in rabbits
and cats when injected intracranially or parenterally in high doses, though hamsters and monkeys
are less sensitive (Ganrot, 1986; Lione, 1985). Such studies have been designed as models for
6
-------
10-23-2006
the possible neurotoxicological effects of A1 in humans. However, it should be noted that the
neurofibrillary changes in affected animals differ in morphological detail from those associated
with SDAT. As discussed further in the Oral Toxicity section, oral doses of A1 can also induce
neurobehavioral effects in adult mice and rats and in their developing offspring. In general, such
neurotoxic effects of A1 appear to be more subtle than those induced through routes of
administration that by-pass the gastrointestinal tract, perhaps reflecting the lower doses of A1
reaching the brain.
Recent reports of studies on the effects of A1 on neurotoxicity in animals have sought to
define the biochemical mechanisms that are impaired when A1 crosses the blood-brain barrier.
However, a unifying concept has yet to emerge, though the passage of the element into various
regions of the brain has been clearly demonstrated (Deloncle et al., 1995). Among the many
biochemical functions and processes that appear to be perturbed by the presence of Al in the
brain are the peroxidation status of biological membranes (Katyal et al., 1997; Deloncle et al.,
1999), inhibition of the neuronal glutamate-nitric oxide-cyclic GMP pathway (Cucarella et al.,
1998), and the marked reduction of protein- and non-protein-bound thiols and the specific
activity of Na+/K+ and Mg++ ATPases (Katyal et al., 1997). The relative importance of each of
these mechanisms and how (or whether) they interact to bring about the observed physiological
changes remains unclear.
Other Effects of Aluminum
Osteomalacia was frequently observed among long-term dialysis patients with
neurological signs and is commonly attributed to Al overload (Ganrot, 1986; Li one, 1985). This
bone condition is characterized by widened osteoid (unmineralized bone matrix) with no fibrosis,
reduced mineralization rate, skeletal pain and a strong tendency for fractures, lack of response to
vitamin D therapy and increased Al concentration in bone. Effects on bone histology and
elevated bone Al levels have also been observed in patients with normal renal function who
received total parenteral nutrition with Al-contaminated casein as a protein source, and in
parenteral Al loading induced osteomalacia in rats and dogs (Lione, 1985).
There are a number of published reports of studies in which the carcinogenicity of
aluminum compounds has been evaluated. These include oral exposure studies in which the
compounds were made available to experimental animals in the drinking water or diet
(Schroeder and Mitchener, 1975a,b: Oneda et al., 1994), and inhalation epidemiological studies,
in which the incidence of tumor formation in persons exposed to aluminum-containing dusts and
fumes in an occupational setting was compared to unexposed individuals (Spinelli et al., 1991;
Theriault et al., 1984, 1990; Armstrong et al., 1986; Tremblay et al., 1995; Selden et al., 1997;
Cullen et al., 1996; Dufresne et al., 1996; Ronneberg and Langmark, 1992). However, it has
been generally concluded that the inferential association between exposure to Al and marginally
increased incidences of tumors of the bladder and/or lung are confounded because of the co-
exposure of subjects in such settings to other harmful and potentially carcinogenic substances,
such as polycyclic aromatic hydrocarbons (PAHs and coal tar pitch volatiles (CTPV) (Ronneberg
and Langmark, 1992). Therefore, the issue of the potential carcinogenicity of Al compounds
remains uncertain.
7
-------
10-23-2006
Human Studies
Oral Exposure
Few reports have been identified that address the toxicological effects of A1 in humans
exposed orally. Furthermore, in a review, Reiber et al. (1995) pointed to the conflicting findings
that have been reported when the incidence of neurological symptoms has been assessed in
relation to Al exposure in either cross-sectional, ecological or case-control epidemiological
studies. Among the more recent studies that have used this approach, Martyn et al. (1997)
discussed the findings of a case-control study involving 441 men in England and Wales who
were afflicted with either Alzheimer's disease, brain cancer, dementia or other neurological
conditions. Assessing the historical exposure of these subjects failed to establish a link between
Al in drinking water at the prevailing concentrations (below 0.2 mg/L) and the incidence of one
or more of the conditions under investigation. No data were located regarding the oral
carcinogenicity of aluminum compounds in humans.
Inhalation Exposure
Neurobehavioral effects were evaluated in a group of 87 Al foundry workers who were
occupationally exposed to 4.6-11.5 mg/m3 Al fumes and dust for a mean of 12.0 years [standard
deviation (SD) 4.5 years, shortest exposure 6 years] compared to an unexposed control group
(n=60) who were matched for age, job seniority and social status to exposed subjects (Hosovski
et al., 1990). It is reported that environmental Al concentrations were measured for each worker
separately during the winter and summer, implying that personal sampling may have been used
and that the contributing concentrations are time-weighted averages. In certain places, the
number of particles ranged as high as 329-1020/cm2 air, and dust particle sizes were <1, 1-5 and
<5 microns in 65.6, 26.6 and 7.6% of the samples, respectively. Tests of psychomotor ability
(simple and complex reaction time, oculomotor coordination), intellectual ability (Wechsler
intelligence, performance intelligence and verbal intelligence quotients and Wechsler subtests on
information processing, memory, understanding, calculation, coding, picture completion, picture
grouping, object assembling, assembling of cubes and common concepts) and cerebral damage
(Bender visual motor test) were conducted. Performance of the exposed workers was found to
be significantly (p<0.02) impaired on the complex reaction time, oculomotor coordination,
memory, coding, picture completion and object assembling tests. However, the investigators
noted that the performance deficits had no clinical manifestations, and that additional studies
were probably needed to confirm the possibility of cerebral damage. The study yielded a lowest
available non-duration adjusted LOAEL of 4.6 mg Al/m3 for psychomotor and cognitive
impairment during repeated 8-hour occupational exposures (Hosovski et al., 1990), that could be
corrected for discontinuous exposure (10 m3/20 m3 and 5 days/7 days) to yield a LOAELrec of
1.64 mg/m3 Al.
Aluminum oxide powders were administered to Canadian miners (mainly underground
gold and uranium miners) in known exposures as a means of prophylaxis against silicosis
(Stokinger, 1981; Rifat et al., 1990). Data in which more than 42 million Al treatments
(-150,000 man-years) had been given over a period of 27 years ending in 1971 were reviewed
8
-------
10-23-2006
by Stokinger (1981). The effectiveness of this treatment is uncertain but no lung damage or
other ill effects (not specified) were observed. The powders (Mclntyre powder) were prepared
by grinding A1 pellets so that 96% of the particles were <1.2 |im in diameter. During this
process most of the particles became oxidized to aluminum oxide; the powder contained 85%
aluminum oxide and 15% elemental Al. According to Stokinger (1981), recommended exposure
concentrations were 30,000 particles of respirable size per cubic centimeter (ppcc) for 10
minutes/day or 10,000-20,000 ppcc for 20 minutes/day (total treatment days not indicated). Rifat
et al. (1990) stated that the recommended exposure was to an Al dust concentration of 20,000-
34,000 parts per ml air in the miners' changing rooms before each shift for 10 minutes.
Stokinger (1981) reported that the 30,000 ppcc concentration corresponds to -350 mg/m3, which
is equivalent to an 8-hour average concentration of 2 mg/m3. Based on the Stokinger (1981) data
and the fact that one unspecified study used levels 30 times higher than advised, the TLV of 10
mg/m3 is recommended for Al dust (ACGIH, 1998).
The increasing awareness of the potential neurotoxicity of Al has resulted in a number of
investigations of the incidence of neurotoxicological symptoms in Al workers. Although
treatment with Mclntyre powder had not produced apparent adverse effects, a neurobehavioral
evaluation of male miners (261 exposed to Mclntyre powder, 346 unexposed) who started
working between 1940 and 1979 (additional duration data not reported) was performed in
1988-1989 (Rifat et al., 1990). There were no significant differences between exposed and
unexposed miners in reported diagnoses of neurological disorder. Results of cognitive testing
(Mini-Mental State Examination for general cognitive function, Ravens colored progressive
matrices test for reasoning and Symbol Digit Modalities Test for spatial perceptual accuracy and
information processing), however, showed that the exposed group had significantly (p<0.001)
impaired performance on at least one test, and when all test scores were summed. Also, the
likelihood of scores in the impaired range increased with duration of exposure.
A neurologic syndrome was described in Al smelting plant potroom workers (White et
al., 1992). Twenty-five men were evaluated for suspected work-related neurologic illness based
on findings in three patients studied previously. The average duration of employment was 18.7
years (SD, 3.6; range, 12-23 years), 15 of the patients were working at the time of evaluation,
and 10 had taken early retirement or medical leave due to workplace-related symptoms (mean
length of time since exposure was 1.3 years ranging from 0.2-5 years). Quantitative exposure
level data were not reported, but 21 of the workers had been employed in the potroom prior to
installation of fume hoods for a mean duration of 5.3 years (range 3-7 years). Symptoms most
often reported by the patients were frequent loss of balance (88%), memory loss (84%) and joint
pain (84%>); other symptoms included dizziness (80%), numbness (80%), parasthesias {12%) and
tremor (68%). Neurologic examinations showed mild to moderate signs of lack of coordination
(tremor, dyssynergy of upper extremity limb movement or ataxia) in 84% of the patients.
Neuropsychologic effects were evaluated in 21 of the patients using the Wechsler Adult
Intelligence Scale-Revised (intellectual functioning), Wide Range Achievement Test-Revised
(academic functioning), Halstead-Reitan Neuropsychological Test Battery (neuropsychological
assessment) and Minnesota Multiphasic Personality Inventory (personality functioning).
Memory function was assessed with the Wechsler Memory Scale (14 patients) and Wechsler
Memory Scale-Revised (8 patients). The memory function evaluation showed mild to moderate
impairment on subtests of immediate recall for verbal or visual information (70-75%) of the
9
-------
10-23-2006
tested patients) and delayed verbal or visual recall (50-70%). Other effects included mild or
moderate impairment on Halstead-Reitan tests of abstract reasoning and flexible thinking (42%
of the tested patients), memory for tactile information (53%) and sustained attention and
discrimination of tonal and speech patterns (44 and 64%, respectively). On the Wechsler
memory and Halstead-Reitan tests, mild and moderate impairment was defined as scores 1.5-2
and >2 standard deviations below the mean of the normal population, respectively. Most (89%)
of the patients tested with the Minnesota Multiphasic Personality Inventory had abnormally
elevated scores (>2 SDs above the population mean) indicative of clinical depression.
Significant positive correlations were found between severity of incoordination (signs and
symptoms) and degree of exposure (qualitative) before the introduction of the ventilation hoods.
White et al. (1992) noted two other studies that described neurologic problems among A1
smelter workers. Thus, an evaluation of 444 electrolysis workers found neuropsychiatric
changes in 123 (28%), "neurotic syndromes" in 89 (20%) and "slight pyramidal and cerebellar
changes" in 39 (9%) (Langauer-Lewowicka and Braszczynska, 1983). In the second study,
symptoms including mental confusion, concentration and memory problems were described in
six potroom workers (Cawthon, 1988).
In another study of Al production workers, neuropsychological effects were assessed in
38 elderly men who had been exposed for at least 10 years exclusively in the potroom (n=14),
foundry (n=8) or other manual labor departments of the same plant (n=16, control group) (Bast-
Pettersen et al., 1994). The mean ages and employment durations of the groups were in the
ranges of 62.5-63.5 and 19.2-19.6 years, respectively. The men were examined soon after or just
before retirement in 1991. Limited environmental monitoring data indicates that the degree of
Al exposure varied between the subgroups and over the years. Average annual total dust
concentrations in the potroom were reduced significantly from 9.5 mg/m3 in 1977 to 3.0 mg/m3
in 1990. Aluminum levels were not specifically reported, but the average Al content in the total
potroom dust was approximately 20% by weight; other constituents of the dust included fluoride
and coal tar pitch components. Data from an Al uptake/excretion study of workers from the
same plant indicated that the level of Al exposure was approximately 8 times higher in the
potroom than in the foundry (0.48 and 0.06 mg/m3, respectively) (Drablos et al., 1992). Medical
examinations (including lung function, standard laboratory tests and serum and urine Al
concentrations) and a neuropsychological test battery were performed. The battery assessed six
mental functions (neuropsychiatric symptoms, motoric/sensoric, reaction time, psychomotor
speed/efficiency, memory/learning and intelligence) using a questionnaire and 15 different
objective tests. Some subtle deficits were found in potroom workers that were not considered to
be indicative of a significant neurological syndrome. The findings in potroom workers included
a subclinical tremor as indicated by results of a static steadiness test [time scores on one of two
test indices were significantly worse in comparison with the control group (84% slower,
p=0.03)], and possible tendencies (i.e., test results that were about 1 SD below normal mean
values but not statistically significant) for increased risk of impaired visuospatial organization
(Block Design subtest of the Wechsler Adult Intelligence Scale) and psychomotor tempo (one
Halstead ReitanTrail Making test). Although these findings were not considered to be indicative
of a neurologic syndrome, it was suggested that they may be early signs of CNS impairment.
