Revised External Review Draft
October 1978
AN ASSESSMENT OF
THE HEALTH EFFECTS OF
ARSENIC GERMANE TO
LOW-LEVEL EXPOSURE
NOTICE
This document is a preliminary draft. It has been
released by EPA for public review and comment
and does not necessarily represent Agency policy.
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
OFFICE OF HEALTH AND ECOLOGICAL EFFECTS
WASHINGTON, D.C. 20460 °
-------
PREFACE
The Environmental Protection Agency has prepared three
documents concerning the health effects of arsenic on the general
population:
1. A health effects assessment,
2. An environmental exposure assessment, and
3. A population risk assessment based on the data pre-
sented in the first two documents.
This report, the health'effects document, will be used by
the Environmental Protection Agency's Office of Air and Waste
Management, and by the Administrator, to determine the scientific
basis for possible actions regarding arsenic under the Clean Air
Act. The report was prepared under the direction of the Criteria
Development and Special Studies Division, Office of Health and
Ecological Effects, with participation by the following Division
personnel:
Dr. Alan Carlin
Dr. Roger Cortesi
Dr. Arthur Saz
Drafts of the three documents were reviewed by the Environ-
mental Protection Agency's Science Advisory Board Study Group on
Arsenic as a Hazardous Air Pollutant in public session on May
22-23, 1978. Members of this panel were:
Dr. Ruth R. Levine, Chairperson
Division of Medicine and Dentistry
Boston University
Boston, Massachusetts
111
-------
Dr. Bertram Carnow
School of Public Health
University of Illinois at Chicago Circle
Chicago, Illinois
Dr. Ursula M. Cowgill
Department of Biology
University of Pittsburgh
Pittsburgh, Pennsylvania
Dr. Samuel S. Epstein
School of Public Health
University of Illinois at Chicago Circle
Chicago, Illinois
Dr. Eva Killam
Department of Pharmacology
University of California
Davis, California
Dr. Harold Peck
Merck, Sharpe and Dohme
Westport, Pennsylvania
Ms. Anne Wolven
Private Consultant
Atlanta, Georgia
Review copies of this document were also provided -to other
government agencies and to industrial and public interest groups,
as the result of a notice that appeared in the Federal Register
April 26, 1978, on page 18246.
Comments and criticisms received at these meetings and in
response to the Federal Register notice have been reviewed and
incorporated into the report as deemed appropriate.
IV
-------
TABLE OF CONTENTS
Preface
Figures
Tables
Acknowledgments
Summary
1. Introduction 1
2. Physical and Chemical Properties 6
Route of Entry, Distribution, Elimination 6
Metabolism 9
Biochemistry and Toxicity 11
3. Experimental Studies 15
Carcinogenicity. 15
Mutagenicity 21
Teratogenicity 26
4. Effects of Human Exposure to Arsenic 29
Smelter Workers and Community Residents 29
Community Exposures 38
Manufacture of Arsenical Pesticides 42
Agricultural Workers 45
Medicinal Use of Arsenic Compounds 50
Ingestion of Arsenic from Contaminated ^"\
Food and Water f 5.2X'
5. Noncarcinogenic Toxic Effects of Arsenic
on Humans 63
Low-level Effects of Arsenic Exposure 67
Organic Arsenic 68
6. Assessment of the Health Effects of Arsenic 70
Introduction 70
Ingestion of Arsenic 72
Occupational Exposure 74
Smelter Worker Mortality Rate Studies 76
Biological Monitoring 77
Air Sampling 80
Evaluating the Exposure 82
Assessment of the Effects of Community
Exposure 88
7. References 97
v
-------
LIST OF FIGURES
No. Page
1 Dose-response Data on Arsenic Concentration in
Well Water and Skin Cancer 59
2 Comparisons of Urinary Arsenic Excretion and
Concentration of Inhaled Arsenic 81
LIST OF TABLES
No. Page
1 Selected Arsenic Compounds and Their Chemical
Structures 2
2 Arsenic in the Environment 3
3 Effect of Route of Administration on Arsenic
Distribution and Excretion of Pentavalent Arsenic
in Rats " 10
4 Experimental Carcinogenesis Studies with Arsenic 16
5 Experimental Studies of Cocarcinogenesis, Tumor
Promotion, and Initiation 18
6 Summary of Studies on Inorganic Arsenic Mutagenicity 22
7 Summary of Epidemiological Studies of Smelter Workers 31
8 Summary of Epidemiological and Clinical Studies of
Nonsmelter Occupational Exposure to Arsenic 46
9 Summary of Epidemiological and Clinical Studies of
Effects of Ingested Arsenic 53
10 Toxic Effects of Arsenic 65
11 Standardized Respiratory Cancer Mortality Rates
Observed Among Several Smelter-worker Cohorts 78
VI
-------
LIST OF TABLES (continued)
No. . Page
12 Respiratory Cancer SMR's as a Function of Duration
and Degree of Exposure 79
13 Atmospheric Arsenic Concentrations in 1965 Smelter
Survey 83
14 Relationship Between Arsenic Exposure and Lung
Cancer Mortality Calculated by Pinto et al. 84
15 Comparison of Urinary Arsenic Excretion 86
16 Derivation of Air Equivalents from Urinary
Arsenic Levels and Corresponding SMR's as a
Function of Duration 87
17 Respiratory Cancer Mortality and Smoking Among
Smelter Workers 89
18 Comparison of U.S. National Average .Lung Cancer
Mortality Rates with the Rates Experienced by
Counties in Which Copper Smelters Are Located 92
19 Distribution of Lung Cancer SMR's in U.S. Counties
with Copper Smelters and Refineries, and with Only
Copper Smelters 93
20 Arsenic in Smelter Feeds and Lung Cancer Rates 94
21 U.S. Counties Engaged in the Primary Smelting and
Refining of Nonferrous Ores in 1963 96
-------
ACKNOWLEDGMENTS
This document was prepared by EPA's Office of Research and
Development with extensive help from a team of consultants led by
Jeanne M. Stellman, Ph.D. Geoffrey Kabat, Ph.D., was a major
contributor.
vm
-------
SUMMARY
Arsenic, which occurs naturally in the environment, may
present a health hazard to humans when it is released into the
environment as a consequence of industrial, manufacturing, and
agricultural processes. The preponderance of clinical and
epidemiological evidence regarding the effects of arsenic
pertains to trivalent inorganic arsenic. Much of this evidence
suggests that trivalent inorganic arsenic is a carcinogen. Some
limited evidence suggests that pentavalant inorganic arsenic may
also be carcinogenic.
The main routes by which arsenic enters the human body are
inhalation and ingestion. Experimental studies show that
following injection of trivalent inorganic arsenic, arsenic is
concentrated initially in the liver, kidneys, lungs, and spleen.
After 24 hours the level in the liver and kidneys decreases,
while that in the skin increases.
Both trivalent and pentavalent arsenic are mutagenic and
teratogenic in animal tests. Attempts to induce tumors in
experimental animals by arsenic usually have been unsuccessful,
although, significantly in some instances, positive results have
been obtained. Animal toxicity studies indicate that trivalent
arsenic is several times more toxic than pentavalent arsenic.
Epidemiological studies of smelter workers show excess lung
cancer risk in individuals exposed to arsenic trioxide, and
several studies indicate that the risk increases with increasing
duration and level of exposure. Risk increases of up to 8-fold
have been reported. It is possible that the increased risk is
not due to arsenic alone, since smelter workers are exposed
concomitantly to sulfur dioxide and other toxic substances.
Workers engaged in the manufacture of arsenical pesticides also
IX
-------
showed elevated risk for pulmonary cancer. Such workers, though
not exposed to sulfur dioxide, may be exposed to other toxic
substances. Similar results have been recorded in vineyard
workers in France and Germany, where the workers were exposed to
pesticides containing trivalent and pentavalent arsenic. In the
United .States, agricultural workers exposed to lead arsenate
spray (pentavalent) evidenced excess lung cancer associated with
this process.
The currently available experimental and epidemiological
evidence does not provide an adequate basis for gauging the effects
of chronic low-level exposure to arsenic compounds. A clear
dose-response effect has been noted in a large-scale study in
which ingestion of drinking water containing arsenic was associated
with subsequent development of skin cancer. It is reasonable
to assume that smaller dosages of inhaled arsenic could be
involved with the development of cancer, since inhalation is
a more efficient route of entry to the body than ingestion. In
addition, as noted above, dose-response relationships have been
reported in smelter workers.
It is difficult to determine a specific level of exposure
associated with a specific level of risk because the precision
of ambient air measurements is low; the level may be incorrect
by as much as one order of magnitude. Nevertheless, the air
levels yielding standard mortality ratios of 800 (i.e., an
8-fold increase in lung, cancer deaths) have been estimated at
23 to 323 yg of arsenic per cubic meter of air.
-------
SECTION 1
INTRODUCTION
Arsenic is a metal that occurs naturally in compounds with
sulfur and with other metals (copper, cobalt, iron, lead, zinc,
etc.). In its various forms arsenic occurs in trace amounts
throughout the environment in water, solids, rocks, and living
organisms. Arsenic is the 20th most common trace element in the
earth's crust (Schroeder and Balassa, 1966). In addition to its
occurrence in natural forms, arsenic trioxide is produced by
smelters and by coal-burning power plants, and refined arsenic
trioxide is used as the raw material for a large number of indus-
trial and agricultural products, both inorganic and organic.
Some of the important natural and commercial arsenic com-
pounds are shown in Table 1 with their chemical structures. Some
of the sources of arsenic in various phases of the environment
are shown in Table 2. . •
The toxicity of arsenic compounds varies greatly, depending
on valence state, chemical structure, and route of entry.
Elemental arsenic is considered nontoxic (Buchanan, 1962), where-
as arsine gas and lewisite [dichloro (2-chlorovinyl) arsine] are
extremely toxic (Buchanan, 1962). The arsenic in seafood (some-
times referred to as "shrimp arsenic") appears to be organically
bound, is excreted rapidly in humans, and is generally thought to
be nontoxic (National Academy of Sciences, 1977).