Additionally, the finding of a subclinical tremor seems to be consistent with the tremor and other
10
-------
10-23-2006
signs of incoordination observed in 84% of the patients in the White et al. (1992) study
summarized above.
Studies of Al welders are consistent with those of Al smelter workers in indicating that
occupational exposure to Al can be neurotoxic. CNS function was evaluated in 17 welders who
had an average of 15 years (range 5-27 years) experience, with the last 4 years exclusively with
Al (Hanninen et al., 1994). Most of the welders had equipment that ventilated the welding
masks but the respiratory protection was not always used. The assessment included
measurements of serum and urinary Al, neuropsychological tests (simple reaction time, three
tests for psychomotor speed, two tests for visual and spatial ability, four memory tests and two
verbal ability tests), a symptom questionnaire and neurological interview, quantitative
electroencephalography (QEEG) and P-300 event-related auditory-evoked responses. Serum and
urine Al levels were 3.5 and 8.5 times higher, respectively, than an unexposed reference
population. The welders performed normally on the neuropsychological tests, although
correlation analysis of test scores and exposure parameters showed weak negative associations
between the four memory tests and urinary Al level and a positive association between the
variability (standard deviation) of visual reaction times and serum Al levels. Analysis of the
QEEG data showed that serum Al levels were positively correlated with the amount of delta and
theta activity in the brain frontal region and negatively correlated with the amount of alpha
activity in the frontal region. Results of this study (disturbances of memory and attention, QEEG
changes similar to those in patients with Al encephalopathy) were interpreted as consistent with
known CNS effects of Al, but insufficient for establishing a definite relationship between Al
exposure and effects.
In another study of Al welders, CNS evaluations were performed on 38 men who had at
least 5 years exposure (mean 17.1 years) and a control group of 44 railway track welders
exposed to metal fumes other than Al (mean 13.8 years) (Sjogren et al., 1996). Limited
monitoring data indicated that the median exposure to welding fumes was 10 mg/m3 and that the
Al content was 40% of the total fumes. Symptom questionnaires, psychological tests (simple
reaction time, finger tapping speed and endurance, digit span, vocabulary, tracking, symbol digit
coding, cylinders, olfactory threshold and Luria-Nebraska motor scale), neurophysiological
indices [electroencephalography, P-300 auditory-evoked responses, brain-stem auditory evoked
responses and diadochokinesis (ability to perform rapidly alternating movements with one limb)]
and blood and urine Al levels were assessed. The blood and urine Al concentrations were
approximately 3 and 7 times higher in the Al welders than in the controls, but there were no clear
correlations between duration of exposure to Al and concentration of Al in blood or urine. The
Al welders reported more acute CNS symptoms (e.g., concentration difficulties) and had
decreased motor function in five tests (finger tapping in non-dominant hand, two tasks from the
Luria-Nebraska motor scale, pegboard peg movement with dominant hand, amplitude of
diadochokinesis in dominant hand) when compared to the control group. Urinary Al
concentration was significantly correlated with acute CNS symptoms, but not with any of the
performance measures. To further study possible dose-effect relationships of Al exposure, the
Al welders were combined with the control group and divided into three exposure categories
according to urinary Al levels, using the 50th and 75th percentiles as category dividers. The
group with the highest mean urinary Al level had significantly more acute CNS symptoms and
significantly reduced performance on one of the motor function tests (a Luria-Nebraska motor
11
-------
10-23-2006
scale task) when compared to the group with the lowest A1 level. In an earlier study of 65
welders with >10 years of exposure to A1 fumes, the highest exposure category (based on
exposure duration) was 2.8 times more likely than unexposed workers to have three or more
neuropsychiatric symptoms (Sjogren et al., 1990).
A body of epidemiological evidence has pointed to an increased incidence of cancers of
various kinds in workers employed in the aluminum production industry. However, as discussed
in a review by Ronneberg and Langmark (1992), the concern about potential cancer hazards in
the aluminum industry has primarily arisen because of exposures to polycyclic aromatic
hydrocarbons (PAHs) and coal tar pitch volatiles (CTPVs) rather than to Al per se. Thus, while
a number of studies have provided inferential data linking occupationally exposed aluminum
workers with an increased risk of developing tumors of the bladder or lung (Gibbs, 1985;
Theriault et al., 1984, 1990; Armstrong et al., 1986; Spinelli et al., 1991; Pearson et al., 1993;
Tremblay et al., 1995), it would be unwise to ascribe any excess tumor formation to the effects of
Al in view of the concurrent exposure to well-documented carcinogenic PAHs such as
benzo(a)pyrene. The issue is further complicated by the likely exposure of production workers
to other substances such as fluorides, sulfur dioxide, aromatic amines and asbestos (Ronneberg
and Langmark, 1992; Tremblay et al., 1995; Dufresne et al., 1996), and to the possible effects of
cigarette smoking in affected individuals. Consequently, these studies have failed to provide
direct evidence for the carcinogenicity of Al fumes and dusts.
Animal Studies
Oral Exposure
Numerous subchronic animal studies were located in the biomedical/toxicological
literature but only those that define the threshold region of the oral dose-response relationship are
summarized in this paper. A major limitation of many of the studies of Al toxicity is the lack of
complete information on total dietary (e.g., food and drinking water) intake of Al and of other
elements that are known to effect Al biokinetics and toxicity (e.g., calcium and magnesium).
Estimated or reported dosages used in studies in which Al content of the basal diets are not
reported must be assumed to underestimate the actual experimental dosages. The magnitude of
the underestimate may be considerable. For example, a range of Al contents of 200-1200 mg
Al/kg for commercial grain-based diets (Golub et al., 1992b) would provide 30-200 mg Al/kg
bw-day in a subchronic or chronic mouse bioassay [based on U.S. EPA (1988) default values for
body weight and food intake]. On this basis, studies in which complete dietary Al intakes were
not reported or could not be estimated may provide some information about the hazards of oral
exposure to Al but are inappropriate for establishing NOAELs or LOAELs for the critical effect
of Al. NOAELs and LOAELs from studies that provide estimates of total Al dosages, or
otherwise provide information relevant to determining the NOAEL/LOAEL boundary for the
critical effect of Al are presented in Table 1 and are summarized below.
Systemic toxicity
Groups of 10 female Sprague-Dawley rats were administered aluminum nitrate
nonahydrate in sugar-containing drinking water at doses of 360, 720 and 3600 mg/kg-day (26, 52
12
-------
10-23-2006
and 259 mg Al/kg bw-day, respectively) for 100 days (Domingo et al., 1987). A control group
received sugar-containing distilled water only. Sugar had been added to the drinking water of all
groups to reduce the taste-aversive effects of Al. The level of Al in the diet was not reported.
Animals were housed in metabolic cages to facilitate the collection of fecal and urine samples.
Food and water consumption were measured daily, body weights were noted weekly and blood
samples were taken at monthly intervals and at termination to monitor clinical chemistry and
hematological parameters. At termination, all animals were necropsied, and the weights of major
organs (brain, heart, lungs, kidneys, liver and spleen) were monitored. Aluminum concentrations
were measured in various tissues, pieces of which were processed for histopathological
examination. A significant decrease (p<0.05) in body weight gain was observed in the 259 mg
Al/kg-day group, attributed by the authors to decreased food intake. Overall, no consistent
variations in hematological (hemoglobin, hematocrit) or clinical chemistry (SGOT, SGPT,
alkaline phosphatase, urea, creatinine, total protein, cholesterol, glucose) parameters were
observed. No histopathological alterations in the heart, liver, kidney, spleen, brain and
cerebellum were observed. Interpretation of these data was complicated by the concurrent
exposure of the rats to high doses of nitrate of up to 475 times the RfD for nitrate (1.6 mg nitrate-
nitrogen/kg-day) which is based on methemoglobinemia in humans (U.S. EPA, 1999).
Therefore, because of nitrate co-exposure, the absence from the study design of a food-restricted
control group and uncertainty surrounding the contribution of Al in food, the apparent effect of
Al on body weight gain cannot be conclusively attributed to Al alone.
Some recent studies have identified a number of potential toxicological responses in
laboratory animals exposed orally to Al compounds in a subchronic or chronic dosing regimen.
In most cases, however, only one dose level was employed in the study compared to controls,
and since the amount of Al in the diet was not given, the resulting dose level represents an
incremental dose of Al compared to that of controls as baseline. However, while these studies
may offer inadequate quantitative dosimetric information for NOAEL/LOAEL identification and
consequent RfD development, they provide an qualitative indication of a range of potential
toxicological responses that might be induced in humans exposed to the element. For example,
Garbossa et al. (1998) studied the potential for water-soluble Al to affect the erythropoietic
integrity of late erythroid progenitor cells in the bone marrow. Three groups of five male Wistar
rats/group were either (1) gavaged with citrate at a dose of 1.0 [j.m Al/g-day (27 mg/kg-day), 5
days/week, for 15 weeks, (2) had drinking water containing 100 mmol Al/L made available to
them as the citrate for the same length of time or (3) maintained as controls. As calculated by the
authors, the dose associated with the applied concentration of Al in drinking water approximated
to 14-17 [j,mol/g-day (420 mg/kg-day). Rats had access to a standard chow diet, though with no
indication of the baseline concentration of Al provided therein. At the end of the in-life phase of
the study, all rats were sacrificed, and samples of blood were obtained for hematological
investigation. Femoral bone marrow cells were flushed with physiological medium, stimulated
with recombinant human erythropoietin, then monitored for the comparative incidence of
colony-forming units-erythroid (CFU-E). Further tests were carried out to monitor the osmotic
fragility and average life-span of erythrocytes from each test group. The animals in the group
receiving Al at the higher dose showed decreased hematocrit, hemoglobin concentration, median
osmotic fragility and erythrocyte life-span values compared to controls. The content of Al
increased in the serum and bone of both exposed groups, the distribution of concentrations in
bone correlating inversely with the extent of an animal's CFU-E development.
13
-------
10-23-2006
That A1 in drinking water may have the ability to cause histopathological changes and
altered hepatic enzyme activities was suggested by Basu et al. (1997) who made available
aluminum chloride in drinking water to groups of eight male Sprague-Dawley rats at a dose of 50
mg/kg-day (10.1 mg Al/kg-day) for 40 days. Additionally, other groups of similarly-treated rats
received drinking water containing either 0, 50, 100, 200 or 400 ppm (mg/L) added calcium
(Ca), as the chloride. The authors reported increased specific activities of acid and alkaline
phosphatases in liver 10,000 x g supernatants from Al-receiving animals versus controls, and in
alkaline phosphatase activity in equivalent kidney preparations. The presence of Ca in the
drinking water appeared to reverse these changes, plus the accompanying histopathological
features associated with them.
Konishi et al. (1996) examined the ability of Al and Ca to cause opposite and potentially
harmful effects in laboratory animals, in relation to the well-documented association between Al
and the onset of osteomalacia. Male STD Wistar rats were divided into four groups (n=4),
receiving either (1) a normal diet (Group I), (2) a normal diet supplemented with Al (Group II),
(3) a Ca-deficient diet (Group III) or (4) a Ca-deficient diet with supplemental Al (Group IV), for
10 weeks. Blood samples were taken at termination, and then animals were perfused with
paraformaldehyde/glutaraldehyde fixative. Levels of Ca, iron (Fe) and Al in serum and bone
were measured by atomic absorption spectrophotometry, and sections of the resected right tibia
were prepared for histopathological examination after decalcification in 5% formic acid in 10%
formalin.
There were statistically-significant changes in body weight gain when those of groups 3
and 4 were compared to animals from groups 1 and 2, the values for the latter groups remaining
constant from about 4 weeks of dosing. In discussing their histopathological findings, the
authors described no decrease in the thickness of cortical bone in Group II compared to control,
while bone specimen from Groups III and IV showed "an increase in osteoid as well as
osteoblasts and osteoclasts", in addition to other disturbances of ossification. Such effects were
considered to suggest bone fragility, with changes being more marked in Group IV compared to
III. The amount of Al in the tibia of exposed rats was significantly greater in Group II than in
Group I, whereas the average levels in Groups III and IV showed a further increase in Al
deposition, most notably in group IV. There were also differences among the groups in the
concentration of Fe in bone (tibia), and in the concentrations of Al, Ca, Fe and the levels of
parathyroid hormone in blood. The authors concluded that Ca deficiency appeared to potentiate
the deposition of orally administered Al in bone, and the attendant inhibition of ossification.
Iron deposition was also thought to play a role in the osteogenic disturbance, where Ca is
deficient.
A histopathological investigation indicated profound changes in the cerebrovascular and
neuronal integrity when male Long-Evans rats (n=9) were exposed for 52 weeks to 0.5 ppm
aluminum fluoride in drinking water (Varner et al., 1998). This corresponded to an Al dose of
0.019 mg/kg-day, based on a default drinking water consumption of 0.057 L/day, and a default
body weight of 0.472 kg for male Long-Evans rats (U.S. EPA, 1988). Duel control groups
received either NaF (fluoride controls) or double distilled deionized water. Tissue levels of Al
were measured in brain, liver and kidney by the use of a direct current plasma technique.
14
-------
10-23-2006
Animals receiving aluminum fluoride showed poor survival compared to the other groups, with
6/9 having died by week 48. The tissue concentrations of A1 were increased in the brain and
kidney compared to both the control groups, with Al-fluorescence being used to demonstrate that
A1 deposition was mostly in the vasculature. Morphological and histopathological changes due
to treatment were apparent in the liver, kidney and spleen. Some changes in neuronal integrity
were also evident in the hippocampus and neocortex. Other cytological changes in the brain
were associated with chromatid clumping, pyknosis and vacuolation.
A report by Somova et al. (1997) describes a study in which 10 male Wistar rats/group
received either 0, 5 or 20 mg/kg-day aluminum chloride by gavage in water for 6 months. At
termination, all animals were exsanguinated, then subjected to a necropsy in which excised
pieces of liver, kidney and cardiac and skeletal muscle were taken for histopathological
examination. Pieces of brain were examined by electron as well as light microscopy, and all
tissues were monitored for Al concentration by atomic absorption spectrophotometry. As
tabulated by the authors, Al in plasma and all of the listed tissues was dose-dependently
increased to levels that were statistically significantly greater than controls. However, though
described in qualitative terms and illustrated photographically, the Al-induced lesions did not
receive a quantitative treatment in the report. Thus, while at least some of the low dose rats
displayed NFD (neuro fibrillar degeneration) of the hippocampal region of the brain, insufficient
data are provided in the report to apply this observation to the identification of a NOAEL or
LOAEL.
Dietary experiments
Six Beagle dogs/sex/group were fed a diet providing either, in males, 0, 118, 317 or 1034
mg/kg-day sodium aluminum phosphate (0, 3.4, 9.0 or 29.4 mg Al/kg-day, respectively) or, in
females, 0, 112, 361 or 1087 mg/kg-day sodium aluminum phosphate (0, 3.2, 10.3 or 30.9 mg
Al/kg bw-day, respectively), for 6 months (Katz et al., 1984). No information was available on
the level of Al in the diet, and no compound-related effects on body weight gain, hematological
and clinical chemistry parameters (parameters not specified) or histopathological endpoints
(major organs and tissues examined) were observed. A highest NOEL of 30.9 mg Al/kg-day
could be tentatively identified in this study, but this would not include the contribution of Al
from the basal diet, nor reflect the identification of any toxicological effects, since the NOEL
occurred at the upper limit of the dose-response curve.
Neurotoxicity
A number of studies have been reported in which neurotoxicological/neurobehavioral
effects have been explicitly evaluated. In others, the effects of Al on neurological developmental
have been addressed. For example, Golub et al. (1989) fed diets containing Al as the lactate at
25 (controls), 500 or 1000 mg Al/kg diet (3.3, 65 or 130 mg Al/kg-day) to groups of 15 female
Swiss-Webster mice for 6 weeks (Golub et al., 1989). No mice were exposed to lactate alone.
While no statistically significant differences in food intake or body weight gain were observed,
mice fed the highest Al concentration gained less weight than the controls or low-dose group.
As reported by the authors, a significant decrease (20%) in spontaneous motor activity (i.e., total,
vertical and horizontal movement) was observed in the 130 mg Al/kg-day group. Activity in the
15
-------
10-23-2006
65 mg Al/kg-day group was not significantly different than the controls. Thus, the highest
NOAEL is 65 mg Al/kg-day and the LOAEL is 130 mg Al/kg-day.
Neurobehavioral effects of aluminum lactate were evaluated in groups of 12 female
N:NIH Swiss-Webster mice (4.5-5.5 weeks old) that were fed 25 (controls) or 1000 mg Al/g diet
for 90 days (Golub et al., 1992a). Based on a food factor of 0.19 kg diet/kg body weight/day
calculated using an algorithm relating food consumption to body weight (U.S. EPA, 1988) and
reported body weight data (the time-weighted average weight is 25.4 g), the dosage in the treated
mice is estimated to be 190 mg Al/kg bw-day. No mice were exposed to lactate alone. A
neurobehavioral test battery used by Donald et al. (1989) was administered at the beginning of
the experiment (day 0) and after 45 and 90 (±3) days, with motor activity evaluated at the latter
two time points. Aluminum levels were measured in brain, femur and liver at the end of the
exposure period.
Body weight was significantly increased in the treated mice but no exposure-related
changes in food intake or overt signs of neurotoxicity were observed. Results of the
neurobehavioral tests showed significantly decreased hindlimb grip strength at 90 days,
decreased air puff startle response at 90 days and decreased auditory startle response at 45 days
in the treated mice. Spontaneous motor activity was reduced at 90 days as indicated by
decreased total activity counts, horizontal activity counts and percentage of intervals with high
activity counts. Aluminum concentrations in the brain and liver were increased approximately
3-fold in the treated mice, but brain and liver lipid peroxidation indices were not altered.
Male Wistar rats (6-8 per group) were exposed continuously for 6 months to food
containing 1.52 mg Al/kg (normal diet) or 1000 mg Al/kg as aluminum chloride with citrate
(Florence et al., 1994). The average daily Al intake was estimated to be 0.13 or 84 mg Al/kg
bw-day, assuming a body weight of 0.305 kg (arithmetic mean of default mature weight of male
Wistar rats and the starting weight in this study of 0.11 kg) and a food intake of 0.026 kg food/kg
bw-day, calculated using an algorithm relating food intake to body weight (U.S. EPA, 1988).
The citrate content of the diet was in a 1:1 stoichiometric proportion to Al, therefore, the
estimated daily intake was 598 mg/kg-day. Rats exposed to Al developed histopathological
abnormalities in brain tissue, not specific to any brain region, characterized by extensive
cytoplasmic vacuolization in astrocytes, swelling of astrocytic processes, particularly of astrocyte
end-feet abutting blood vessels. Neurons also exhibited vacuolization and nuclear inclusions.
Although no specific behavioral assays were reported, the investigators noted that "no significant
behavioral changes were observed". Accordingly, the functional significance of the
histopathological lesions is uncertain. The lesions appear to differ from the NFD observed with
parenteral Al exposures (Kowall et al., 1989; Wakayama et al., 1993); or from exposures to Al in
combination with calcium deprivation (Garruto et al., 1989; Kihira et al., 1995; Mitani, 1992).
The LOAEL for histopathological changes in the brain was 84 mg Al/kg-day.
Male Sprague-Dawley rats (40 per group) were exposed in drinking water to 0, 50 or 100
mg Al/kg bw-day as aluminum nitrate with citric acid for 6.5 months beginning at 21 days of
age, 8 months of age or 16 months of age (Domingo et al., 1996). The citric acid dosage was
355 or 710 mg/kg-day in the 50 or 100 mg Al/kg bw-day groups, respectively. Controls did not
receive citric acid. Dietary Al intake was not reported; the rats were maintained on Panlab rat
16
-------
10-23-2006
chow. Animals from control and exposed groups were subjected to a number of neurobehavioral
tests, and at termination, A1 levels were measured in various excised regions of the brain. The
authors observed the highest A1 levels in the olfactory bulb and rhachidical bulb, while the cortex
and thalamus were the regions showing the lowest A1 content. However, compared to controls,
there were no significant effects (p>0.05) of A1 (with citric acid) on spontaneous motor activity
(open-field) or passive avoidance operant training or performance (grid floor shock, light/dark
shuttle box). Thus, the NOAEL was 100 mg Al/kg-day with citric acid; although this does not
include the A1 contribution from food. This study is listed on Table 1 because the NOAEL,
although probably underestimated because of unreported A1 intake from food, is still lower than
the LOAELs from other studies.
Groups of six male albino rats were administered 0 or 25 mg Al/kg bw-day as aluminum
nitrate in normal saline by gavage, 10% ethanol in drinking water, or 25 mg Al/kg bw-day by
gavage combined with 10% ethanol in drinking water, 6 days/week for 6 weeks (Flora et al.,
1991). The level of Al in the diet was not reported. Urinary A-aminolevulinic acid (ALA),
blood ALA-dehydratase (ALAD), blood zinc protoporphyrin (ZPP), glutamic oxaloacetic
transaminase (GOT) and glutamic pyruvic transaminase (GPT) in serum and liver and brain
biogenic amines and their metabolites [dopamine (DA), norepinephrine (NE),
5-hydroxytryptamine (5-HT), homovanillic acid (HVA) and 5-hydroxyindolacetic acid
(5-HIAA)] were evaluated at the end of the treatment period. Treatment with Al alone caused
significantly increased blood ALAD (p<0.01), decreased liver GPT (p<0.05), decreased brain
DA (p<0.01), increased brain NE (p<0.05) and decreased brain 5-HT (p<0.05). Compared to
treatment with Al alone, concurrent exposure to ethanol and Al produced significantly decreased
ALAD, increased ALA, increased ZPP, increased liver GPT, increased serum GOT and
increased brain HVA. Significant changes found only in the combined Al and ethanol group
included increased serum GPT, increased brain NE and decreased brain 5-HT. Treatment with
ethanol alone only inhibited blood ALAD. The rats were co-exposed to relatively high levels of
nitrate [comparable to those in the Domingo et al. (1987) subchronic study], but it seems likely
that some of the changes (i.e., effects on brain chemicals) are related to aluminum which is
known to be neurotoxic. Because the toxicological significance of the changes is unclear due to
lack of evaluation of neurobehavioral performance and other endpoints, there is uncertainty
whether the 25 mg Al/kg-day dose is a NOAEL or a LOAEL, an uncertainty compounded by the
absence of information about the level of Al in the basal diet.
Reproductive/developmental toxicity
A number of studies have been carried out to examine the effects of Al compounds on
developmental toxicity, particularly their effects on postnatal neurobehavioral development. For
example, Bernuzzi et al. (1989) exposed groups of 6-12 pregnant Wistar rats to aluminum
chloride or aluminum lactate in the diet on gestational days 1 through 21. The rats received
nominal daily doses of 0, 100, 300, 400 mg Al/kg as aluminum chloride or 0, 100, 200 or 400 mg
Al/kg as aluminum lactate. No rats were exposed to lactate alone, and information regarding
level of Al in the basal diet was not reported. On the average, there was a less than 10%
decrease in maternal body weight gain and no effect on food or water intake. No significant
difference in litter size was observed. However, postnatal mortality increased 55% and 26% in
offspring of the rats exposed to 300 or 400 mg Al/kg-day, respectively. The offspring of dams
17
-------
10-23-2006
fed >300 mg Al/kg-day weighed significantly less than controls on postnatal day 1. Decreased
body weight was also observed on postnatal days (PD) 4 and 14 in the offspring of rats fed 400
mg Al/kg-day as aluminum lactate. The following tests were used to assess neuromotor
development (maturation): righting reflex, grasping reflex, negative geotaxis, suspension test
and locomotor coordination. The tests were performed on PDs 4, 6, 9, 12 and 20, respectively.
Impairment of neuromotor development (righting and grasping reflexes) was observed in the
pups exposed to >200 mg Al/kg-day. Impaired grasping reflex was also observed in the 100
mg/kg-day aluminum lactate group. Offspring of rats fed 400 mg/kg-day also exhibited altered
performance on the locomotor coordination test.
A follow-up study by the same research group found that ingestion of 400 mg Al/kg bw-
day as aluminum lactate had no effect on postnatal mortality, body weight and righting and
grasping reflex tests (Muller et al., 1990), although significant differences between control and
exposure groups were noted in locomotor coordination and operant conditioning tests.
Significant differences between controls and exposed groups in the negative geotaxis test were
limited to those pups of dams treated during the second and third weeks of gestation, a finding
interpreted by the authors to indicate the possibility of long-term effects on the central nervous
system of trans-placenta exposure to Al during a later organogenic phase. According to Muller
et al. (1990), the contradictions between this and their earlier study (Bernuzzi et al., 1989) could
be related to environmental modifications. In particular, the mothers and pups were much more
protected in the Muller et al. (1990) study than in the previous one because they were housed in
plastic cages instead of wire mesh cages and received cotton to build nests. Body temperature of
the pups, therefore, may have been more adequately maintained in the Muller et al. (1990) study.
As discussed in this study, toxicity in pups can be confounded by insufficient body temperature,
and delayed pup weight gain could explain the differences in neuromotor performance.
Muller et al. (1990) administered diets supplemented with 0 or 400 mg Al/kg bw-day as
aluminum lactate to groups of 6-9 pregnant Wistar rats on days 1-7, 1-14 or 1-21 of gestation.