Since this report deals with the human health effects of
airborne arsenic, the complex questions of the distribution and
circulation of arsenic throughout the environment and the pos-
sible hazards to wildlife and humans by arsenic in the soil,
foodstuffs, and drinking water are beyond its scope. Case
studies involving ingestion of arsenic in beverages, drinking
1
-------
TABLE 1. SELECTED ARSENIC COMPOUNDS AND THEIR
CHEMICAL STRUCTURES3
AsS, AsS4 As2S3
REALGAR SULFURET AURIPIGMENTUM
ORPIMENT
C1-CH-CH-As-Cl2 As203
LEWISITE ARSENOUS OXIDE
WHITE ARSENIC
As=0
OH
MAPHARSIDE MAPHARSEN
ARSENOXIDE OXOPHENARSINE
OH .
i
PHENYL ARSONIC ACID
FeAsS
ARSENOPYRITE
MISPICKEL
AS205
ARSENIC OXIDE
H
i
i
H
PHENARSAZINE
As = As
H2N-l^j) iL^l- NHCH2S02Na
OH
NEOARSPHENAMINE
HC1-NH
OH OH
ARSPHENAMINE "606" SALVARSAN
NaO-
OH
TRYPARSAMIDE
ONa
NaO - As =• 0
NH
MELARSEN OXIDE
Taken from: Vallee et al. (1960)
-------
TABLE 2. ARSENIC IN THE ENVIRONMENT
u>
Milieu
Air
Soil
Mater
Food
Compound
Arsenic trioxide
Lead arsenate
Arsenic acid
Sodium arsenitc
Calcium arsenate,
etc.
Cacodylic acid
MS MA
DSMA, etc.
Unidentified arsenic
compounds
Arsenic compounds
in fruits and vege-
tables
Unidentified arsenic
in seafood (thought to
be organically bound)
Source
Milling and refining;
smelters; coal-burning
power plants; petroleum
Weathering of arsenic -
containing rocks
Pesticides;
insecticides
Herbicides
Dissolution of pyrites,
minerals and ores;
industrial effluents
containing arsenic
Arsenical herbicides
and pesticides;
arsenical feed additives
Weathering and leaching
of minerals containing
arsenic.
Levels reported (as elemental arsenic)
Up to 2.5 M9/m at property line of
Tacoma smelter3
Less than 10 ppm to 500 ppro
1100 ug/ liter downstream from an
industrial complex; 0.14-1.0 ppm
in seawater near mouths of estuaries
draining industrial areas °
Up to 174 ppm in prawns0
Nelson (1977)
Luh et al. (1973)
Vallee (1960)
-------
water, and medicinal preparations are mentioned only as they
pertain to the association between arsenic and cancer.
This report is concerned primarily with inorganic arsenic.
None of the scientific evidence to date implicates organic arse-
nic as a carcinogen, and few scientific studies have dealt with
the effects of long-term exposure to organic arsenic compounds,
such as the herbicides monosodium methanearsonate, disodium
methanearsonate, and cacodylic acid. For these reasons it is
difficult to draw conclusions as to whether organic arsenic
compounds pose a hazard to the general population. The available
evidence and its implications are discussed in Section 6.
Most of the available data on the human toxicity of in-
organic arsenic relate to human exposure to trivalent arsenic.
Much of this information indicates that inorganic trivalent
arsenic is a human carcinogen, and that at high levels it can
induce other serious, irreversible effects, such as cardiovas-
cular disease and severe peripheral neuropathy. Trivalent in-
organic arsenic is emitted into the air along with sulfur dioxide
and other contaminants by smelters and coal-burning power plants.
Although there is more clinical and epidemiological evidence
for the carcinogenicity of trivalent than pentavalent inorganic
arsenic, several studies suggest that pentavalent inorganic
arsenic may also be a carcinogen (Frohn, 1938; Roth, 1957; Ott et
al. , 1974; NIOSH, 1975: reevaluation of Nelson et al. , 1973).
This evidence is discussed in a recent EPA report (Carcinogen
Assessment Group, 1978).
It has been demonstrated that both valency forms of in-
organic arsenic interfere with normal genetic functioning,
although the trivalent form is more toxic than the pentavalent.
This document pertains to arsenic in particulate matter in
the air, rather than to the levels of both inorganic and organic
arsenic compounds in waterways, soils, and foods, a concern that
deserves consideration elsewhere. A recent review of arsenic
toxicity (Luh et al., 1973) points out that no standard has been
set for arsenic levels in industrial effluents. The authors
-------
further note that organisms in the aquatic food chain concentrate
arsenic by factors of hundreds to thousands over the amounts
present in the surrounding water. The only incidents related to
human ingestion of contaminated drinking water have resulted from
contamination with inorganic arsenic, and these are discussed in
this report.
-------
SECTION 2
PHYSICAL AND CHEMICAL PROPERTIES
ROUTE OF ENTRY, DISTRIBUTION, ELIMINATION
The main routes by which arsenic can enter the body are
inhalation and ingestion, although absorption through the skin is
also a possible minor route of entry (Sollman, 1964; Oehme, 1972;
Garb and Hine, 1977). Arsenic compounds are generally absorbed
onto particulate matter. The biological fate of the arsenic thus
depends on particle size and rate of solution (i.e., solubility).
The optimum range for deposition in the lower tracheobronchial
tree, where it can either lodge or be absorbed into the blood
stream, is 0.1 to 2 ym. . Particles smaller than 0.1 ym remain in
suspension in the inhaled air and are exhaled. Particles larger
than 2 ym are trapped by the mucous- membranes of the nose and
throat and can be swallowed; these enter the gastrointestinal
tract, either to be absorbed or to be excreted (Falk and Kotin,
1961).
Emissions of arsenic and other toxic trace elements from
high-temperature combustion sources such as smelters have been
observed to be mainly in the size range of less than 1 ym in
diameter. Thus they are respirable and are capable of being
absorbed through the lungs, the most efficient route of entry
(Natusch and Wallace, 1974).
Some information on the distribution of arsenic and its
elimination is available from experimental studies involving
administration of specific arsenic compounds (usually radio-
actively labelled) and measurement of arsenic levels in specific
tissues and in-urine and feces. Arsenic metabolism is poorly
understood, however, owing primarily to the scarcity of information
-------
about the form in which arsenic occurs normally in the diet and
in human tissues and the interconversions (reduction, oxidation,
methylation) of different arsenic compounds in the body (Crecelius,
1977) .
Vallee et al. (1960), who reviewed the literature on the
toxicology of arsenic in animals and man, report that.in most
species arsenic concentrates in the liver, kidneys, lung, spleen,
and skin in the first 24 hours following oral and parenteral
administration. Thereafter, the concentration in the skin in-
creases while that in the liver and kidneys decreases. Low
levels of arsenic are distributed throughout the body tissues,
but bone, muscle, and skin contain proportionately greater
amounts because of their greater mass. They represent the "major
storage areas." Following oral administration high levels of
arsenic appear temporarily in the gastrointestinal tract. Arse-
nic is excreted predominantly in the urine and to a much lesser
degree in the feces. Excretion takes place rapidly during the
first 2 days following intake and more slowly for 7 to 10 days
thereafter. Small amounts of arsenic are excreted at a much
lower rate through ectodermal tissue (mainly the hair and nails)
(Vallee et al., 1960).
The general pattern, then, is rapid clearance from the
blood; short-term concentration in the liver, kidneys, lung, and
spleen; and rapid excretion of the bulk of the administered dose
in the urine. This pattern appears to apply to trivalent arsenic
in the rabbit, mouse, guinea pig, and man, and to pentavalent
arsenic in the rabbit (DuPont et al., 1942) and cow (Peoples,
1964). Excretion of arsenic by the rat is much slower than in
other species owing to the rat's ability to store arsenic in its
j
red blood cells.
A study of inhaled sodium arsenite (74^s) in humans has
shown that the rate of absorption from the bronchial tree was
rapid for the first several days and then decreased slowly.
Three patients excreted 45 percent of the inhaled arsenic in the
urine in 10 days and excreted 2.5 percent in the feces. The
-------
remainder was assumed to have been deposited in the body, ex-
haled, or excreted over a long period of time (Holland et al.,
1959).
Hunter et al. (1942) found that subcutaneously injected
potassium arsenite ('^As) was concentrated in the blood in rats
but was generally distributed throughout the tissues of guinea
pig, rabbit, chimpanzee, one baboon, and man. Excre.tion took
place mainly through the kidney and was essentially complete in
6 days. On examining different fractions of tissues from this
experiment, >Lowry et al. (1942) found that most of the arsenic
was in the protein fraction, a small amount in the acid-soluble
fraction, and a trace amount in the lipid fraction.
Ducoff et al. (1948) compared the distribution and excretion
of sodium arsenite ('"As) in rat, rabbit, mouse, and man. They
confirmed the observation that arsenic concentrates in the blood
of the rat and found that in the rabbit it is distributed by the
bloodstream and concentrates in liver, kidneys, and lungs.
In a woman with carcinoma of the parotid, arsenic concentra-
tion was highest in the liver and kidney 20 hours after injection
of sodium arsenite ( As).
Cumulative excretion as a percentage of the administered
dose was most rapid in rabbits (more than 90% within 96 hours
following injection), slower in two human subjects (40 and 50%
within 96 hours following injection), and slowest in rats (10%
within 72 hours following injection). (No exact measurements
were made on excretion rates in mice.)
74
Lanz et al. (1950) administered sodium arsenate ( As)
intramuscularly in the rat, dog, cat, chick, guinea pig, and
rabbit. After 48 hours, less than 0.27 percent of the arsenic
was present in the organs studied in all species except the rat
and cat, which retained 79 and 5.6 percent, respectively, in the
blood. Arsenic in the blood of rats was primarily in the hemo-
globin. Because of this storage of arsenic in the red blood
cells, where it remains for the lifetime of the cell, the rat is
a poor model for the fate of arsenic in man.
8
-------
Dutkiewicz (1977) observed the effects of the route of
administration on the distribution and elimination of pentavalent
arsenic in rats. Both intravenous and intratracheal administra-
tion of sodium arsenate caused immediate distribution of arsenic
to most tissues, whereas skin application and gastrointestinal
administration caused much lower arsenic concentrations in
tissues. These data are summarized in Table 3.
Peoples (1964) fed arsenic acid (pentavalent arsenic) at
levels of 0, 0.05, 0.25, and 1.25 mg/kg daily for 8 weeks to
lactating cows in order to determine whether arsenic appeared in
their milk. After 8 weeks the arsenic levels in tissue were low,
even in cows receiving the highest doses. Highest concentrations
were in the liver, kidneys, and spleen. Analysis of the blood
showed no increase in arsenic. Lack of arsenic storage in the
tissues was explained by the arsenic acid content of the urine,
which nearly equalled the daily intake. Peoples concluded that
the tissues store little arsenic and that "these low levels are
rapidly depleted . . . and represent a 'transit1 period ra'ther
than true storage of arsenic." Peoples found that all tissue-
bound arsenic was in the pentavalent form and that none was
reduced to the trivalent form. No arsenic appeared in the milk,
an indication of a blood-mammary barrier to arsenic in cows.