No rats were exposed to lactate alone, and information regarding level of Al in the basal diet was
not reported. Neuromotor development was assessed on postnatal days 4, 6, 9, 12 and 20 using
tests of righting reflex, grasping reflex, negative geotaxis, suspension and locomotor
coordination, respectively. Learning ability was also tested on PD 65 using operant
conditioning. No effects on maternal body weight or food intake were observed in dams exposed
on gestational days 1-7 or 1-14. In the dams exposed on gestational days (GD) 1-21, a
significant decrease in maternal body weight (26 and 35%, respectively) was observed on days
16 and 19 of gestation. Decreased food intake was also observed on day 19 of gestation. No
effects on litter size, postnatal mortality or postnatal body weight were observed. Impairment of
neuromotor development (p<0.05) was observed in two of the five tests (negative geotaxis and
locomotor coordination); no differences between the three treated groups were observed. For the
operant conditioning test, there were significant differences (p<0.05) between the treated and
control young rats. No differences between the three treated groups were observed. The
LOAEL for developmental toxicity is 400 mg Al/kg-day, but this does not include the
contribution of Al from the basal diet.
Groups of 10 pregnant Sprague Dawley rats were administered 180, 360 or 720 mg/kg-
day aluminum nitrate nonahydrate by gavage (13, 26, 52 mg Al/kg bw-day) on GDs 6-14
18
-------
10-23-2006
(Paternain et al., 1988). A vehicle (water) only control group was used. The level of A1 in the
diet was not reported. Aluminum exposed dams gained significantly less weight than the
controls. No significant effects on the numbers of litters, corpora lutea, total implants, live
fetuses, resorptions or runt fetuses were observed. Significant decreases in fetal body weight and
tail length were observed at all three Al doses; decreased fetal body length was also observed at
the 52 mg Al/kg-day dose level. No dose-related external or visceral malformations were
observed in the offspring. However, a significant increase in the incidence of skeletal
malformations (delayed ossification, hypoplastic deformed ribs) was observed at all three
treatment levels. In addition, the incidence of hematomas was significantly increased at the high
dose. Because the rats were co-exposed to relatively high levels of nitrate [comparable to those
in the Domingo et al. (1987) subchronic study], the effects of treatment cannot be conclusively
attributed to Al alone, in the absence of a nitrate-exposed control group.
By contrast to the striking findings of potentially teratogenic effects of aluminum nitrate
in Sprague-Dawley rats, as described above (Paternain et al., 1988), equivalent experiments by
Domingo et al. (1989) in Swiss mice did not reveal any reproductive, developmental or
teratogenic effects of Al, when administered to dams as the hydroxide. Domingo et al. (1989)
administered by gavage 0, 66.5, 133 or 266 mg/kg-day aluminum hydroxide (0, 23.9, 47.8 or
95.5 mg Al/kg bw-day) to groups of 20 pregnant Swiss mice on GD 6-15. The level of Al in the
diet was not reported. The dams were killed on GD 18. No compound-related effects were
observed on maternal mortality, clinical signs, body weight, food intake or absolute or relative
heart, lung, spleen, liver, kidney and brain weights. In addition, no compound-related effects
were observed on numbers of implantations, resorptions, live and dead fetuses, sex ratio and the
incidences of external malformations, internal soft-tissue defects or skeletal abnormalities.
Therefore, this study identifies a NOEL of 95.5 mg Al/kg-day by default for reproductive,
developmental and teratogenic toxicity in mice. However, neuromotor development was not
assessed and the contribution of Al from the basal diet was not stated in the report.
A number of studies have been designed to evaluate the influence of citrate or lactate on
the potential developmental toxicity of Al. For example, Gomez et al. (1991) exposed groups of
15-19 pregnant Sprague-Dawley rats to either distilled water (controls) or 133 mg Al/kg bw-day
in the form of either aluminum hydroxide (384 mg/kg-day), aluminum citrate (1064 mg/kg-day)
or aluminum hydroxide (384 mg/kg-day) concurrent with citric acid (62 mg/kg-day) by gavage
on GD 6-15. The level of Al in the diet was not reported and no rats were exposed to citric acid
alone. Terminations were performed on GD 20. Maternal and fetal evaluations showed
exposure-related effects only in the group exposed to aluminum hydroxide and citric acid
concurrently. Significant changes included reduced maternal body weight gain on GDs 6-20 (but
not at sacrifice on day 20), reduced fetal body weight and some skeletal variations (increased
delayed occipital and sternebrae ossification and increased absence of xiphoides). No effects
were seen on maternal food consumption or clinical signs, maternal absolute or relative liver,
kidney or brain weights, gravid uterine weight, corpora lutea/dam, implantations/litter, pre- or
postimplantation loss/litter, viable or nonviable implants/litter, fetal sex ratio or fetal
malformations (external, visceral or skeletal). This study identified a stand alone minimum
LOAEL of 133 mg Al/kg-day for non-neurobehavioral developmental toxicity of aluminum
hydroxide and aluminum citrate in rats. Although confidence in this LOAEL is low (because
aluminum hydroxide administered concurrently with citric acid induced did developmental
19
-------
10-23-2006
effects and because the dose does not include a contribution of A1 from the basal diet) the value
is consistent with the developmental NOAEL of 95.5 mg Al/kg-day for aluminum hydroxide in
mice (Domingo et al., 1989).
In a similar experimental protocol, groups of 11-13 pregnant female Swiss albino (CD-I)
mice were administered 57.5 mg Al/kg bw-day as either aluminum hydroxide (166 mg/kg-day),
aluminum lactate (627 mg/kg-day) or aluminum hydroxide (166 mg/kg-day) concurrent with
lactic acid (570 mg/kg-day) by gavage on gestation days 6-15 (Colomina et al., 1992). Other
groups were treated with lactic acid alone (570 mg/kg-day, equivalent to the amount in 627
mg/kg of aluminum lactate) or distilled water (controls). The level of Al in the diet was not
reported. Fetal evaluations were performed on GD 18, including examinations for skeletal and
visceral abnormalities in approximately two-thirds and one-third of the pups, respectively. The
investigators noted that the dose of Al (57.5 mg/kg-day) is equivalent to ingestion of 3.5 g
Al/day by a 60 kg person, which is higher than the usual quantities of Al ingested therapeutically
for peptic disorders. Maternal body weight gain was significantly lower than control values in
the aluminum lactate-treated mice when evaluated over GDs 6-9 (92%), 6-12 (55.6%) and 0-18
(38.5%>) and in the mice treated with combined aluminum hydroxide and lactic acid evaluated
over GDs 6-12 (37.8%>), 6-15 (42.7%) and 0-18 (15.7%). The decreased maternal weight gain in
the aluminum lactate group was accompanied by significantly reduced food consumption during
gestation days 6-18. Significant developmental and/or teratological effects in the aluminum
lactate group included 16% reduced fetal body weight (p<0.01) and increased incidences of cleft
palate (13.2%, p<0.05), dorsal hyperkyphosis (i.e., excessive flexion of spine) (13.5%, p<0.05)
and delayed parietal ossification (15.4%, p<0.01). These developmental effects were not
observed in any of the control or aluminum hydroxide exposed pups, and the only other
significant changes in the other groups were decreased maternal relative liver weight and delayed
fetal parietal ossification in the lactic acid only exposure group. Other types of internal or
skeletal malformations or variations were not found in any of the fetuses. Additionally, no
effects were seen on maternal absolute or relative kidney weight, gravid uterine weight, numbers
of implantation sites/litter, live or dead fetuses, resorptions, postimplantation loss/litter, litters
with dead fetuses or fetal sex ratio in any of the groups. By analogy to the findings of the
Domingo et al. (1989) and Gomez et al. (1991) studies, the lack of developmental effects of
aluminum hydroxide at the tested dose could be related to low solubility and absorption.
In a more recent study, pregnant Swiss mice were administered gavage doses of 0 or 104
mg Al/kg bw-day as aluminum hydroxide on days 6-15 of gestation (Colomina et al., 1994).
Dietary Al intake was not reported; the mice were maintained on Panlab rodent chow.
Compared to controls, there were no effects (p>0.05) of Al on maternal body or organ weight,
number of implantations per litter, number of resorptions per litter, number of dead fetuses per
litter, percentage of positive post-implantation loss, sex ratio or fetal body weight per litter.
Gross external, visceral or skeletal examination of fetuses revealed no abnormalities or
developmental variations. Thus, the NOAEL for development effects from this study is 104 mg
Al/kg-day, however, this does not include the Al contribution from food. Thus, based on this
study and the previous study (Colomina et al., 1992), aluminum lactate appears to be more potent
as a developmental toxicant in mice than the less water soluble aluminum hydroxide.
20
-------
10-23-2006
Groups of 16 pregnant Swiss-Webster mice were fed 25 (control group), 500 or 1000 mg
Al/kg diet as aluminum lactate throughout gestation and lactation (Donald et al., 1989). The
control diet was fed to pups that were selected for post-weaning neurobehavioral assessment.
Reported maternal doses were 5, 100 and 200 mg Al/kg bw-day at the beginning of pregnancy
and 10.5, 210 and 420 mg Al/kg bw-day near the end of lactation. No mice were exposed to
lactate alone. There were no treatment-related changes in maternal survival, body weight
(measured on GD 0 and 16 and PDs 0, 5, 10, 15 and 20), food intake, toxic signs or
neurobehavior (evaluated after pups were weaned at PD 21 using the same test battery used for
the pups and described below), or on litter size or postnatal growth and development in pups as
assessed by body weight, toxic signs on PDs 0-55, and by crown-rump length on PDs 0 and 20.
Neurobehavioral maturation was tested in two pups per litter on PDs 8-18 with a 12-item test
battery (fore- and hindlimb grasp, fore- and hindpaw placement on sticks of 2 widths, vibrissa
placing, visual placing, auditory and air puff startle, eye opening and screen grasp, cling and
climb). A neurobehavioral test battery was administered to six pups per litter at age 25 days (4
days postweaning) or 39 days (fore- and hindlimb grip strengths, temperature sensitivity of tail,
negative geotaxis, startle reflex to air puff and auditory stimuli) or age 21 and 35 days (foot
splay). The pre-weaning neurobehavioral testing showed that a significant (p=0.007) number of
pups in the high dose group had impaired vertical screen climb performance. The postweaning
neurobehavioral assessment showed significantly (p<0.05) altered performance on several tests.
These included decreased forelimb grip strength at age 39 days in the low dose group, increased
hindlimb grip strength at age 25 days in both low and high dose groups, increased foot splay
distance at age 21 days in both low and high dose groups and at age 35 days in the low dose
group, and increased forelimb grip strength at age 25 days and decreased thermal sensitivity at
age 25 and 39 days in the high dose group. There were no treatment-related changes in
concentrations of Al in pup liver or bone (brain tissue was not analyzed).
In a more recent study of similar design by the same group of investigators, groups of 14
and 9 female Swiss Webster mice (6-8 weeks old) were fed 25 (control) or 1000 mg Al/g diet as
aluminum lactate, respectively, during gestation and lactation (Golub et al., 1992b). The 1000
mg/g concentration was selected based on the demonstration of neurobehavioral effects in
weanlings at this level (Donald et al., 1989). No mice were exposed to lactate alone. Using food
intake and body weight values estimated from reported data, maternal doses are estimated to be
approximately 4.3 and 174 mg Al/kg bw-day at the beginning of gestation and 4.8 and 607 at the
end of the lactation period. At birth, litters were fostered either within or between groups to
provide four groups of offspring that were exposed to excess Al via maternal diet during
gestation, lactation, both or neither (i.e., 25 ppm during gestation and lactation, 1000 ppm during
gestation and 25 ppm during lactation, 25 ppm during gestation and 1000 ppm during lactation,
and 1000 ppm during gestation and lactation). Maternal effects included significantly (p<0.015)
reduced (10-12%) body weight gain and food intake in the treated group during late pregnancy
and lactation, and signs of neurotoxicity (hindlimb splaying and dragging) in one treated dam at
postnatal day 21 (weaning); this dam had seizures and died 4 days later. No treatment-related
effects on litter size, birth weight, crown-rump length, righting ability at birth, sex ratio or
postnatal survival were observed. Both gestation-only and lactation-only exposure caused
significantly (p<0.05) decreased body weight gain in the treated pups beginning on postnatal day
10; combined gestation and lactation exposure produced the greatest decrease (approximately
24% at weaning). Neurobehavioral testing using the same battery as Donald et al. (1989) was
21
-------
10-23-2006
performed at weaning on the dams and on a total of 12, 16, 12 and 6 pups (1 male and 1 female
pup per litter) from the control, gestation-only, lactation-only and combined gestation and
lactation groups, respectively. Results of this testing showed effects only in pups, including
significantly decreased forelimb grip strength after gestation-only exposure, increased hindlimb
grip strength after both gestation and lactation exposure, decreased temperature sensitivity after
lactation-only exposure, and longer negative geotaxis latency after lactation-only exposure. In
general, the findings of this study are consistent with those of Donald et al. (1989) in showing
neurodevelopmental effects at the 1000 mg/kg dietary concentration, although intake dosages are
dissimilar at the end of lactation. Using the dosage at the beginning of gestation, this study
defines a LOAEL of 174 mg/kg-day for developmental effects.
The Donald et al. (1989) study differs from that of Golub et al. (1992b) in that offspring
were not fostered, were tested at a later age (25 vs. 21 days), were allowed 4 days of recovery
from the treated diet prior to testing, participated in other behavioral tests currently, and
experienced no growth retardation. The effects found only in the cross-fostered groups in the
Golub et al. (1992b) study (lower forelimb strength after gestation exposure and altered negative
geotaxis latencies after lactation only exposure) were not observed by Donald et al. (1989).