Mealey et al. (1959) administered radioactive arsenic
trioxide to 11 patients dying of intracranial disease and found
that arsenic accumulated mainly in the liver, kidneys, and spleen
and disappeared rapidly from the blood.
METBOLISM
It is generally accepted that trivalent arsenic is largely
oxidized and excreted in the urine in the pentavalent form
(Overby and Frederickson, 1963; Peoples, 1964; Schroeder and
Balassa, 1966). Winkler (1962) examined the livers of rats fed
with sodium arsenite and found that most of the arsenic was
pentavalent. While oxidation of trivalent arsenic appears to
take place in vivo, it is not clear x^hether the reverse reaction
9
-------
TABLE 3. EFFECT OF ROUTE OF ADMINISTRATION ON DISTRIBUTION
AND EXCRETION OF PEMTAVALENT ARSENIC IN RATSa
Route of
administration
Percentage of dose/g tissue at various
times after administration
Intravenous
Intratracheal
Gastrointestinal
Skin resorption
Intravenous
Intratracheal
Gastrointestinal
Skin resorption
1 h
2 h
5 h
24 h
120 h
240 h
Distribution in liver
3.85
2.49
0.34
3.73
2.59
0.27
3.47
3.25
0.32
2.75
3.08
0.26
0.066
2.17
1.69
0.19
0.36
1.73
1.02
0.37
0.23
Distribution in spleen
1.60
0.71
0.19
1.96
1.07
0.40
2.15
1.55
0.65
1.72
1.69
1.07
0.29
1.90
2.48
0.78
0.63
1.70
2.75
1.20
0.96
Elimination in urine and feces after
10 days, % of absorbed dose
Intravenous
Intratracheal
Gastrointestinal
Skin
Urine
39
35
20
30
Feces
5
15
30
30
Taken from Dutkiewicz (1977).
10
-------
occurs, i.e., reduction of the pentavalent to the trivalent form.
When sodium arsenate and sodium arsanilate (both pentavalent)
were fed to rats, the livers contained only pentavalent arsenic
(Winkler, 1962). Peoples (1964) also found only pentavalent
arsenic in the urine of cows fed arsenic acid (pentavalent).
Ginsburg (1965) found, however, that in dogs arsenate was in part
reduced to arsenite, which appeared in the urine and blood. It
has been suggested that such reduction of the pentavalent organic
arsenicals used as medicines may be necessary for their medicinal
effects (Vallee et al., 1960) .
Urinary excretion following human ingestion of known arsenic
species shows that biomethylation occurs. When sodium arsenate
and potassium arsenite were fed to cows and dogs, more than 50
percent of the trivalent and pentavalent inorganic arsenic was
methylated in the urine (Lakso and Peoples, 1975). After inges-
tion of arsenite--rich wine, approximately 10 percent of the
arsenic was excreted as arsenite, but most of the arsenic was
excreted as methylarsonic acid and dimethylarsinic acid. Of 63
jig of arsenic ingested, approximately 80 percent was excreted in
the urine within 61 hours. After ingestion of arsenate-rich
water, analyses showed higher levels of both arsenate and di-
methylarsinic acid in the urine. After ingestion of crabmeat
(containing an as-yet-unidentified organo-arsenic compound),
elevated levels of dimethylarsinic acid were observed, but only
after the urine had been heated in 2N NaOH (Crecilius, 1977).
BIOCHEMISTRY AND TOXICITY
The toxicity of arsenic compounds varies with their valence
state and structure. On the basis of experimental studies with
animals, trivalent arsenic appears to be considerably more toxic
than pentavalent (Webb, 1966; Luh et al., 1973). Pentavalent
arsenic does not readily bind to tissues and is excreted rapidly
in the urine, whereas trivalent arsenic binds readily and is
excreted more slowly (Webb, 1966). Toxicity is thought to be a
function of binding of various arsenicals (Vallee et al., 1960).
11
-------
Inorganic arsenic has been said to be generally more toxic than
organic arsenic, but Webb, in an exhaustive review of the bio-
chemistry of arsenic compounds, notes that certain organic com-
pounds (lewisite and phenylarsenoxide) are 50 to 1000 times more
potent than arsenite (Webb, 1966, p. 655).
Trivalent arsenic (arsenite), can, owing to its tetrahedral
structure with a lone electron pair, complex with hydroxyl groups
in biologically important monosaccharides and in catechol (which
occurs in epinephrine). Pentavalent arsenic, on the other hand,
was found not to complex with hydroxyl groups (Roy et al.,
1957). Trivalent arsenic has also been found to inhibit the
activity of many enzymes containing sulfhydryl groups by binding
to the sulfhydryl groups, with which it has been postulated to
form an arsenical-thiol mercaptide (Vallee et al, 1960).
It is thought that such sulfhydryl binding occurs with
lipoic acid, thereby inhibiting the pyruvate oxidase system. By
blocking pyruvate oxidation, arsenite interferes with cellular
respiration, the main energy-producing process in aerobic cells.
An extreme example of such binding is the war gas, lewisite,
which binds to two thiol groups in lipoic acid to form a stable
ring. Inhibition of pyruvate oxidase probably accounts for some
of the most obvious acute effects of arsenic poisoning.
The reaction between lipoic acid and lewisite can be repre-
sented as:
/c'
\
-SH
H.
+ Cl2AsCH:CHCl-
AsCHrCHCl
H
C
-SH
'H
C
(CH2)4
COOH
Lipoic acid Lewisite
(CH2)4
COOH
12
-------
British antilewisite (BAL), which resembles lipoic acid, binds
more readily to arsenicals and forms an even more stable five-
member ring with lewisite. Thus it can be used as a remedy for
arsenic poisoning.
H0C SH H_C
2| 2|
HC SH + Cl^AsCH:CHCl HC
H2C OH H2C OH
BAL ' Lewisite
In addition to pyruvate oxidase, some of the other enzyme
systems inhibited by trivalent arsenic are: cholinesterase,
deoxyribonuclease, d-amino acid oxidase, 2-glutamic acid oxidase,
monoamine oxidase, transaminase, liver choline oxidase, choline
dehydrogenase, glucose oxidase (Buchanan, 1962). An exhaustive
list if given by Webb (1966).
Owing to its similarity to phosphorus, it has been suggested
that arsenic, in both its trivalent and pentavalent states,
competes with phosphate in many enzymatic reactions. Arsenite
uncouples oxidative phosphorylation. Arsenate can replace
phosphate in phosphoglyceraldehyde dehydrogenase (Vallee et al.,
1960). The arsenical esters formed in these reactions are un-
stable and are immediately hydrolyzed. This phenomenon is
known as "arsenolysis." The uncoupling by arsenite has been
explained as follows: "the unstable arsenlyated oxidation product
rapidly undergoes irreversible hydrolysis and allows oxidation to
proceed at an increased rate, but without the formation of the
high energy phosphate bond" (Vallee et al., 1960). The reaction
can be represented as:
f 1 + arsenate
glucose-arsenate + enzyme
+ H20
glucose + arsenate
13
-------
Rosen (1971) has suggested that arsenic can substitute for phos-
phorus in DNA. It has been hypothesized that arsenic inhibits
dark repair in human epidermal cells by binding to DNA-polymerase
(Jung and Trachsel, 1970).
14
-------
SECTION 3
EXPERIMENTAL STUDIES
CARCINOGENICITY
In an EPA-commissioned project, the National Academy of
Sciences (1977) has reviewed the data on the carcinogenicity of
arsenic; the Academy concludes that: "In general, animal studies
have not shown carcinogenicity for arsenic compounds, even when
administered at near the maximally tolerated dosage for long
periods." The important experimental studies of carcinogenesis
and cocarcinogensis with arsenic are reviewed by NAS (1977) and
IARC (1973) and are summarized in Tables 4 and 5.
The NAS ci,tes two exceptions to the generally negative
findings. One is a study by Halver (1962), reviewed by Kraybill
and Shimkin (1964), in which 50 trout were exposed to carbasone
at 480 mg/100 g of diet; five developed-heptomas, whereas no
heptomas were found in a large control group fed the same diet
but without carbasone. The second exception, a study by Osswald
and Goerttler (1971), reported a considerable increase in the
incidence of leukemia in female Swiss mice given daily subcuta-
neous injections of sodium arsenate during pregnancy and in their
offspring. Pregnant mice were given a total of 20 injections,
each containing arsenic at 0.5 mg/kg of a 0.005 percent aqueous
sodium arsenate solution. Some groups of offspring from the
treated females were given an additional 20 subcutaneous injec-
tions of arsenic (0.5 mg/kg) at weekly intervals and others were
left untreated. Eleven of 24 mothers (46%) developed leukemia or
lymphoma within 24 months after the beginning of the experiment,
whereas none of 20 untreated mothers that died during the 2-year
period developed lymphoma.
15
-------
TABLE 4. EXPERIMENTAL CARCINOGENESIS STUDIES WITH ARSENIC
Study
Hueper and
Payne (1962)
Baronl et al.
(1963)
Knoth (1966)
Kanisawa and
Schroeder
(1967)
Fa i Hi all and
Miller
(1941)
Hueper and
Payne (1962)
' Byron et al .
(1967)
Kanisawa and
Schroeder
(1969)
Species
Rat
House
Mouse
Mouse
Rat
Rat
Rat
Rat
Route
of admin.
Oral
Oral
Oral
Oral
Oral
Oral
Oral
Oral
Compound
Arsenic trloxlde
Arsenic trloxlde
Fowler's solution
(potassium arsenlte)
Sodium arsenlte
Lead arsenate
Calcium arsenate
Arsenic trloxlde
Sodium arsenlte
Sodium arsenate
Sodium arsenite
Concentration
0.0004% solution of
122 aqueous ethanol In
drinking water
O.OU In drinking
water
Total dose of 7 mg As
as As .O-
5 mg As/ml
10 mg/day
Equivalent amount
0.00042 1n drinking
water
0.0004-0.00342 In 122
aqueous ethanol
10, 15.6. 31.2. 62.5
125, 250 ppm As as
. sodium arsenlte
0. 31.2, 62.5. 125.
250, 400 ppm As as
sodium arsenate
5 ppm
Frequency
Dally
1 drop/week
Ad libitum
Ad libitum
Duration
15 months
60 weeks
5 months
Llfespan of the
mice
Up to 2 years
22-24 months
2 years
Lifespan
Results
No excess tumors
No excess tumors
Adenocarclnomas of the
skin, lung and lymph
nodes; tumors were ob-
served In offspring
No tumors
No tumors
No tumors
No excess tumors
No excess tumors
"Pretumorous lesions of
the liver" In treated
rats
(Ti
(continued)
-------
TABLE 4 (continued)
Study
Leitch and
Kennaway
(1922)
Osswald and
Goerttler
(1971)
Osswald and
Goerttler
(1971).
Heuper (1954)
Holland and
Acevedo
(1964)
Species
Mouse
House
Mouse
Rat
Rabbit
Rabbit
Route
of admin.
Skin ap-
plications
Subcuta-
neous
Intra-
muscular
Intra-
venous
Intra-
medullary
injection
in the
femur
Inhala-
tion
Compound
Potassium arsenite
In alcohol (containing
1.8% arsenous oxide;
later reduced to 0.12%)
Sodium arsenate
Sodium arsenate
Metallic arsenic
in lanolin
Arsine gas
Concentration
0.6 ing/kg body weight
as 0.005% aqueous
solution of sodium
arsenate
0.5 mg As (as 0.005%
^solution of sodium
arsenate)
0.43 mg arsenic
0.64 mg
Small doses
Frequency
3 times/week
Dally throughout
pregnancy
Meekly
Single Injection
Dally
Duration
3-5.5 months
2-year observa-
tion period
20 weeks
20-26 months
Results
Negative
11 treated mice (45%)
developed lymphocytic
leukemia or lymphoma
13 of the 71 untreated
progency of the arse-
nate-mothers developed
lymphoma during a 2-year
observation period com-
pared with 41 of 97
progency given weekly
subcutaneous Injections
of 0.5 mg/kg body weight
(as sodium arsenate)
11 of 19 mice had
lymphoma
1 sarcoma at site of
Injection
No tumors
One rabbit developed a
malignant mesothelioma
of the pleura after 17
months exposure
-------
TABLE 5. EXPERIMENTAL STUDIES OF COCARCINOGENESIS, TUMOR PROMOTION, AND INITIATION
00
Study
Salaman
and Roe
(1956)
Baron 1
(1963)
Boutwel 1
(1963)
Species
House
House
House
Route of
admin.
Skin
applica-
tion
Oral
Skin
applica-
tion
Oral
Skin
applica-
tion
Compound
Potassium
arsenite
Arsenic
trioxide
Sodium
arsenate
Potassium
arsenate
i
Concentration
IX solution In
methanol
O.OU
solution
2.4 mg
KAs02/
mouse
676 mg
KAs02/kg
of diet
169 mg
KAsOp/kg
diet*
1.24 mg/
mouse
2.2 rag/
weeks
Frequency .
duration
Once/week
for 10
weeks
40-60
weeks
Over
5-day
period
2 weeks
48 weeks
Over
5-day
period
29
weeks
Promoter or
initiator
Promoter
0.17% or
0.085%
croton oil
in acetone
Croton oil ,
OMB A, or
urethane
Initiators:
urethane
CUBA
Promoter:
croton oil
Prompter :
croton oil .
Initiator:
DMBA
croton oil
Promoter:
croton oil
In benzene
Initiator:
DMBA
Results
Three treated mice
developed skin
paplllomas, but
controls receiving
only croton oil
also developed
skin tumors
No excess tumors
No excess tumors
In test mice
receiving potas-
sium arsenite
compared with
control mice
receiving only
croton oil or
DMBA + croton oil
(continued)
-------
TABLE 5 (continued)
Study
Milner
(1969)
Kroes et
al. (1974)
Spectes
Mouse
Rat
Route of
admin.
Oral
Oral
Compound
Arsenic
trioxide
Lead
arsenate
Sodium
arsenate
Concentration
0.01* in
drinking water
463 and 1850 ppm
in diet
416 ppm in diet
Frequency,
duration
4-13
weeks
Up to 29
months
Promoter or
initiator
Initiator:
MCA
DENA
Results
No enhancement
of skin cancer
No evidence of
carcinogeniclty
of sodium arse-
nate; inconclu-
sive evidence for
lead arsenate; no
enhancement of
carcinogenic
effect attributable
to DENA
OMBA - dimethylbenz (a) anthracene;
MCA ° methylocholanthrene;
ENA •= dlethylnitrosamine.
-------
Among the untreated offspring, 7 of 34 males (21%) and 6 of
37 (16%) females developed leukemia. Among the treated off-
spring 17 of 41 males (41%) and 24 of 50 females (48%) developed
leukemia. Eleven of 19 treated offspring (58%) also developed
lymphoma. Among untreated control mice only 3 of 35 male (9%)
and none of 20 female offspring developed leukemia.
IARC (1973) comments that this study is "difficult to inter-
pret since 20 out of the 55 control animals and some of the
experimental animals were still alive at the date of reporting.".
The NAS criticism that- the credibility of this study is limited
by the "failure to give the vehicle solution to the controls" is
difficult to understand because the sodium arsenate was admin-
istered in aqueous solution.
Two other reports that may indicate positive effects are
those of Knoth (1966) arid Kanisawa and Shroeder (1969).
Knoth (1966) reported in a preliminary study that oral
administration of one drop of Fowler's solution (containing
potassium arsenite) per week for 5 months to groups of 30 mice
(equivalent to a total dose of 7 mg calculated as As«O_) caused
a significant increase in the number of tumor-bearing mice.
Tumors including adenocarcinomas of the skin, lung, and lymph
.nodes were observed at 14 months, the time at which the experi-
ment was ended. Fifteen control mice showed no tumors. Off-
spring of the treated mothers also showed some tumors, whereas
the offspring of the control mice showed no tumors. Knoth1s data
are difficult to interpret because the preliminary report is very
brief and the followup study he mentioned was not published.
Kanisawa and Schroeder (1969) fed 91 Long Evans rats sodium
arsenite in their drinking water at a concentration of 5 ppm over
their lifespan. The incidence of tumors in treated rats was
similar to that in untreated controls. The authors note, how-
ever, that 19 of 91 rats given arsenic developed unspecified
"pretumorous lesions" of the liver, whereas only 10 of 82 con-
trols developed such lesions. IARC (1973) called attention to
the very low level of arsenic used in this experiment.
20
-------
Thus, some evidence can be taken to indicate that arsenic
causes cancer in test animals. These interesting leads should be
followed-up, refined, and replicated in order to fully substan-
tiate the experimental carcinogenicity of arsenic. The NAS
(1977) discusses factors to be considered in future efforts to
find an experimental model for arsenic carcinogenesis.
MUTAGENICITY
Both point mutational and chromosomal tests indicate that
trivalent and pentavalent inorganic arsenic are mutagenic.
Furthermore, both forms of arsenic appear to interfere with DNA
repair in a variety of cell types, and sodium arsenate has this
effect in human epidermal cells. These studies are summarized in
Table 6.
Jung et al. (1969) found that exposure of living human
epidermal cells to inorganic arsenic causes a temporary reduction
in enzymatic repair of DNA. They hypothesized that prolonged
inhibition by arsenic of repair of DNA damage caused by ultra-
violet radiation, chemical agents, or viruses could lead to
scattered lesions, which "may constitute the starting point of
carcinogenic changes."
Burgdorf (unpublished Master's Thesis, 1977) studied sister
chromatid exchange (the exchange of genetic material between the
two chromatids of a single chromosome during mitosis) in six
patients with histories of arsenic use and with multiple skin
malignancies and other signs of chronic arsenicism. In all six
the frequencies of sister chromatid exchange were significantly
elevated. Burgdorf notes, however, that the meaning of elevated
sister chromatid exchange is not yet understood. It may reflect
either DNA damage or successful DNA repair.
Paton and Allison (1972) tested the effects of sodium
arsenite, sodium arsenate, and acetylarsan in cultures of human
leukocytes and diploid fibroblasts. Subtoxic doses of the arse-
nic compounds were added to leukocyte cultures and fibroblast
cells at various times between 2 and 24 hours of fixation. A
21
-------
TABLE 6. SUMMARY OF STUDIES ON INORGANIC ARSENIC MUTAGENICITY'
to
Study
Dominant lethal
effects
DNA repair
(Mlcroblal)
(Hanina 11 an)
Metabolic and
cellular toxlcity
effects
(In vivo)
Author) s)
Sram and Bencko (1974)
Rossman et al. (1975',
' 1976)
Jung et al.(1969)
Jung and Trachsel (1970)
Skipper et al.(1951)
Sawada and Rebhun (1969)
Chemlcal(s)
Sodium arsenlte
Sodium arsenlte
Sodium arsenate
Sodium arsenate
l
Potassium
arsenlte
Sodium arsenlte
Cell type
1CR-SP
mice
£. coll (uv-
ex posed)
Human epidermal
cells (exposed
to xenon lamp)
Human epidermal
cells (exposed
to xenon lamp)
Mice
Annelid worm
eggs
Effects
An acute oral dose (250 mg/kg) given to the male
produced negative results. Chronic application
over 4 generations of males at 10 mg/ liter of
drinking water gave positive results (Increase In
overall dominant and prelmplantatlon lethality);
subtoxic dose (100 mg/Hter) gave no change In
frequency of dominant lethality or male fertility
Decreased mutation frequency of Irradiated
cells that were excision-repair deficient;
decreased survival rate of uv-exposed cells
(wild type, excision-repair deficient); no effect
was found on strain deficient 1n post-replication
repair
DNA dark repair activity decreased In presence
of sodium arsenate
DNA synthesis (reduced mitotlc index) and DNA
dark repair both Inhibited in presence of
sodium arsenate
Reduced Incorporation of C-formate (precur-
sor of mouse DNA-purlnes) into nucleic acid
purines
Inhibition of the formation of the mitotlc
apparatus (before fertilization)
(continued)
-------
TABLE 6 (continued)
to
u>
Study
Point mutation
Chromosomal effects
In vitro
In vivo
cytogenetlcs
Sister chromatld
exchange
Author(s)
Ffcsor and Piccolo (1972)
Moore (1976)
Nlshloka (1975)
Casto (1977 letter)
Oppenhelm and
Flshbeln (1965)
Patton and
Allison (1972)
Petres et al. (1970)
Burdorf et al.