Increased footsplay was observed by Donald et al. (1989) but not by Golub et al. (1992b),
perhaps due to an opposing effect of smaller pup body size in this study. Neither gestation or
lactation exposure affected pup brain or liver Al concentrations, but lactation exposure caused
significantly lower manganese and iron concentrations in liver and manganese concentrations in
brain.
In a further extension of the two previous studies (Donald et al., 1989; Golub et al.,
1992b), pregnant female Swiss-Webster mice were exposed continuously to a semi-purified diet
containing 7 (control), 500 or 1000 mg Al/kg from the time of conception, through pregnancy
and lactation (Golub et al., 1995). At weaning, pups were exposed to the same Al diet as their
mothers (500 or 1000 mg Al/kg) until they were 150-170 days of age or were switched to the
control diet (7 mg Al/kg) for the same time period. Based on reported dosages in previous
studies by the same investigators, estimated daily dosages for mice exposed to 1000 mg Al/kg
diet were as follows: 200 mg/kg bw-day in pregnant mice, 420 mg/kg-day in lactating mice and
130 mg/kg-day in offspring (Golub et al., 1994); doses for the mice exposed to 500 mg Al/kg
diet were assumed to be approximately half of that of mice fed 1000 mg Al/kg, or 100 mg/kg-
day in pregnant mice, 210 mg/kg-day in lactating mice and 65 mg/kg-day in offspring.
Compared to the control diet, the Al diet had no effect on dam weight, gestation length, litter
size, pup weight, offspring growth or organ weights. Operant conditioning (nose poke) of
offspring for delayed spatial alternation or discrimination reversal tasks was initiated at 50 days
of age and continued 5 days/week for a total of 35 sessions. A neurobehavioral test battery was
conducted when the offspring were 150-170 days of age (forelimb and hindlimb grip strength,
temperature sensitivity, negative geotaxis, air puff and auditory startle response). Maternal and
pre-weaning exposure to 500 mg Al/kg significantly affected (p<0.05) operant training in the
offspring, but not performance after training in delayed spatial alternation or discrimination
reversal tasks (i.e., decreased number of training sessions to achieve the training criteria). This
exposure also significantly decreased forelimb and hindlimb grip strength and puff startle
response (p<0.05). Pre-weaning and combined pre- and post-weaning exposure to 1000 mg
Al/kg significantly increased (p<0.05) incidence of cagemate aggression at the time behavioral
22
-------
10-23-2006
testing. No effects were observed on auditory startle response, temperature sensitivity or
negative geotaxis in offspring. Histopathological examination of the brain and spinal cord
revealed no treatment-related changes. Thus, the LOAEL for combined maternal and pre-
weaning exposure on neurobehavioral effects in mice would approximate to 100 mg Al/kg-day
(estimated daily maternal dosage).
Pregnant Charles River CD rats were administered gavage doses of 0, 250, 500 or 1000
mg Al/kg bw-day ("experiment A") or 0, 5, 25, 50, 250 or 500 mg Al/kg bw-day ("experiment
B") as aluminum lactate in distilled water on GDs 5-15 (Agarwal et al., 1996). Dietary A1 intake
was not reported. Offspring were examined for body weight, anogenital distance, oestrus cycle
regularity (after puberty), duration of pseudopregnancy induced by mechanical stimulation of the
cervix, oocyte production induced by an injection of human chorionic gonadotropin, and male
and female gonad weights. Aluminum had no effect on litter size and no consistent effects on
birth weight were observed. For example, birth weights were decreased in male offspring from
dams that received 250 mg Al/kg-day, but not at higher dosages, and the effect was observed
only in experiment A. Female offspring birth weights decreased at certain dosage levels in
experiment A and increased at these same dosage levels in experiment B. Similar
inconsistencies between experiment A and B were observed for gonadal weights, anogenital
distance, time to puberty (vaginal opening), duration of pseudopregnancy or numbers of
superovulated oocytes. A significantly increased (p<0.05) number of abnormal oestrus cycle
lengths (defined as less than 4 days or greater than 5 days) occurred in offspring from dams that
received 250 mg Al/kg-day (in experiment A, the endpoint was not measured in experiment B).
However, the effect was most pronounced in the first three oestrus cycles (of five observed) and
not detected by the 5th cycle. Thus, the NOAEL for temporary disturbance of the oestrus cycle
in offspring of dams administered Al is 250 mg Al/kg-day. NOAELs for all other reproductive
endpoints in this study were 1000 mg Al/kg-day. These NOAELs do not include the
contribution of Al in food.
In a three-generation study, Ondreicka et al. (1966) exposed initial groups of seven
female and three male Dobra Voda mice to either 0 or 19.3 mg Al/kg bw-day as aluminum
chloride in drinking water. The diet also contained 160 to 180 ppm Al, giving an estimated
intake of 27-31 mg/kg-day based on default values for food consumption and body weight for
chronic exposure of mice (U.S. EPA, 1988). Using this estimate, the total Al intakes (drinking
water and food) were 27 mg/kg-day (controls) and 46.3 mg/kg-day (exposed group). The Po
group produced three litters (designated Fia, Fib and Fic) and the Fia group produced two litters
(designated F2a and F2b) from which the weanlings were exposed to Al in the drinking water
starting at 4 weeks of age. There was no difference in body weight gain among the groups in the
Po generation, a result that contrasted with the striking decrease in this parameter in the treated
Fib, Fic, F2a and F2b groups. Though no effects on erythrocyte count, hemoglobin levels or
histopathology of the liver, spleen and kidneys were observed in the Po, Fi or F2 generations at
the end of the study and no significant differences were seen in the number of litters or offspring
between the exposed and control groups, the study identified a LOAEL of 46.3 mg Al/kg-day,
based on the observed changes in body weight gain.
23
-------
10-23-2006
Other toxicological effects of aluminum
In a study designed to determine the effects of oral A1 exposure on susceptibility to
bacterial infection, female Swiss-Webster mice (13-14 per group) were exposed to a diet
containing 25 (control), 500 or 1000 mg Al/kg as aluminum lactate during pregnancy, through
lactation and for 10 days following weaning of the pups (Yoshida et al., 1989). Based on
reported dosages in previous studies by the same investigators, estimated daily dosages for mice
exposed to 1000 mg Al/kg diet are as follows: 200 mg/kg-day during pregnancy and 420 mg/kg-
day during lactation; doses for the mice exposed to 500 mg Al/kg diet are assumed to be
approximately half of that of mice fed 1000 mg Al/kg, or 100 mg/kg-day in pregnant mice and
210 mg/kg-day in lactating mice (Golub et al., 1994). At weaning, dams and pups were
inoculated with a tail vein injection of Listeria monocytogenes and monitored for mortality for
10 days. In a separate experiment, female mice, 6 weeks of age, were exposed to the same
dietary Al levels for 6 weeks and then inoculated with L. monocytogenes. Estimated Al dosages
were 5, 98 or 195 mg Al/kg bw-day for the 25, 500 or 1000 mg Al/kg dietary levels,
respectively, based on a default food factor of 0.195 kg diet/kg bw-day assuming a reference
"subchronic" food intake and body weight for female B6C3F1 mice over the period from
weaning to 90 days (U.S. EPA, 1988). Inoculation resulted in significantly greater (p<0.025)
mortality in dams exposed to 500 or 1000 mg Al/kg diet compared to controls. There were no
differences in mortality between the groups of inoculated pups or between groups of inoculated
adult mice exposed to Al for 6 weeks. The LOAEL for pregnant mice was 100 mg Al/kg bw-day
and the NOAEL for adult, non-pregnant mice was 195 mg Al/kg bw-day. Although the exposure
duration in this study was only 7 weeks, it is included in Table 1 because it provides the only
dose-response data on the effects of Al on resistance to pathogens.
Carcinogenicity studies
Schroeder and Mitchener (1975a) exposed 52 Long-Evans rats/sex/group to 0 or 5 ppm
Al as potassium aluminum sulfate in drinking water for life. Based on default values for
drinking water consumption and body weight for this strain of rat in a chronic study (U.S. EPA,
1988), these values are equivalent to Al doses of 0.472 and 0.67 mg/kg-day, for males and
females, respectively. Study endpoints included body and heart weight; serum glucose,
cholesterol and uric acid; and urinary protein, glucose and pH. All animals were necropsied at
the time of natural death, and histological examinations were carried out on heart, lung, kidney,
liver, spleen and gross tumors, for approximately 50% of the animals in the group. The only
remarkable finding was a significant increase (p<0.005) in gross tumor incidence in exposed
male rats [13/25 (52%) compared to 4/26 (15%) in controls], although the tumor sites were not
reported. Six of the tumors in the exposed males (46% of total) were considered malignant
compared to two malignant tumors (50% of total) in the male controls. There were no
significant differences in tumor incidences between exposed and control females.
In another study by the same investigators, 54 Swiss mice/sex/group were exposed to
drinking water containing 0 or 5 ppm Al as aluminum potassium sulfate for life (Schroeder and
Mitchener, 1975b). Based on default values for drinking water consumption and body weight for
B6C3F1 mice in a chronic study (U.S. EPA, 1988), these values approximate to Al doses of 1.2
mg/kg-day in both males and females. Study endpoints included body weight, gross pathology,
24
-------
10-23-2006
and some limited histology of the heart, lung, liver, kidney and spleen. The incidences of gross
tumors were 15/41 (36.6%) and 11/38 (28.9%) in exposed and control males, respectively, and
19/41 (46.3%>) and 14/47 (29.8%>) in exposed and control females, respectively, differences that
did not achieve statistical significance by Fisher's exact test, although incidences of multiple
tumors and lymphoma leukemia were considered by the authors to be significantly increased in
females (p<0.025 and p<0.05, respectively). However, a definitive assessment of aluminum
carcinogenicity in both this and the rat study (Schroeder and Mitchener, 1975a) is precluded by
the limitations of the pathology examinations and reporting.
In a more recent study, the tumorigenic potential of aluminum potassium sulfate was
assessed in B6C3F1 mice chronically exposed in the diet (Oneda et al., 1994). Sixty
animals/sex/group were fed a diet containing 0, 1.0, 2.5, 5.0 or 10.0%> (w/w) for 20 months.
These concentrations of aluminum potassium sulfate (as the dodecahydrate) are equivalent to 0,
569, 1422, 2844 and 5687 ppm Al. Using food factors calculated with an algorithm relating food
consumption to body weight (U.S. EPA, 1988) and body weight data estimated from growth
curves reported by the investigators, the dosages of aluminum are estimated to be 0, 95, 237, 483
or 1024 mg Al/kg-day in males and 0, 97, 242, 512 or 1110 mg Al/kg-day in females. Clinical
signs, food consumption, and body weight were evaluated weekly. Hematology, clinical
chemistry or urine endpoints were not assessed. Necropsies that included organ weight
measurements and comprehensive histological examinations (including brain) were performed
on all animals, including those that died during the course of the study. Survival rates were
higher than control values in all treated male and female groups, ranging from 86.7-95.0%)
compared to 73.3%> in males and 86.7-91.7%) compared to 78.3%> in females. No changes in
food consumption were observed, but body weight gain was increased in both sexes at 95-97 and
237-242 mg Al/kg-day (weights were 10-23%) higher than controls at end of study), was similar
to controls in both sexes at 483-512 mg Al/kg-day, and decreased in both sexes at 1024-1110 mg
Al/kg-day (11-16%> lower than controls at end of study). There were no exposure-related
increased incidences of tumors, other proliferative lesions or non-neoplastic lesions. In fact, the
incidence of spontaneous hepatocellular carcinomas was significantly decreased in males at 1024
mg Al/kg-day (5.5%> compared to 20.5%> in controls, p<0.01).
Inhalation Exposure
Groups of 20 weanling Fischer 344 rats/sex and 20 weanling Hartley guinea pigs/sex
were exposed to 0, 0.25, 2.5 or 25 mg/m3 aluminum chlorhydrate [A12(0H)5C1>x(H20)] for 6
hours/day, 5 days/week for 6 months (Steinhagen et al., 1978). Analysis of the aluminum
chlorhydrate by the investigators showed it to contain 24.5%> Al, indicating that the animals were
exposed to 0, 0.061, 0.61 and 6.1 mg Al/m3. Body weights were measured weekly for the first 8
weeks and biweekly thereafter. At the end of the exposure period, 10 animals (5/sex) of each
species were sacrificed for organ weight measurements (heart, lung, liver, kidney, spleen and
brain) and histological examination of the lungs, liver and kidney. In addition, comprehensive
histological examinations were performed on animals in the control and 6.1 mg AL/m3 groups.