(1977)
Chemical (s)
Sodium arsenate
Arsenic trloxlde
Sodium arsenate
Sodium arsenlte
Sodium arsenlte
Potassium
Arsenlte
Sodium arsenlte
Sodium arsenlte
Sodlunvarsenate
Unspecified
(probably
potassium
arsenlte and
lead arsenate)
KAS02
(Fowler's
solution)
Cell type
E. Coll
S. typhlmurlum
Bad Ins subtlHs
Bacllus subtil is
and E. coll
Hamster lung
cells
Human peripheral
leukocytes
Human dlplold
flbroblasts
Human dlplold
Flbroblasts and
leukocytes
Human leukocytes
(from psoriasis
patients and
winegrowers)
Human lymphocytes
Effects
No reversions Induced
Negative Ames' test (without metabolic activation)
Positive mutagen
Positive mutagen > effect stronger than sodium
arsenate
Positive mutagen •> effect stronger than sodium
arsenate
Positive mutagen; Increased the frequency of
mutation In 8-asaguanlne resistant cells
Chromosome gaps, breaks translocatlon,
dlcentrlcs, rlngforms; cell division
suppressed; Increased number of broken
metaphase plates
Chromatld breakage
Chromatld breakage > effect higher with sodium
arsenlte
Chromatld breakage > effect higher with sodium
arsenlte
Compared with controls, arsenic treatment
yielded a much higher percentage of secondary
chromosomal constrictions, achromatic lesions,
chromosome gaps, chromatld breaks, acentric
fragments, dlcentrlc chromosomes, and aneuploldy
Compared with controls, patients treated with
Fowler's solution had elevated rates of sister
chromatld exchange
(continued)
-------
TABLE 6 (continued)
to
Study
(In vitro)
Author(s)
Tsuda (1974)
Petres and Hundelker
(1968)
Petres and Berger
(1972)
Petres et al. (1974)
Petres et al. (1975)
Baron et al. (1975)
Chemical (s)
Sodium arsenate
Sodium arsenate
Sodium arsenate
Sodium arsenate
Sodium arsenate
Sodium arsenate
Cell type
House
flbroblasts
Human
peripheral
lymphocytes
Same
Same
Same
Same
Effects
Indices for prophase, metaphase, and
mitosis Increased, but Indices for
telophase decreased
Number of mitoses reduced to 1/5
the expected number; chromosome pulver-
ization occurred consistently
At high concentrations, the mitotic index
decreased; metaphase plate pulver-
izations at low doses; inhibition
of 3H-thym1d1ne incorporation into
ONA
Inhibition of 3H-thymid1ne and
14C-thym1dine incorporation into DNA
at lower doses than Inhibition of 3H-ur1d1ne
and 14c-uridine Into RNA
After 1 h of exposure, Increasing
inhibition of 14c-thymid1ne incorpor-
ation into DNA
Inhibition of 14C-TTP Into DNA and
HC-UTP RNA; Inhibition of corporation
of labeled alanine and leuclne Into
cellular proteins
Source: EPA Carcinogen Assessment Group.
(continued)
-------
significantly increased incidence of chromosome breakage was
-9
found in leukocytes treated with sodium arsenite (2.9 x 10 to
— 8 — 8
1.8 x 10 M) and acetylarsan (6.0 x 10 M) in the last 48 hours
of the culture period. In the leukocytes treated with sodium
arsenite 60 percent of 148 metaphases examined had chromatid
breaks. In leukocytes treated with acetylarsan, 20 percent of 50
metaphases observed had chromatid breaks. Sodium arsenate,
however, administered in the highest nontoxic concentration,
caused no significant increase in the number of breaks. Diploid
— 9
fibroblast cultures exposed to sodium arsenite (2.9 x 10 to
_ Q
5.8 x 10 M) in the last 24 hours of culture showed a signifi-
cantly increased incidence of chromosomal damage.
Beckman et al. (1977) looked at chromosomal aberrations in
short-term cultured lymphocytes from nine employees exposed to
arsenic at the Ronnskar smelter works. In lymphocytes from
smelter workers there were 87 aberrations in 819 mitoses. The
frequency of aberrations was significantly higher (p<0.001) among
the arsenic-exposed workers than among controls. Individual
variations ranged from 0 to 25 aberrations per 100 cells. Since
the arsenic-exposed workers also experienced concomitant exposure
to other chemicals, it could not be determined whether arsenic
alone or arsenic in conjunction with other agents was respon-
sible.
Petres et al. (1977) examined lymphocytes from 31 patients
at a dermatological clinic. These patients had been exposed to
arsenic and displayed the characteristic arsenical hyperkeratoses
of the hands and feet. Thirty-one people with no known arsenic
exposure served as a control group. The group with arsenic
exposure showed a frequency of chromosomal aberrations signifi-
cantly above that of the coiitrols. The frequency of chromatid
breaks was especially high (34 times greater than in the con-
trols) . Petres et al. also found that in vitro addition of
sodium arsenate to cultured lymphocytes from healthy subjects
induced the same chromosome changes that were found in the
exposed group. Radioactive incorporation' studies indicated that
25
-------
arsenate inhibited the incorporation of radioactively labelled
nucleotides in RNA and DNA, and the inhibitory effect increased
with increased dose. Finally, Petres et al. noted that arsenic
blocked the lymphocytes in the S- and G~ phases. Petres et al.
suggested that arsenic interferes with enzyme systems involved in
nucleic acid metabolism by binding to the sulfhydryl groups in
certain enzymes and possibly by uncoupling phosp^hprylation or
substituting for phosphorus in nucleic acids.
Rossman et al. (1977) showed that the presence of sodium
arsenite significantly decreased survival of wild type E. coli
after ultraviolet irradiation, an indication that arsenite is
mutagenic in E_. coli normally capable of carrying out post-
replication repair. This finding would support the hypothesis
that arsenite may act as a cocarcinogen by interfering with DNA
repair, although no strong experimental evidence of such cocar-
cinogenicity is available.
Administration of arsenic together with the referential
mutagen (chemosterilant TEPA, tris(1-aziridinyl) phosphine oxide)
resulted in a significant increase in dominant lethal mutations
in F, generation mice. This effect, seen only at high dosages,
has been explained by the possible interference of arsenic in
chromosome repair by its blocking sulfhydryl groups (Bencko,
1977).
Negative results have been reported (Hodge, 1977) in tests
for a dominant lethal effect by administering (intraperitoneally)
to male mice single doses of sodium arsenate (5 mg/kg), sodium
arsenite (5 mg/kg), sodium cacodylate (200 mg/kg), arsenoacetic
acid (50 mg/kg), methane arsenic acid (250 mg/kg), and a compo-
site flue dust (2 mg/kg). The implications of a negative dom-
inant lethal test are thought to be limited.
TERATOGENICITY
Several investigators have shown that sodium arsenate in-
duces developmental malformations in a variety of test animals:
26
-------
embryo chick, hamster, rat, and mouse (Ancel, 1946; Ridgeway and
Karnovsky, 1952; Ferm and Carpenter, 1968; Hood and Bishop, 1972;
Beaudoin, 1974).
Pregnant golden hamsters injected with sodium arsenate (15
to 25 mg/kg body weight) produced offspring with a range of
developmental malformations including anencephaly, renal agenesis,
rib malformation, cleft lip and palate, and anophthalmia. The
percentages of living embryos with various selected malformations
followed maternal treatment with 20 mg/kg sodium arsenate on the
8th day of gestation were as follows: nearly 90 percent with all
malformations; over 80 percent with anencephaly; nearly 70 per-
cent with rib malformations; 30 percent with exencephaly. The
spectrum of malformations varied with the time of injection
during critical stages of embryogenesis. Malformations induced
by arsenate differed from those induced by other teratogenic
agents including certain heavy metals (Ferm et al., 1971).
In another study, single intraperitoneal injections of
sodium arsenate (45 mg/kg) in Swiss-Webster mice between the 6th
and llth days of gestation consistently caused an increase in
fetal resorptions, a significant decrease (p<0.05) in fetal
weights compared to controls, and a number of fetal malforma-
tions, most frequently the following: exencephaly, shortening of
the jaws with consequent protrusion of the tongue, exophthalmos,
missing pinna, cleft lip, hydrocephalus, umbilical hernia, even-
tration, ectrodactyly, micromelia, and shortened or twisted tail
or limb, or both. Malformations were dependent on the stage of
embryogenesis. Exencephaly occurred in 54 percent of the fetuses
when the injection was administered on day 9 of gestation;
fusion of the ribs occurred in 100 percent of the fetuses when
the injection was given on day 9; and fusion of the vertebrae
occurred in 73 percent when the injection was given on day 10
(Hood and Bishop, 1972).
In a later report, Ferm (1977) demonstrated that administra-
tion of 20 mg/kg of sodium arsenate intravenously or intraperi-
toneally to Golden hamsters during day 8 to 9 of gestation
27
-------
induced a specific spectrum of malformations including exen-
cephaly, encephaloceles, skeletal defects, and malformations of
the genitourinary system. The last effect, which appears to be
unique to arsenate occurred in both sexes and with high fre-
quency.
74
Perm (1977) further showed that sodium arsenate C As)
injected intravenously into Golden hamsters on day 8 of gestation
was transmitted across the placenta during the critical stage of
embryogenesis and appeared in the fetal tissues. Ferm refers to
a report by Lugo et al. (1969) concerning a case of arsenic
trioxide poisoning during human pregnancy, which demonstrated the
"ease with which inorganic arsenic crosses the human placenta at
term with extremely high levels in the fetal liver, brain, and
kidneys" (Ferm, 1977). Introduction of arsenic into fertilized
bird eggs has led to malformations of beak and brain (Peterkova
and Puzanova, 1975).
Hood et al. (1977) compared the prenatal effects of oral and
intraperitoixeal administration of sodium arsenate in mice.
Intraperitoneal administration had a considerably greater effect
than oral administration on prenatal mortality, reduction of
fetal weights, and occurrence of fetal malformations. The dos-
ages were 40 mg/kg (intraperitoneal) and 120 mg/kg (oral).
Hood et al. further noted that although arsenite is con-
siderably more toxic than arsenate, it has received less atten-
tion from teratolegists. Intraperitoneal injection of mice in
utero with 10 to 12 mg/kg of sodium arsenite on one of days 7 to
12 of pregnancy caused significant increases in prenatal mor-
tality (p<0.05), and treatment on days 8, 9, and 10 resulted in
gross and skeletal malformations similar to but less frequent
than those induced by comparably toxic levels of arsenate (Hood
et al. , 1977) .
28
-------
SECTION 4
EFFECTS OF HUMAN EXPOSURE TO ARSENIC
We have noted that the respiratory tract represents a major
route of entry for arsenical compounds. Much of the evidence
regarding the toxic potential of inhaled arsenic derives from
data on occupationally exposed populations, although epidemiolo-
gical investigations of the relative cancer rates in communities
near arsenic smelters have also been reported. Epidemiological
and clinical studies of worker groups and of communities located
near smelters are presented below.