The remainder of the animals was used for hematology evaluation (RBC, WBC, hematocrit and
hemoglobin) and Al measurements in blood and tissues. Apparent effects of Al included
multifocal granulomatous pneumonia in both species at >0.61 mg Al/m3, significantly increased
absolute and relative lung weights in both species, and decreased body weight gain in rats and
25
-------
10-23-2006
minimal lung edema in guinea pigs at 6.1 mg Al/m3. The granulomatous reaction was
characterized by foci of giant vacuoled particle-containing macrophages in the lungs and
macrophages that did not appear to contain vacuoles or other evidence of phagocytized material
in the peribronchial lymph nodes. There was a significant dose-related accumulation of A1 in the
lungs of both species at >0.061 mg Al/m3. However, a NOAEL of 0.061 mg/m3 could be
identified for the onset of compound-induced histopathological effects.
In other studies, groups of 14-30 guinea pigs, rats and hamsters were exposed to fine
metallic A1 powders (pyro, atomized and flaked) at concentrations of 15, 30, 50 or 100 mg
powder/m3 air for 6 hours/day, 5 days/week for 6 months (Gross et al., 1973). Alveolar
proteinosis occurred in exposed animals of all three species after 2 months of exposure, but
fibrosis or other pulmonary changes did not develop. Similarly, groups of 23 or 46 rats and 48
hamsters were exposed to undetermined concentrations of Al fumes or Al powder (20% Al, 80%
Al(OH)3) for morning hours only or morning and afternoon for up to 20 months (Christie et al.,
1963). Effects were similar for both forms of Al in both species, including initial increased
alveolar macrophage proliferation followed by nodular hyalinized areas, with development of
pneumonia but no fibrosis.
Exposure to 2.18 mg Al fibers/m3 for 6 hours/day, 5 days/week for up to 86 weeks
produced slightly increased alveolar macrophages and some irritation of the nasal passages in a
group of 50 Alderly Park rats (Pigott et al., 1981). Finally, a study by Drew et al. (1974)
observed the development of granulomatous nodules also developed in male hamsters that were
exposed to 8 mg Al/m3 of Alchlor (a propylene glycol complex of aluminum-chloride-hydroxide)
for 6 hours/day, 5 days/week for 20 or 30 exposures. The alterations persisted at the longest post
treatment observation (6 weeks) and consistently developed at the bifurcation of the
bronchioloalveolar ducts, which is a likely site of particulate deposition.
DERIVATION OF A PROVISIONAL CHRONIC RfD
FOR ALUMINUM
This survey of the toxicological effects of Al in rodents suggests that neurotoxicological
and developmental (including neurodevelopmental) endpoints are among the most sensitive
indicators of Al toxicity. However, as vehicles for the development of toxicity values such as a
provisional chronic RfD, the latter group of studies are considered to be more appropriate, since
the level of exposure to Al appears to be better characterized. In fact, neurobehavioral deficits
have been observed in mice and rats exposed during various stages of development and in
subchronic studies (Bernuzzi et al., 1989; Donald et al., 1989; Golub et al., 1989, 1992a, b, 1995;
Muller et al., 1990), as described above. These deficits include impaired operant learning,
changes in grip strength, altered startle response and impaired motor coordination. In addition,
several studies have shown that oral Al can produce histopathological changes in the CNS,
although the histopathological lesions have yet to be causally related to the neurobehavioral
deficits. Thus, Florence et al. (1994) reported histopathological changes in the brain of rats
exposed to dietary Al for 6 months, the changes including the appearance of vacuolation of the
cell body and cell processes of astrocytes in the brain and swelling of astrocytic processes. In
addition, more localized vacuolization of neurons in the brain also was observed. These changes
26
-------
10-23-2006
were observed in rats exposed to elevated A1 in the diet and are distinct from the NFD that has
been observed in rats, rabbits and monkeys maintained on elevated dietary A1 in combination
with reduced dietary calcium (Garruto et al., 1989; Kihira et al., 1994; Mitani, 1992; Yano et al.,
1989; Yoshida et al., 1990) or in rabbits administered intracisternal or intraventricular injections
of Al (Kowall et al., 1989; Wakayama et al., 1993). Interpretation of the low-calcium studies is
complicated by the observation that NFD was observed in animals maintained on low-calcium
diets without excess Al and was enhanced by the addition of excess Al to these diets (Garruto et
al., 1989; Kihira et al., 1994). Furthermore, Al has been shown to inhibit the gastrointestinal
absorption of calcium (Orihuela et al., 1996), an effect that may exacerbate the calcium
deprivation induced by low calcium diets. Thus, it is not clear whether calcium deprivation
enhances the neurotoxicity of Al or Al exacerbates the adverse effects of calcium deprivation.
Donald et al. (1989) and Golub et al. (1995) are co-principal studies that identify a
LOAEL of 100 mg Al/kg-day for minimal neurotoxicity in the offspring of mice exposed to
dietary aluminum lactate (soluble aluminum) during gestation and lactation. The neurotoxicity
associated with this LOAEL is consistent with LOAELs from other developmental and
subchronic neurobehavioral studies in mice and rats which used higher dietary dosages of
aluminum lactate or aluminum chloride (Golub et al., 1989, 1992a,b; Bernuzzi et al., 1989;
Muller et al., 1990). Of the above, Golub et al., (1995) is the only study in which a
histopathological examination of the brain and spinal cord was conducted and no abnormalities
were reported. The Florence et al. (1994) study indicates that histopathological abnormalities of
the CNS can occur in rats exposed subchronically to 84 mg/kg-day; although this is lower than
the LOAEL for neurobehavioral effects, it was not chosen as the principal study because the
functional significance of the histopathological lesions are uncertain.
A number of studies were identified that, at face value, appeared to indicate LOAELs at
lower doses than the 100 mg Al/kg-day value selected herein, for example, Paternain et al.
(1988) and Colomina et al. (1992). However, in these as in many of the studies under
consideration, insufficient information on dietary Al (Al content and/or feed type) was reported
to permit a reliable estimation of the overall dose level to which the animals were subjected.
Other developmental studies with aluminum hydroxide and/or citrate in mice and rats
identified a NOAEL which are equivalent (95.5 mg Al/kg-day), or a minimum LOAEL that was
greater (133 mg Al/kg-day) than the 100 mg Al/kg-day critical LOAEL (Domingo et al., 1989;
Gomez et al., 1991), an overlap potentially related to differences in effective doses due to
variations in unreported Al dietary content and factors affecting absorption such as chemical
form (e.g., the use of less absorbable aluminum hydroxide). In addition, the LOAEL of 43.3 mg
Al/kg-day for decreased body weight gain in mice exposed to aluminum chloride for 180-390
days (Ondreicka et al., 1966) was thought be inappropriate for risk assessment due to the small
sample size and to the poor reporting of study details. Aluminum nitrate caused alterations in
levels of brain biogenic amines and hepatic and hematological indices in rats exposed to 21.4 mg
Al/kg-day for 6 weeks (Flora et al., 1991). This dose is not a LOAEL because insufficient
information is available to determine if the effects are adverse.
Therefore, the LOAEL of 100 mg Al/kg-day for minimal neurotoxicity in the offspring of
mice (Donald et al., 1989, Golub et al., 1995) is selected as the basis for the provisional chronic
27
-------
10-23-2006
RfD. The LOAEL is considered minimal because the results of the postweaning neurobehavioral
test battery indicate that performance deficits may be marginal. In particular, of the three
observed effects (decreased forelimb and increased hindlimb grip strengths, increased hindlimb
foot splay distance), one effect (increased grip strength) has unclear toxicological significance
and two effects (increased grip strength and foot splay distance) did not persist after 2 weeks of
no further exposure.
Application of an uncertainty factor (UF) of 100 (3 for use of a minimal LOAEL, 10 for
interspecies extrapolation and 3 for intrahuman variability where the critical effects have been
observed in a sensitive sub-group) results in a provisional RfD of
p-RfD = 1E-0 mg Al/kg-day.
The provisional RfD of 1E-0 mg Al/kg-day is approximately 3-fold higher than estimated
normal daily A1 intake of approximately 0.2-0.3 mg/kg-day (Iyengar et al., 1987; Ganrot, 1986;
Wilhelm et al., 1990). Chronic users of medications such as antacids, buffered aspirins and
antiulceratives would be expected to ingest much larger amounts of Al, possibly as high as 10-70
mg/kg-day. However, these subjects would not represent the most sensitive population
(developing infants), as indicated by the animal data.
Low confidence is placed in the co-critical studies, because they only identify a LOAEL
for a sensitive effect and evaluated comparatively small numbers of animals. Confidence in the
data base is low because the most reliable supporting data for neurotoxicity of Al in humans are
of limited general relevance (e.g., dialysis encephalopathy is manifested in patients with
impaired renal function and excessive Al uptake from intravenous exposure). In fact,
neurotoxicity remains to be assessed in animals chronically exposed to Al, and developmental
morphology has not been adequately investigated in two animal species. These limitations in the
Al data base do not increase uncertainty in the RfD; therefore, a data base uncertainty factor was
not used. However, reflecting the low confidence in the co-critical studies, there is low overall
confidence in the RfD.
DERIVATION OF A PROVISIONAL CHRONIC RfC FOR ALUMINUM
Al seems to be the most likely cause for the generally and consistently reported
psychomotor and cognitive effects (particularly signs of impaired coordination) in Al production
workers and welders (Bast-Pettersen et al., 1994; Rifat et al., 1990; Hosovski et al., 1990; White
et al., 1992; Hanninen et al., 1994; Sjogren et al., 1990, 1996). In addition, there is strong
evidence that Al is neurotoxic by other routes of exposure. Thus, a degenerative neurological
syndrome (dialysis dementia) has been documented in humans with chronic renal failure,
apparently due to an increased exposure to Al from dialysis treatment and/or ingestion of
phosphate binding agents which contain Al (Alfrey, 1993). This syndrome is characterized by
gradual loss of motor, speech and cognitive functions. Neurotoxicity, particularly neuromuscular
effects such as decreased motor activity, startle responsiveness and grip strength, has also been
observed in mice following subchronic oral exposure and in the offspring of mice and rats
exposed orally during gestation and/or lactation. Based on this information, as well as evidence
28
-------
10-23-2006
that A1 is absorbed by A1 production workers and welders, the hypothesis that the occupational
studies are indicative of a neurotoxic effect of A1 appears to be justified. However, the only
occupational study that has yielded suitable monitoring data is that of Hosovski et al. (1990), in
which workers were exposed to presumed time-weighted average (TWA) concentrations of 4.6-
11.5 mg Al/m3 magnitude for an average of 12 years. Using 4.6 mg Al/m3 as the LOAEL for
psychomotor and cognitive impairment for an 8-hour occupational exposure (Hosovski et al.,
1990) and corrections for discontinuous exposure (10 m3/20 m3 and 5 days/7 days), the
LOAELhec is 1.64 mg/m3. Applying an uncertainty factor of 300 for intrahuman variability
(10), use of a LOAEL (10) and an incomplete database (3) yields a provisional RfC of
p-RfC = 1.64 mg/m3/300 = 5E-3 mg/m3.
The lack of inhalation developmental studies may increase uncertainty in the database
because oral data in animals indicate that neurotoxic and morphological developmental effects
may occur at lower doses than neurotoxicity in adults. Additionally, there is uncertainty related
to the lack of corroborating data on air concentrations associated with neurotoxicity. Confidence
in the critical study is low to medium because only a LOAEL was identified. Confidence in the
database is medium because (1) there are no corroborating data on effect levels (NOAELs and
additional LOAELs), (2) no data are available for developmental neurotoxicity by the inhalation
route and (3) a well-designed two-generation reproduction study is lacking. Reflecting the low
to medium confidence in the critical study and database, there is low to medium confidence in
the provisional RfC.
PROVISIONAL CARCINOGENICITY ASSESSMENT FOR
ALUMINUM
Weight-Of-Evidence Classification
A considerable number of epidemiological studies have examined the incidence of excess
tumor formation in persons occupationally exposed to Al in the form of dusts or fumes. In
general, a body of inferential evidence exists for an increase in cancer of the bladder and lung
through such occupational exposure to Al, although conclusions linking these responses to the
effects of Al are confounded by attendant co-exposure to other harmful emissions such as PAHs
and by cigarette smoking. A 20-month exposure of B6C3F1 mice to Al potassium sulfate
dodecahydrate in the diet at concentrations up to 10% w/w displayed no indication of compound-
related carcinogenicity and, in general, no indication of adverse toxicological effects of any kind
(Oneda et al., 1994). Similarly, the life-time exposure of Swiss mice and Long-Evans rats to 5
ppm Al as aluminum potassium sulfate in drinking water provided no convincing evidence for
the carcinogenicity of Al compounds (Schroeder and Mitchener, 1975a,b). Gene reversion
experiments on Al compounds resulted in negative results in S. typhimurium (Ahn and Jeffrey,
1994). Taking all of the evidence of Al carcinogenicity together, and in accordance with the
U.S. EPA (2005) cancer guidelines, aluminum is classified as inadequate information to assess
carcinogenic potential. The basis for this classification is insufficient evidence in
epidemiological/occupational studies, lack of demonstrated carcinogenicity or mutagenicity in
29
-------
10-23-2006
available animal studies, lack of positive evidence of non-carcinogenicity and lack of mode of
action data for aluminum.
Quantitative Estimates of Carcinogenic Risk
Due to insufficient data, a provisional oral slope factor and inhalation unit risk could not
be developed.
REFERENCES
ACGIH (American Conference of Governmental Industrial Hygienists). 1998. 1998 Threshold
Limit Values for Chemical Substances and Physical Agents and Biological Exposure Indices.
ACGIH, Cincinnati, OH.
Agarwal, S.K., L. Ayyash, C.S. Gourley, J. Levy, K. Faber and C.L. Hughes Jr. 1996.
Evaluation of the developmental neuroendocrine and reproductive toxicology of aluminum.
Food Chem. Toxicol. 34: 49-53.
Ahn, H-W. and E.H. Jeffrey. 1994. Effect of aluminum on fluoride uptake by Salmonella
typhimurium TA98: Implications for the Ames mutagenicity assay. J. Toxicol. Environ. Health.
41:357-368.
Alfrey, A.C. 1993. Aluminum toxicity in patients with chronic renal failure. Ther. Drug.
Monit. 15:593-597.
Armstrong, B.G., C.G. Tremblay, D. Cyrr and G.P. Theriault. 1986. Estimating the relationship
between exposure to tar volatiles and the incidence of bladder cancer in aluminum smelter
workers. Scand. J. WorkEnv. Health. 12 486-493.
ATSDR (Agency for Toxic Substances and Disease Registry). 1998. Toxicological Profile for
Aluminum (Update). U.S. Public Health Service. Atlanta, GA.
Bast-Pettersen, R., P.A. Drablos, L.O. Goffeng, Y. Thomassen and C.G. Torres. 1994.
Neuropsychological deficit among elderly workers in aluminum production. Am. J. Ind. Med.
25:649-662.
Basu, S., D. Chaudhuri and A.N. Chaudhuri. 1997. Influence of calcium on the toxic effects of
dietary aluminum. J. Food Sci. Toxicol. 34:264-266.
Bernuzzi, V., D. Desor and P.R. Lehr. 1989. Developmental alterations in offspring of female
rats orally intoxicated by aluminum chloride or lactate during gestation. Teratol. 40:21-27.
Cawthon, M.L. 1988. Chronic cyanide toxicity: A cause of neurological disease in aluminum
workers? Presented at the Northwest Occupational Health Conference. October 1988; Spokane,
WA. (as cited in White et al., 1992).
30
-------
10-23-2006
Christie, H., R.J. MacKay and A.M. Fisher. 1963. Pulmonary effects of inhalation of aluminum
by rats and hamsters. Ind. Hyg. J.24:47-56.
Colomina, M.T., M. Gomez, J.L. Domingo, J.M. Llobet and J. Corbella. 1992. Concurrent
ingestion of lactate and aluminum can result in developmental toxicity in mice. Res. Commun.
Chem. Pathol. Pharmacol. 77:95-106.
Colomina, M.T., M. Gomez, J.L Domingo and J. Corbella. 1994. Lack of maternal and
developmental toxicity in mice given high doses of aluminum hydroxide and ascorbic acid
during gestation. Pharmacol. Toxicol. 74:236-239.
Cucarella, C., C. Montoliu, C. Hermenegildo et al. 1998. Chronic exposure to aluminum
impairs neuronal glutamate-nitric oxide-cyclic GMP pathway. J. Neurochem. 70:1609-1614.
Cullen, M.R., H. Checkoway and B.H. Alexander. 1996. Investigation of a cluster of pituitary
adenomas in workers in the aluminum industry. Occup. Environ. Med. 53:782-786.
Deloncle, R., O. Guillard, F. Huguet and F. Clanet. 1995. Modification of the blood-brain
barrier through chronic intoxication by aluminum glutamate. Possible role in the etiological of
Alzheimer's disease. Biol. Trace Element Res. 47:227-238.
Deloncle, R., F. Huguet, P. Babin, B. Fernandez, N. Quellard and O. Guillard. 1999. Chronic
administration of aluminum L-glutamate in young mature rats: Effects on iron levels and lipid
peroxidation in selected brain areas. Toxicol. Lett. 104:65-73.
Deng, Z., C. Coudray, L. Gouzoux, A. Mazur, Y. Rayssiguier and D. Pepin. 1998. Effects of
oral aluminum and aluminum citrate on blood level and short-term tissue distribution of
aluminum in the rat. Biol. Trace Element Res. 63:139-147.
Domingo, J.L., J.M. Llobet, M. Gomez, J.M. Tomas and J. Corbella. 1987. Nutritional and
toxicological effects of short-term ingestion of aluminum by the rat. Res. Commun. Chem.
Pathol. Pharmacol. 56:409-419.
Domingo, J.L., M. Gomez, M.A. Bosque and J. Corbella. 1989. Lack of teratogenicity of
aluminum hydroxide in mice. Life Sci. 45:243-247.
Domingo, J.L., M. Gomez, J.M. Llobet and J. Corbella. 1991. Influence of some dietary
constituents on aluminum absorption and retention in rats. Kidney Int. 39:598-601.
Domingo, J.L., J. Llorens, D.J. Sanchez, M. Gomez, J.M. Llobet and J. Corbella. 1996. Age-
related effects of aluminum ingestion on brain aluminum accumulation and behavior in rats. Life
Sci. 58: 1387-1395.
Donald, J.M., M.S. Golub, M.E. Gershwin and C.L. Keen. 1989. Neurobehavioral effects in
offspring of mice given excess aluminum in diet during gestation and lactation. Neurotoxicol.
Teratol. 11:345-351.
31
-------
10-23-2006
Drablos, P. A., S. Hetland, F. Schmidt and Y. Thomassen. 1992. Uptake and excretion of
aluminum in workers exposed to aluminum fluoride and aluminum oxide. In Gitelman, H.J.
(ed.): Proceedings of the second international conference on aluminum and health. The
Aluminum Association, Washington, DC. p. 157-160.
Drew, R.T., B.N. Gupta, J.R. Bend and G.E.R. Hook. 1974. Inhalation studies with a glycol
complex of aluminum-chloride hydroxide. Arch. Environ. Health. 28:321-326.
Dufresne, A., P. Loosereewanich, B. Armstrong, G. Theriault and R. Begin. 1996. Inorganic
particles in the lungs of five aluminum smelter workers with pleuro-pulmonary cancer. Am. Ind.
Hyg. Assoc. J. 57:370-375.
Flora, S.J.S., M. Dhawan and S.K. Tandon. 1991. Effects of combined exposure to aluminum
and ethanol on aluminum body burden and some neuronal, hepatic and hematopoietic
biochemical variables in the rat. Hum. Exp. Toxicol. 10:45-48.
Florence, A.L., A. Gauthier, C. Ponsar, P. Van den Bosch de Aguilar and R.R. Crichton. 1994.
An experimental animal model of aluminum overload. Neurodegeneration. 3:315-323.
Ganrot, P.O. 1986. Metabolism and possible health effects of aluminum. Environ. Health
Perspect. 65: 363-441.
Garbossa, G., G. Galvez, M.E. Castro and A. Nesse. 1998. Oral aluminum administration to rats
with normal renal function. 1. Impairment of erythropoiesis. Hum. Exp. Toxicol. 17:312-317.
Garruto, R.M., S.K. Shankar, R. Yanagihara, A.M. Salazar, H.L. Amyx and D.C. Gajdusek.
1989. Low-calcium high-aluminum diet-induced motor neuron pathology in cynomolgus
monkeys. Acta Neuropathol. 78:210-219.
Gibbs, G.W. 1985. Mortality and aluminum reduction plant workers, 1950 through 1977. J.
Occup. Med. 27:761-770.
Glynn, A.W., A. Sparen, L-G. Daniellson, B. Sundstrom and L. Jorhem. 1999. Concentration-
dependent absorption of aluminum in rats exposed to labile aluminum in drinking water. J.
Toxicol. Environ. Health. 56:501-512.
Golub, M.S., J.M. Donald, M.E. Gershwin and C.L. Keen. 1989. Effects of aluminum ingestion
on spontaneous motor activity of mice. Neurotoxicol. Teratol. 11:231-235.
Golub, M.S., B. Han, C.L. Keen and M.E. Gershwin. 1992a. Effects of dietary aluminum
excess and manganese deficiency on neurobehavioral endpoints in adult mice. Toxicol. Appl.
Pharmacol. 112(1): 154-160.
Golub, M.S., C.L. Keen and M.E. Gershwin. 1992b. Neurodevelopmental effect of aluminum in
mice: Fostering studies. Neurotoxicol Teratol. 14(3): 177-182.
32
-------
10-23-2006
Golub, M.S., B. Han, C.L. Keen and M.E. Gershwin. 1994. Auditory startle response in Swiss-
Webster mice fed excess aluminum in diet. Neurotoxicol. Teratol. 16:423-425.
Golub, M.S., B. Han, C.L. Keen, M.E. Gershwin and R.P. Tarara. 1995. Behavioral
performance of Swiss-Webster mice exposed to excess dietary aluminum during development or
during development and as adults. Toxicol. Appl. Pharmacol. 133:64-72.
Gomez, M., J.L. Domingo and J.M. Llobet. 1991. Developmental toxicity evaluation of oral
aluminum in rats: Influence of citrate. Neurotoxicol. Teratol. 13(3):323-328.
Gross, P., R.H. Harley and R.T.P. de Treville. 1973. Pulmonary reaction to metallic aluminum
powders. Arch. Environ. Health. 26:227-236.
Hanninen, H., E. Matikainen, T. Kovala, S. Valkonen and V. Riihimaki. 1994. Internal load of
aluminum and the central nervous system function of aluminum welders. Scand. J. Work
Environ. Health. 20:279-285.
Hosovski, E., Z. Mastelica, D. Suderic and D. Radulovic. 1990. Mental abilities of workers
exposed to aluminum. Med. Lav. 81(2): 119-123.
Ittel, T.H., A. Griesner and H.G. Siberth. 1991. Effect of lactate on the absorption and retention
of aluminum in the remnant kidney rat model. Nephron. 57:332-339.
Iyengar, G.V., J.T. Tanner, W.R. Wolf and R. Zeisler. 1987. Preparation of a mixed human diet
material for the determination of nutrient elements, selected toxic elements and organic nutrients:
a preliminary report. Sci. Total Environ. 61:235-252.
Katyal, R., B. Desigan, C.P. Sodhi and S. Ojha. 1997. Oral aluminum administration and
oxidative injury. Biol. Trace Element Res. 57:125-130.
Katz, A.C., D.W. Frank, M.W. Sauerhoff, G.M. Zwicker and R.I. Freudenthal. 1984. A
6-month dietary toxicity study of acidic sodium aluminum phosphate in beagle dogs. Food
Chem. Toxic. 22:7-9.
Kihira, T., S. Yoshida, Y. Uebayashi, I. Wakayama and Y. Yase. 1994. Experimental model of
motor neuron disease: oral aluminum neurotoxicity. Biomed. Res. 15(l):27-36.
Kihira, T, S.Yoshida, J. Komoto, I. Wakayama and Y. Yase. 1995. Aluminum-induced model
of motor neuron degeneration: subperineurial injection of aluminum in rabbits. Neurotoxicol.
16:413-424.
Konishi. Y., K. Yagyu, H. Kinebuchi, N. Saito, T. Yamaguchi and Y. Ohtsuki. 1996. Chronic
effect of aluminum ingestion on bone in calcium-deficient rats. Pharmacol. Toxicol.
78:429-434.
33
-------
10-23-2006
Kowall, N.W., W.W. Pendlebury, J.B. Kessler, D.P. Perl and M.F. Beal. 1989. Aluminum-
induced neurofibrillary degeneration affects a subset of neurons in rabbit cerebral cortex, basal
forebrain and upper brainstem. Neuroscience 29:329-37.
Langauer-Lewowicka, H. and Z. Braszczynska. 1983. A trial evaluation of combined action of
certain physical and chemical and harmful factors on the nervous system. Neurol. Neurochir.
Pol. 24:91-96. (as cited in White et al., 1992).
Lione, A. 1985. Aluminum toxicology and the aluminum-containing medications. Pharmacol.
Therap. 29:255-285.
Marquis, J.K. Neurotoxicity of aluminum. 1989. Environmental Chemistry and Toxicology of
Aluminum. T.E. Lewis, Ed. Lewis Publishers, Inc., Chelsea, Michigan, p. 289-298.
Martyn, C.N., D.N. Coggon, H. Inskip, R.F. Lacey and W.F. Young. 1997. Aluminum
concentrations in drinking water and risk of Alzheimer's disease. Epidemiology 8:281-286.
Mitani, K. 1992. Relationship between neurological diseases due to aluminium load, especially
amyotrophic lateral sclerosis, and magnesium status. Magnes. Res. 5:203-213.
Muller, G., V. Bernuzzi, D. Desor, M-F. Hutin, D. Burnel and P.R. Lehr. 1990. Developmental
alterations in offspring of female rats orally intoxicated by aluminum lactate at different
gestation periods. Teratol. 38:253-261.