The National Institute of Occupational Safety and Health
(NIOSH) has estimated that 1.5 million workers are potentially
exposed to inorganic arsenic (National Institute of Safety and
Health, 1975). Occupational exposure to arsenic occurs prin-
cipally in workers employed in the smelting and refining of
nonferrous ores containing arsenic, in workers employed in pro-
ducing arsenical pesticides and insecticides, and in agricultural
workers using arsenical desiccants and pesticides. A relatively
small number of workers and commmunities is exposed to organic
and inorganic arsenic used as desiccants in cotton-ginning opera-
tions.
Environmental exposure to airborne arsenic may be a cause
for concern in the vicinity of smelters, coal-fired power plants,
glass-manufacturing plants, and other sources of arsenical air
pollution.
SMELTER WORKERS AND COMMUNITY RESIDENTS
Smelter workers are a population exposed to high levels of
arsenic trioxide fumes in conjunction with high levels of sulfur
29
-------
dioxide. Other air contaminants, such as lead and other heavy
metals, are also present but in smaller quantities. Toxic
exposures of smelter communities and workers have been much
greater in the past than they are now (Nelson, 1977), so that
epidemiological evidence relating to workers with a long history
of smelter work is of particular significance. This evidence is
summarized in Table 7.
The working population at the smelter in Tacoma, Washington,
has been studied extensively over the years. ASARCO, which
operates the smelter, has collected and published a great deal of
data on atmospheric and biological exposure levels; the series of
reports based on these studies represent an important contribu-
tion to the available knowledge about the effects of arsenic on
humans. The papers are especially interesting when presented in
chronological order because they illustrate the changes in
recognition of the effects of chronic arsenic exposure. Milham
and others not employed by ASARCO have also studied this popula-
tion, and their results are relevant.
In early study Pinto and McGill (1953) reported the urinary
arsenic levels of 348 workers exposed to arsenic trioxide dust at
the Tacoma smelter and concluded from physical examinations of
the population that the urinary arsenic levels were not asso-
ciated with systemic arsenic poisoning. The values ranged from
0.10 to 6.44 mg arsenic/liter. The average urinary arsenic level
for the whole group was 0.82 mg/liter. In 147 men actively
working in the industry, but considered by Pinto and McGill not
to have exposure to arsenic trioxide, the average urinary level
of arsenic was 0.13 mg/liter.
Pinto and Bennett (1963) reported on an investigation of the
causes of death among 229 active smelter workers and retirees
from the Tacoma plant over the period 1946 to 1960 and calculated
proportional mortality rates (PMR). The PMR1s calculated for the
male population of the State of Washington between the ages of 15
and 95 in the year 1950 served as a reference rate. Among the
smelter workers 43 cancer deaths were observed whereas 36.7 were
30
-------
TABLE 7. SUMMARY OF EPIDEMIOLOGICAL STUDIES OF SMELTER WORKERS
Study
Pinto and McGIll
(1953)
Pinto and
Bennett (1963)
Mil ham and Strong
(1974)
Leg and Fraumenl
(1969)
Kuratsune et al.
(1974)
Mil by and H1ne
(unpublished.
1974)
Type of study,
and period of
observation
Clinical
Proportionate mor-
tality. 1946-1960
Proportionate mor-
tality, 1950-1971
Retrospective cohort
analysis. 1938-1963
Case- control, 1917-
1965
Proportionate mor-
tality. 1950-1972
Process
Smelter producing
ASoO, as byproduct
C J
Smelter workers at
same plant as Pinto
and McGIll
Smelter workers at
same smelter as
Pinto and McGIll
Smelter workers with
exposure to As,0,
£• •)
Same company as
studies by Rencher
and Carter (Utah
smel ter)
Number 1n cohort
229 deaths reported
among 904 active
plant employees and
209 pensioners
>241 deaths
1877 deaths among
8047 white male
smelter workers
Case group: 19 males
who died of lung
cancer; 19 males who
died of diseases
other than lung,
urinary, bladder, or
skin cancer
1910 deaths among
persons who had
worked at least 10
years with company
(Including mine, con-
centrator, refinery)
with no Increased
risk according to
Rencher and Carter
Findings
Smelter workers had In-
creased Incidence of deaths
due to lung cancer as a
proportion of cancer deaths
(41.9* vs. 23. 7* 1n the
state as a whole)
Increased lung cancer; 40
observed deaths, 18 ex-
pected (p< 0.001)
6.7, 4.8, and 2.4-fold lung
cancer excess for heavy,
medium, and light exposure
11 lur\g cancer deaths oc-
curred In former copper
smelter workers vs. only
3 deaths In former copper
smelters In the control
group (p=0.01)
No excess lung cancer
Exposure data
Urine levels; but "nonexposed"
smelter workers might have had
some exposure to arsenic
Urine levels
Heavy, medium, and light ex-
posure groups (concomitant ex-
posure to SO.)
Exposure to SO. and PAH
£
(continued)
-------
TABLE 7 (continued)
u>
NJ
Study
Tokudome and
Kuratsune (1976)
Pinto, Enter) ine.
Henderson, and
Varner (1977)
Rencher and Carter
(1971)
Type of study,
and period of
observation
Retrospective cohort.
1949-1971
Historical prospec-
tive, 1949-1973
Retrospective
1959-1969
Process
Metal refinery
Copper smelter
workers with ex-
posure to As.O..
C J
Copper smelter
workers
Number 1n cohort
157 deaths among 839
•copper smelter workers
324 deaths among 527
male retirees from
copper smelter
651 deaths
l
'
Findings
Significant excess mortality
for lung cancer (SMR=1189)
among smelter workers with
heavy exposure for 15 or
more years; this exposure
reached SMR=25f>n
Significant excess mortality
for all causes (SMR=112.2)
for cancer (SHR=148.4) for
lung cancer (SMR=304.8).
Highest exposure category
SMR=833.3
Smelter workers exhibited
highest % of deaths from
lung cancer (7.0% based on
17 deaths); both smoking
and nonsmoking smelter
workers experienced a
higher relative frequency
of lung cancer deaths
than their counterparts at
the mine and the concentra-
tor
Exposure data
Dose-response relationship
demonstrated between lung can-
cer mortality and degree of
exposure
Measurements of urinary arsenic
for all plant workers showed
direct correlation between air-
borne As and urinary As values;
time-weighted Index of total
lifetime exposure to As was
linearly related to respiratory
cancer mortality
Based on average exposure in
each of 12 work areas and amount
of time worked in each of these
areas, 5 exposure Indices (for
SO? sulfuric acid mist, arsenic.
lead, copper) were computed for
each worker and averaged over
the number of persons in each of
the three categories of cause of
death; all 5 of these average
cumulative exposure Indices were
substantially higher for lung
cancer group, indicating that
these persons had either worked
longer In the smelter or in
areas of higher exposure to the
contaminants than persons dying
of other causes; average hourly
exposure level for the 12 work
areas ranged from a reported 0.
in the engineering building and
warehouse to 22.0 vjg/nr In the
reverberating furnace area;
overall average was 7.38 wg/mj
-------
expected. For lung cancer the rates were 41.9 percent (18 out of
43) for smelter workers and 23.7 percent for Washington males.
The significance levels of these calculations are not reported.
When the smelter population was divided into exposed and
f
unexposed workers on the basis of urinary arsenic levels measured
in the Pinto and McGill study, the authors calculated that
unexposed workers had a higher proportion of deaths from all
cancer (37 versus 6) and from lung cancer (15 versus 3) than
exposed workers. The age at death of exposed workers did not
correlate with arsenic exposure. Pinto and Bennett concluded
that the levels of arsenic trioxide to which the workers were
exposed did not cause systemic cancer or cardiovascular disease.
According the Milham and Strong (1974), however, the 39 lung
cancer deaths among the Tacoma smelter workers between 1950 and
1971 represent a significant excess (p<0.001) over the number
expected (18) on the basis of U.S. mortality rates.
A recent study of the mortality of retired Tacoma smelter
workers by Pinto et al. (1977) confirms the Milham and Strong
results. Pinto et al. found evidence of a dose-response rela-
tionship between exposure to airborne arsenic and lung cancer
mortality. The study group consisted of 527 pensioners from the
Tacoma copper smelter. Overall mortality rate of the retirees
was 12 percent higher than that of all Washington males. Mor-
tality due to lung cancer showed the greatest excess (32 deaths
observed vs. 10.5 expected; standardized mortality rate (SMR) =
304.8; p<0.05). Smoking histories obtained for most retirees did
not appear to account for the elevated lung cancer rate.
An "index of exposure" based on urinary arsenic levels was
calculated for each retiree. The "index" did not indicate an
absolute exposure rate, but rather represented the relative
exposure rates among the cohort members. Intensity of exposure
was better correlated with lung cancer mortality than was length
of exposure; a gradient of increasing lung cancer mortality
corresponded to the increasing exposure gradient. Among the
group with highest exposure, the SMR for lung cancer was 833.3.
33
-------
Although the smoking histories that were obtained with
respect to most of the retirees did not appear to be correlated
with the elevated lung cancer rate in this report, Pinto et al.
later reported that among 377 retirees alive on January 1, 1961,
the SMR for the 189 smokers was 287 (p<0.05). The SMR for 119
nonsmokers was 507 (p<0.05) (Pinto et al. , in press).
Pinto et al. (1977) also observed that among workers with
less than 25 years of exposure the lung cancer rates were lower
than those of workers with more than 25 years 1 Also, the risk
for lung cancer decreased with increasing age beyond 65. These
observations are taken by the authors as possible evidence of a
threshold value for arsenic trioxide, below which no adverse
effect may be expected. This conclusion is discussed further in
the health assessment section. The authors concluded that al-
though other airborne contaminants were present in the atmosphere
of the smelter, exposure to airborne arsenic was "closely re-
lated" to the excess lung cancer observed.
Milham (1977) examined the cause of death among 753 Tacoma
smelter workers for the period 1940 to 1976. The study group
included retirees and men who had worked in the smelter, but had
not retired from it and had continued to reside in Pierce County
after termination of employment. Former employees who died out
of the State of Washington were not included.
The rates of deaths from all cancer, cancer of the large
intestine, and lung cancer showed significant excesses when
compared with the rates among Washington males. The respective
PMR1s are 128, 162, and 222 (for all threee, p<0.05). Excesses
in deaths from circulatory diseases and nephritis/nephrosis were
also significant (PMR's = 119 and 122, respectively; p<0.05 in
both cases) .
These investigations and their implications are discussed
more fully in Section 6. The Tacoma smelter epidemiological data
discussed here are summarized in Table 3.
An extensive study of workers in many smelters was carried
out by Lee and Fraumeni (1969), who examined deaths among 8047
34
-------
white male smelter workers with exposure of 1 year or more to
arsenic trioxide during the period 1938 to 1963. The mortality
of the smelter workers was compared with that of the white male
population of the same states by use of the life-table method.