NIOSH (National Institute for Occupational Safety and Health). 1994. Pocket Guide to
Chemical Hazards. U.S. Department of Health and Human Services, Washington, DC.
Ondreicka, R., E. Ginter and J. Kortus. 1966. Chronic toxicity of aluminum in rats and mice
and its effects on phosphorus metabolism. Brit. J. Industr. Med. 23:305-312.
Oneda, S., T. Takasaki, K. Kuriwaki et al. 1994. Chronic toxicity and tumorigenic study of
aluminum potassium sulfate in B6C3F1 mice. In Vivo. 8:271-278.
Orihuela, D., C.E. Carnovale, J. A. Monti and M.C. Carrillo. 1996. Sex-related differences in
the effect of aluminum on calcium transport in the small intestine of the rat. Toxicol. Lett.
85:165-171.
Partridge, N.A., F.E. Regnier, J.L. White and S.L. Hem. 1989. Influence of dietary constituents
on intestinal absorption of aluminum. Kidney Int. 35:1413-1417.
Paternain, J.L., J.L. Domingo, J.M. Llobet and J. Corbella. 1988. Embryotoxic and teratogenic
effects of aluminum nitrate in rats upon oral administration. Teratol. 38:253-257.
Pearson, R.J.C., M.C. Battigelli and J.T. Gamble. 1993. Mortality from working in the
aluminum reduction industry. Environ. Res. 61:19-24.
34
-------
10-23-2006
Pennington, J. A. and S.A. Schoen. 1995. Estimates of dietary exposure to aluminum. Food
Add. Contam. 12:119-128.
Pigott, G.H., B.A. Gaskell and J. Ishmael. 1981. Effects of long-term inhalation of aluminum
fibres in rats. Br. J. Exper. Pathol. 62:323-331.
Reiber, S., W. Kukull and P. Standish-Lee. 1995. Drinking water aluminum and bioavailability.
J. Am. Waterworks Assoc. 87:86-100.
Rifat, S.L., M.R. Eastwood, D.R. Crapper McLachlan and P.N. Corey. 1990. Effect of exposure
of miners to aluminum powder. Lancet 336 (8724): 1162-1165.
Ronneberg, A. and F. Langmark. 1992. Epidemiologic evidence of cancer in aluminum
reduction plant workers. Am. J. Ind. Med. 22:573-590.
Schroeder, H.A. and M. Mitchener. 1975a. Life-term studies in rats: effects of aluminum,
barium, beryllium, and tungsten. J. Nutr. 105:421-427.
Schroeder, H.A. and M. Mitchener. 1975b. Life-term effects of mercury, methyl mercury, and
nine other trace metals on mice. J. Nutr. 105:452-458.
Selden, A.I., H.B. Westberg and O. Axelson. 1997. Cancer morbidity in workers at aluminum
foundries and secondary aluminum smelters. Am. J. Ind. Med. 32:467-477.
Sjogren, B., P. Gustavsson and C. Hogstedt. 1990. Neuropsychiatric symptoms among welders
exposed to neurotoxic metals. Br. J. Ind. Med. 47:707-707.
Sjogren, B., A. Iregren, W. Freeh et al. 1996. Effects on the nervous system among welders
exposed to aluminum and manganese. Occup. Environ. Med. 53:32-40.
Slanina, P., Y. Falkeborn, W. Freeh and A. Cedergren. 1984. Aluminum concentrations in the
brain and bone of rats fed citric acid, aluminum citrate or aluminum hydroxide. Food Chem.
Toxicol. 22:391-397.
Slanina, P., W. Freeh, A. Bernhardson, A. Cedergren and P. Mattsson. 1985. Influence of
dietary factors on aluminum absorption and retention in the brain and bone of rats. Acta
Pharmacol. Toxicol. 56:331-336.
Slanina, P., W. Freeh, L.G. Ekstrom, L. Loof, S. Slorach and A. Cedergren. 1986. Dietary citric
acid enhances absorption of aluminum in antacids. Clin. Chem. 32:539-541.
Somova, L.I., A. Missankov and M.S. Khan. 1997. Chronic aluminum intoxication in rats:
Dose-dependent morphological changes. Meth. Find. Exp Clin. Pharmacol. 19:599-604.
Spinelli, J.J., P.R. Band, L.M. Svirchev and R.P. Gallagher. 1991. Mortality and cancer
incidence in aluminum reduction plant workers. J. Occup. Med. 33(11): 1150-1155.
35
-------
10-23-2006
Steinhagen, W.H., F.L. Cavender and B.Y. Cockrell. 1978. Six month inhalation exposures of
rats and guinea pigs to aluminum chlorohydrate. J. Environ. Path. Toxicol. 1:267-277.
Stokinger, H.E. 1981. The Metals: Aluminum. In: Patty's Industrial Hygiene and Toxicology,
Volume IIA, 3rd rev ed., G.D. Clayton and F.E. Clayton, eds. John Wiley and Sons, New York,
p. 1493-1504.
Sutherland, J.E. and J.L. Greger. 1998a. Kinetics of aluminum disposition after ingestion of low
to moderate doses of aluminum. Toxicology. 126:115-125.
Sutherland, J.E. and J.L. Greger. 1998b. Effect of the size of an oral dose of aluminum on the
relative importance of biliary v. urinary aluminum excretion in conscious rats. Food Chem.
Toxicol. 36:505-512.
Taylor, G.A., P.B. Moore, I.N. Ferrier, S.P. Tyrer and J.A. Edwardson. 1998. Gastrointestinal
absorption of aluminum and citrate in man. J. Inorg. Biochem. 69:165-169.
Theriault, G., C. Tremblay, S. Cordier and S. Gingras. 1984. Bladder cancer in the aluminum
industry. Lancet. i:949-950.
Theriault, G.P., C.G. Tremblay and B.G. Armstrong. 1990. Bladder cancer screening among
primary aluminum production workers in Quebec. J. Occup. Med. 32(9):869-872.
Tremblay, C., B. Armstrong, G. Theriault and J. Brodeur. 1995. Estimation of risk of
developing bladder cancer among workers exposed to coal tar pitch volatiles in the primary
aluminum industry. Am. J. Ind. Med. 27:335-348.
U.S. EPA. 1987. Health Effects Assessment for Aluminum. Prepared by the Office of Health
and Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH
for the Office of Solid Waste and Emergency Response, Washington, D.C. EPA600/8-88/016.
U.S. EPA. 1988. Recommendations for and Documentation of Biological Values for Use in
Risk Assessment. Prepared by the Office of Health and Environmental Assessment,
Environmental Criteria and Assessment Office, Cincinnati, OH for the Office of Solid Waste and
Emergency Response, Washington, DC. EPA 600/6-87-008. NTIS PB88-179874.
U.S. EPA. 1994. Chemical Assessments and Related Activities (CARA). Office of Health and
Environmental Assessment, Washington, DC.
U.S. EPA. 1997. Health Effects Assessment Summary Tables. FY-1997 Annual and FY-1997
Supplement. Office of Research and Development, Office of Emergency and Remedial
Response, Washington, DC.
U.S. EPA. 2005. Guidelines for Carcinogen Risk Assessment. Risk Assessment Forum,
Washington, DC. EPA/630/P-03/001B. Available at: www.epa.gov/cancerguidelines
36
-------
10-23-2006
U.S. EPA. 2006. Integrated Risk Information System (IRIS). Office of Research and
Development, National Center for Environmental Assessment, Washington, DC. Online.
www.epa.gov/iris/
U.S. FDA. 2000. Aluminum in Large and Small Volume Parenterals Used in Total Parenteral
Nutrition. FR 65(17):4103-4111.
van der Voet, G.B. and F.A. de Wolff. 1986. Intestinal absorption of aluminum in rats: effect
of intraluminal pH and aluminum concentration. J. Appl. Toxicol. 6:37-41.
Varner, J., K.F. Jensen, W. Horvath and R.L. Isaacson. 1998. Chronic administration of
aluminum-fluoride or sodium-fluoride to rats in drinking water: Alterations in neuronal and
cerebrovascular integrity. Brain Res. 784:284-298.
Wakayama, I., V.R. Nerurkar and R.M. Garruto. 1993. Immunocytochemical and ultrastructural
evidence of dendritic degeneration in motor neurons of aluminum-intoxicated rabbits. Acta
Neuropathol. 85:122-128.
White, D.M., W.T. Longstreth, Jr., L. Rosenstock, K.H.J. Claypoole, C.A. Brodkin and B.D.
Townes. 1992. Neurologic syndrome in 25 workers from an aluminum smelting plant. Arch.
Intern. Med. 152(7): 1443-1448.
Wilhelm, M., D.E. Jager and F-K. Ohnesorge. 1990. Aluminum toxicokinetics. Pharmacol.
Toxicol. 60:4-9.
Wilhelm, M., I. Lombeck, B. Kouros, J. Wuthe and F-K. Ohnesorge. 1995. Duplicate study on
the dietary intake of some metals/metalloids by German children. Part II: Aluminum, cadmium
and lead. Zbl. Hyg. 197:357-369.
Yano, I., S. Yoshida, Y. Uebayashi, F. Yoshimasu and Y.Yase. 1989. Degenerative changes in
the central nervous system of Japanese monkeys induced by administration of aluminum salt.
Biomed. Res. 10:33-41.
Yokel, R.A. and P.J. McNamara. 1988. Influence of renal impairment, chemical form, and
serum protein binding on intravenous and oral aluminum kinetics in the rabbit. Toxicol. Appl.
Pharmacol. 95:32-43.
Yoshida, S., M.E. Gershwin, C.L. Keen, J.M., Donald and M.S. Golub. 1989. The influence of
aluminum on resistance to Listeria monocytogenes in Swiss-Webster mice. Arch. Allergy Appl.
Immunol. 89:404-409.
Yoshida, S., I. Yano, I. Wakayama, K. Mitani and Y. Yase. 1990. Morphometric analysis of
neurodegenerative changes induced by low calcium-magnesium and excessive aluminum intake.
Biomed. Res. 11(1):11-18.
37
-------
10-23-2006
Table 1. Summary of oral toxicity data for aluminum11
Study
Type
Species
Al
Exposure
Concentration
(ppm)
Exposure
Dosage
(mg Al/kg-
day)
Exposure
Frequency and
Duration
Critical Effect
NOAEL
(mg Al/kg-
day)
LOAEL
(mg Al/kg-
day)
FEL
(mg Al/kg-
day)
Ondreicka
etal., 1966
Subchronic 3-
gen dietary
Dobra Voda
mice
chloride
--
27 (control),
46
Continuous,
180-390 days
Decreased body weight gain
inFl andF2.
--
46
--
Golub et
al., 1989
Subchronic
dietary
S-W mice
lactate
25 (control),
500,1000
3.3 (control),
65,130
Continuous, 6
weeks
Decreased spontaneous
motor activity; decreased
weight gain.
65
130
Golub et
al., 1992a
Subchronic
dietary
S-W mice
lactate
25 (control),
1000
190
Continuous, 90
days
Decreased hindlimb grip,
decreased
spontaneous motor activity,
decreased startle response.
190
Florence et
al., 1994
Subchronic
dietary
Wistar rat
chloride
(with citric
acid)
1.52 (control),
1000
0.13
(control), 84
Continuous, 6
months
Histopathological changes in
brain astrocytes and neurons.
84
Domingo et
al., 1996
Subchronic
drinking water
Sprague
Dawley rats
nitrate
(with
citric acid)
0, 50, 100
(plus
unreported
dietary Al)
Continuous, 6.5
months
Operant conditioning and
performance
100
Yoshida et
al., 1989
Subchronic
dietary
S-W mice
lactate
25 (control), 500,
1000
5 (control),
98, 195
Continuous, 7
weeks
Increased mortality from L.
monocytogenes inoculation
195
--
Donald et
al., 1989
Developmental
dietary
S-W mice
lactate
25 (control), 500,
1000
5 (control),
100, 200
Continuous,
gestation and
lactation
Neurobehavioral effects.
100
Golub et
al., 1992b
Developmental
dietary
S-W mice
lactate
25 (control),
1000
4 (control),
174
Continuous,
gestation and
lactation
Neurobehavioral effects.
174
Golub et
al., 1995
Developmental
dietary
S-W mice
lactate
7, 500, 1000
1 (control),
100, 200
Continuous,
gestation,
lactation to
maturity
Neurobehavioral effects.
100
38
-------
10-23-2006
Table 1. Summary of oral toxicity data for aluminum3
Study
Type
Species
Al
Exposure
Concentration
(ppm)
Exposure
Dosage
(mg Al/kg-
day)
Exposure
Frequency and
Duration
Critical Effect
NOAEL
(mg Al/kg-
day)
LOAEL
(mg Al/kg-
day)
FEL
(mg Al/kg-
day)
Yoshida et
al., 1989
Developmental
dietary
S-W mice
lactate
25 (control), 500,
1000
4 (control),
100, 200
Continuous,
gestation and
lactation
Increased mortality of dams
fromZ. monocytogenes
inoculation
~~
100
~~
aStudies for which total dosages were reported or could be estimated (unless otherwise noted).
39
------- |