Excess overall mortality among smelter workers was significant
(p<0.01) and was attributable primarily to cancer and cardiac
disease. Among specific causes of death, lung cancer showed a
significant excess over the expected (SMR = 329; p<0.01). Excess
of lung cancer reached 8-fold in workers with more than 15 years
employment and heavy exposure to arsenic. On the basis of air
measurements each work area was classified as involving "heavy,"
"medium," or "light" exposure. In general, excess lung cancer
reflected the degree of exposure to both arsenic and sulfur
dioxide. Distinguishing the effects of arsenic from those of SO-
is not possible because most work areas with heavy arsenic
exposure also involved medium SO- exposure and all jobs with
heavy SO- exposure involved medium arsenic exposure. The highest
excess of lung cancer, however, occurred among those workers with
heaviest exposure to arsenic coupled with moderate or heavy
exposure to SO-. The authors conclude that their findings are
consistent with the "hypothesis that exposure to high levels of
As_O_, perhaps in interaction with SO- or unidentified chemicals
in the work environment, is responsible for the excessive number
of respiratory cancer deaths among smelter workers."
In their study of the mortality at the Kennecott smelter
near Salt Lake City, Rencher and Carter (1977) also report a
significant excess of deaths due to lung cancer among smelter
workers (7.0% of all deaths). The lung cancer mortality among
smelter workers is compared with that of mine workers (2.0%),
concentrator workers (2.2%), and Utah males (2.7%). In addition
to excess lung cancer mortality, a smaller excess due to other
types of cancer was found among smelter workers. Data on smoking
habits were examined, but gave no evidence of an increased risk
for smokers. The excess lung cancer mortality occurred in those
35
-------
with the greatest cumulative exposure to five contaminants: SO2/
H-SO. mist, arsenic, lead, and copper.
All deaths among current and former employees in the 11-year
period from 1959 to 1969 were included in the study. The life-
table method was used in comparing the death rates of smelter and
mine workers with those of State of Utah citizens. Each deceased
employee was classified according to the length of time worked in
each of 12 major work areas, and indices were developed for each
employee for exposure to S02, H-SO. mist, arsenic, lead, and
copper. Information on smoking was obtained for nearly all
deceased smelter workers and for random samples of mine and
concentrator workers. Workers were classified in one of three
groups: smokers, nonsmokers, and those of unknown smoking his-
tory. The proportion of smokers was similar in each of the
operations (approximately 60%). To evaluate the increased risk
of lung cancer among smelter workers with respect to the risk
among smokers, the investigators compared the smelter/mine
mortality ratios with the smoker/nonsmoker mortality ratios for
U.S. males. They found that "any increased risk at the smelter
is far less than the risk due to smoking in the population at
large." V
Rencher and Carter concluded that the excess of lung cancer
mortality at the smelter might be due to high levels of airborne
arsenic prevalent in the plant before 1959. They also noted that
areas of the plant with high levels of airborne arsenic also had
high levels of S0_, similar to the situation reported by Lee and
Fraumeni (1969).
Kuratsune et al. (1974) performed a case-control study of
much smaller scope. The cases consisted of 19 males in one town
who died of lung cancer. Eleven men (58%) had worked as smelters
in a local copper smelter, where they were exposed to high
levels of arsenic. The control group consisted of 19 men who had
died of diseases other than cancer of the lung, bladder, or skin.
The control group had only three smelter workers (15.8%).
36
-------
Smoking habits were similar in the two groups, and the difference
between the two groups was significant (p=0.01).
As a followup of the case-control study, Tokudome and
Kuratsune (1976) conducted a retrospective cohort study of
workers at the smelter in the town. The period of observation
was 1949 to 1971. Observed deaths were compared with expected
deaths based on age- and cause-specific death rates for Japanese
males.- Copper smelter workers had a significant excess of lung
cancer deaths (29 deaths vs. 2.4 expected; SMR = 1189; p<0.01).
In smelter workers with heaviest exposure the SMR for lung
cancer mortality reached 2500. Furthermore, the SMR's for
subgroups of copper workers showed a definite positive gradient
with length of employment, level of exposure, and time when
exposure occurred, although the numbers of deaths in each cate-
gory were small. The authors took this as clear evidence of a
dose-response relationship.
Smoking habits of the smelter workers were not thought to be
different from those of' other workers, and smoking alone could
not account for the magnitude of the increased risk. The authors
concluded that arsenic and sulfur dioxide were probably respon-
sible for the observed excess mortality, although polycyclic
aromatic hydrocarbons were also present in the operations at the
time and cannot be ruled out as contributing to the lung cancer
mortality.
One study that did not show an increased proportion of lung
cancer deaths was carried out by Snegireff and Lombard (1951),
who compared the mortality of workers in a plant (Plant A) that
handled large amounts of arsenic trioxide with that of the popu-
lation of the State. During a 25-year period, 18 deaths from
cancer were reported among workers in the plant. Seven of these
were ascribed to lung cancer. The deaths were not limited to
workers in any specific work area within the plant or to any
occupational group. The authors calculated that the proportion
of cancer deaths among plant workers under 55 (9 cancer deaths
37
-------
out of a total of 62, or 12.5%) was not significantly different
from that for the State (6.1%).
At a second plant (Plant Z, the type of plant is not spec-
ified) in which arsenic was not present, the workers had an
excess proportional mortality rate for cancer (in age groups
below 55) similar to that of workers in the arsenic-handling
plant.
NIOSH (1975) pointed out several deficiencies in this study.
Snegireff and Lombard did not calculate the relative lung cancer
mortality rates in the two plants, which accounted for 38.9 and
50.0 percent of cancer deaths in the "arsenic" plant and the
"nonarsenic" plant, respectively. Furthermore, because they made
no attempt to determine the status of former employees or retirees,
the exposed cohorts are incomplete.
Using the total cancer deaths experienced in each plant,
NIOSH calculated the expected number of lung cancer deaths, by
age group, that should have occurred if the rates for the ap-
propriate U.S. population were applied. Both plants showed large
excesses of lung cancer deaths relative to mortality from all
causes in 1938 (460% for Plant A; 350% for Plant Z). Both
plants also showed large excesses when lung cancer deaths were
compared to all cancer deaths (450% for Plant A; 550% for Plant
Z). These excesses for lung cancer contrasted with the deficits
that both plants showed for total cancer mortality relative to
all causes of death (4% for Plant A; 25% for Plant Z). The
excesses of lung cancer deaths in both plants indicate that it is
lung cancer rather than all cancers that requires detailed
examination and furthermore that Plant Z was an inappropriate
control population since its workers were evidently exposed to
some respiratory carcinogen.
COMMUNITY EXPOSURES
Mortality studies of inhabitants of communities near smelt-
ers emitting arsenic have also been reported. In a study com-
paring lung cancer mortality in 71 counties, Blot and Fraumeni
38
-------
(1975) suggested an association between elevated lung cancer
mortality and environmental air pollution from industrial sources
of inorganic arsenic. The study compared the average mortality
rates for lung cancer among white males and females in counties
with copper, lead, or zinc smelting and refining industries and
in counties with other nonferrous ore-processing industries.
Copper, lead, and zinc ores contain substantial amounts of
inorganic arsenic, whereas aluminum and other nonferrous metals
contain much smaller amounts.
Seventy-one counties with manufacturing units engaged in
primary smelting and refining (SR) of nonferrous ores (and with
an estimated 50,855 persons employed in the SR industry in 1963)
were selected for study. The 71 counties were divided into two
groups according to the type of ore processed: 36 counties with
copper, lead, or zinc processing units and 35 counties with
installations processing aluminum or other nonferrous metals.
Average annual age-adjusted lung cancer mortality rates per
100,000 people in 1950 to 1969 were calculated for the 71 SR-
industry counties and the other 2984 counties of the contiguous
United States. A multiple-regression analysis permitted correc-
tion for differences in demographic and'socioeconomic factors
such as population density, degree of urbanization, schooling,
income, and'geographic region.
This analysis showed that the 36 counties with copper, lead,
and zinc SR industries had a significantly higher lung cancer
mortality among males (p<0.001) and females (p<0.05) than did
counties in the rest of the United States. The mean excess in
lung cancer for these counties, corrected for demographic and
socioeconomic influences, was 17 percent for males and 15 percent
for females. The rates for both sexes were high in each 5-year
interval over the 20-year period. The 35 counties with indus-
tries processing nonferrous ores other than copper, lead, and
zinc showed no excess lung cancer mortality.
The median SMR for all 36 counties with SR industries was
112 for males and 110 for females. Data from three counties with
39
-------
the highest proportion of the total population employed in the 3R
industry (each had >5%) showed an average excess lung cancer
mortality of 92 percent in males and 36 percent in females.
According to Blot and Fraumeni, occupational exposure alone
cannot explain the increased mortality in both males and females.
Although occupational exposure probably contributes to excess
risk in males, the number of workers in the SR industries is
small relative to the total population of the counties (less than
1% for more than half of the 36 counties and less than 3% for all
but four). The relative risk for the workers would have to be
13-fold to account for the excess lung cancer observed. The
authors believe that their correlations are not due to confound-
ing social factors or to differences in smoking habits. They
conclude that "the most likely explanation for the elevated lung
cancer mortality in this study is neighborhood air pollution from
industrial sources of inorganic arsenic." They do not, however,
rule out contributory effects from other toxic substances. The
implications of their results, together with criticisms of their
study made by others, are discussed in detail in Section 6.
Milham and Strong (1974) confirmed that inhabitants of
copper smelter communities absorb excessive' arsenic. Their study
of children living near the copper smelter in Tacoma, Washington,
revealed elevated levels of urinary arsenic that decreased with
distance of residence from the smelter stack. Analyses of
vacuum-cleaner dust showed a decreasing arsenic content with
increasing distance of residence from the smelter. Since the
urinary arsenic levels in children living near the smelter are
similar to those in smelter workers, it was argued that commun-
ities near the smelter might be exposed to an excess risk of lung
cancer. It should be noted that the authors did not consider
seafood ingestion as a source of arsenic, and they did not con-
sider the specific gravity of the spot urine samples.
Newman et al. (1976) studied the histologic types of lung
cancer in two Montana counties, one with a copper smelter (lo-
cated in Anaconda) and the other with several copper mines
40
-------
(adjacent to Butte). Vital statistics indicated an elevated lung
cancer mortality in both counties. The investigators examined
records of 143 lung cancer cases (114 males and 29 females) from
the period 1959 to 1972 and classified them histologically.
Occupational information was obtained for all males and for all
but 2 of the 25 Butte females. The persons were in four groups:
copper smelter workers (those with more than 1 years employment
in the smelter); copper mine workers (with more than 1 years em-
ployment in a mine); "other" men (with less than 1 years employ-
ment in the smelter or mines, this group serving as a control for
the smelter workers and miners); and females from Butte.
Lung cancer mortality of men from both Butte and Anaconda
and women from Butte was significantly elevated. No arsenic
exposure was apparent in Butte, and it was hypothesized that the
excess lung cancer might be due to exposure to asbestos-like dust
from a material used to sand the streets. The lung cancer rate
of women from Anaconda also was significantly elevated relative
to the rate for Montana* women, in calculations over a period of
10 years (2.9/10,000 as compared with 1.4/10,000; p<0.05). It
was noted that the average level of arsenic in Anaconda air was
0.45 mg/m .
Smelter workers were found to have a greater frequency of
poorly differentiated epidermoid carcinomas than copper miners
(p<0.05). Miners were found to have a distribution of histologic
types of lung cancer similar to that of controls ("other" men),
suggesting an etiologic agent in miners different from that in
smelter workers. Butte women showed a distribution of histologic
types similar to that of smelter workers. Histories of smoking,
though incomplete, indicated that smoking habits did not differ
significantly in the three male groups. The women smoked less
than the men.
The high percentage of poorly differentiated epidermoid
carcinomas among the smelter workers was unexpected. The authors
noted that Robson and Jelliffe (1963) had reported poorly dif-
ferentiated carcinomas in six patients with lung cancer who had
41
-------
received arsenical medication. They concluded that the poorly
differentiated epidermoid lung carcinoma among smelter workers
was best explained by exposure to airborne arsenic and further
suggested that poorly differentiated epidermoid lung carcinoma
may serve as a "marker" indicating exposure to arsenic.
Pershagen et al. (1977) studied mortality from different
causes among residents living near the Ronnskarsverken smelter
works in northern Sweden, which for decades has emitted large
amounts of arsenic as well as sulfur dioxide. Mortality ratios
for the exposed population for the period 1960 to 1974 were
compared with ratios for a reference population with similar
degrees of urbanization, occupational profiles, and age distribu-
tion. Lung cancer mortality of men over 40 in the exposed area
was significantly elevated (SMR = 250; p<0.0016). When the
occupationally exposed Ronnskarsverken employees were excluded,
however, the increase was no longer significant- Excess mor-
tality from lung cancer among the Ronnskarsverken employees was
large (SMR = 405). The authors are conducting a followup study
that should yield' more data on mortality rates of the nonoc-
cupationally exposed.
MANUFACTURE OF ARSENICAL PESTICIDES
Several studies of workers exposed to various inorganic
arsenic compounds used in the manufacture of arsenical pesticides
and insecticides have showed an increased lung cancer incidence
associated with such exposure. Unlike smelter workers, pesticide
workers have no concomitant exposure to sulfur dioxide. In some
plants, however, they do undergo exposure to a wide range of
other chemicals.
In 1945 the Medical Research Council in Britain commissioned
two studies to evaluate the relationship between arsenic and lung
cancer. Hill and Faning (1948) conducted a proportional mortal-
ity study of workers in a factory that produced arsenic-contain-
ing sheep-dip, and Perry et al. (1948) measured the levels of
arsenic and examined the workers in the same factory. Exposure
42
-------
to sodium arsenite in the factory was considered heavy among 31
chemical workers and moderate among 20 maintenance workers and 12
packers. Chemical workers all exhibited skin changes that made
it immediately apparent to the authors whether the person was a
chemical worker, even before a work history was obtained. Most
chemical workers were grossly pigmented, and one-third of them
had hyperkeratinization of exposed skin, with a tendency to wart
formation. Chemical workers had significantly higher arsenic
content in their hair (p<0.01) and urine compared with unexposed
workers (Perry et al., 1948).
In the period from 1910 to 1943 a total 75 of the factory
workers died, 22 of them from cancer. The proportion>of deaths
due to cancer (29.3%, 22 out of 75 deaths) was compared with 12.9
percent in other occupations in the same town, yielding a statis-
tically significant difference of 16.4+4.1. In a grouping of
the deaths into three periods (1910 to 1919, 1920 to 1929, 1930
to 1943) cancer deaths among the factory workers in each period
2
were still significantly higher than among controls (sum of x
for the three periods = 11.88; p<0.01). When deaths were stan-
dardized for three age groups (under 55, 44 to 59, 70 and over),
the proportion of deaths due to cancer among factory workers
(29.3%) was still markedly higher than that among other workers
(12.9%; p<0.01) (Hill and Faning, 1948).
Analysis of the deaths by work area showed that the excess
of cancer deaths was accounted for mainly by chemical workers (16
deaths), who had the heaviest exposure to arsenic dust, and by
three deaths among engineers and packers. The difference between
cancer deaths among the chemical workers and among persons in
2
other occupations (nonfactory workers) was significant (x =
3.95; p = 0.047) .
Analysis of cancer deaths by site suggested a relative
excess of lung and skin cancer among factory workers, although
Hill and Faning (1948) state that the numbers of deaths in each
category were too small to be conclusive.
43
-------
Ott et al. (1974) reported the results of both a propor-
tional mortality study and a retrospective cohort study of deaths
among workers at a Dow Chemical plant that manufactured arsenical
insecticides, primarily lead arsenate and calcium arsenate.
The proportional mortality study revealed that lung cancer
accounted for a significantly higher proportion of all deaths in
the exposed group of 173 decedents who had worked for one or more
days with arsenical insecticides (16.2%) than in the controls,
1809 decedents who had worked at the same plant but without
arsenic exposure (5.7%; p<0.001). Lymphatic and hematopoietic
cancer (excluding leukemia) were also found to be significantly
elevated in the exposed group (3.5%) relative to the controls
(1.4%; p<0.05).
The retrospective cohort analysis included 603 chemical
workers with a minimum of 1 month exposure to arsenic. Although
overall mortality of the cohort was low compared with mortality
of U.S. white males, mortality due to lung cancer and lymphatic
and hematopoietic cancer (excluding leukemia) were both con-
siderably higher than expected on the basis of. U.S. white male
deaths (SMR = 345 and SMR = 385, respectively).
Four exposure groups were established, and time-weighted
average concentrations were calculated for each group. From
these average concentrations, the arsenic dosages were calcu-
lated. By both methods of analyses lung cancer mortality in-
creased with increasing cumulative arsenic exposure. The meth-
odology of this study is discussed in Section 6.
Baetjer et al. (unpublished study, summarized in NIOSH,
1975) did a preliminary proportional mortality study of retired
workers from a Baltimore plant manufacturing arsenical pesti-
cides . Cancer mortality among 22 retirees was almost four times
the expected rate (based on age-sex specific.proportional mor-
tality ratios for the city of Baltimore): 17 cases vs. 4.43
expected. The SMR's were 671 for lung cancer, 300 for "lymphatic
and hematological cancers" (lympho-sarcomas), and 149 for all
other cancers. There is a problem in that workers experienced
44
-------
concomitant exposure to several hundred chemicals, some of them
known carcinogens, in the air of the plant.
When the observed death rates were compared with age-cause
specific death rates calculated for the population of Baltimore,
the retirees again showed an excess mortality from lung cancer,
with an SMR of 1667 (95% confidence limits of 7.14 to 32.8%);
from lymphatic cancer, with an SMR of 5000 (confidence limits of
6.05 to 180.50); and from all other cancers, with an SMR of 465
(confidence limits of 1.26 to 11.90). The number of deaths in
this study was small. Baetjar et al. are currently conducting a
broad retrospective cohort mortality study of the workers in this
plant. This more thorough study has not borne out the reported
excess mortality from lymphatic cancer.
These studies are summarized in Table 8.
AGRICULTURAL WORKERS
Agricultural workers engaged in the application of arsenical
insecticides constitute another occupational group with exposure
to inorganic arsenic. Some of these workers are known to be
exposed both through inhalation of insecticide sprays and through
ingestion of wine contaminated by arsenic.
Many cases of chronic arsenic poisoning occurred in Germany
in the 1930's among winegrowers who used arsenical insecticides
including calcium arsenate and copper acetoarsenite (Frohn,
1938) . The studies described here provide clinical evidence of
the carcinogenic effects of arsenic exposure, probably by both
inhalation and ingestion. The use of arsenical insecticides was
banned in Germany in 1942. Roth (1957) reported on 27 autopsies
he performed on Moselle vintners between 1950 and 1956. All 27
workers had been exposed 20 to 30 years earlier, between 1925 and
1938, and their exposures had lasted several years. Symptoms of
arsenic poisoning appeared in most of the workers after 1935.
Sixteen of the 27 Moselle winegrowers had a total of 38
malignant tumors. Six had "arsenic cirrhosis" of the liver.
Eleven (or 40%) had lung cancer. All of the lung cancer cases
45
-------
TABLE 8. SUMMARY OF EPIDEMIOLOGICAL AND CLINICAL STUDIES OF
NONSMELTER OCCUPATIONAL EXPOSURE TO ARSENIC
Study
Hill and Fantng
(1948)
Perry et al.
(1948)
Sneglreff and
Lombard (1951)
Roth (1957)
Daetjer et al.
(1974)
Type of study.
and period of
observation
Proportionate mor-
tality. 1910-1943
Environmental and
clinical
Proportionate mor-
tality. 1922-1949
Clinical
Proportionate mor-
tality. 1960-1972
Process
Sodium arsenlte
sheep-dip manu-
facturing plant
Same plant as Hill
and Fanlng
Plant handling large
quantities of A$203
German vlnegrowers
with exposure to
arsenical Insec-
ticides and As-con-
taminated wine
Arsenical pesticide
among workers '
Number In cohort
75 deaths
146 deaths
27 deaths
Findings
Two- fold excess for lung
cancer; 16 deaths from
cancer out of 41 deaths
among chemical workers
No significant excess of
cancer deaths determined
by proportionate mortality
18 lung cancer deaths out of
47 deaths. 7 hemanglosar-
comas of the .liver; cancer
of esophagus
Proportionate mortality
ratio of observed to ex-
pected lung cancer deaths
was 6.71 ; 3.0 for lym-
phatic and hematolonlcal
cancer (lymphosarcoma);
death rate compared with
age-cause specific death
rates for Baltimore showed
significantly Increased
cancer mortality for lung
cancer with 0/E ratio of
16.67 (95% confidence
limits of 7.14-32.84);
mortality from lymphatic
cancer (5.0 0/E ratio.
6.05-180.50 confidence
limits)
Exposure data
Median concentrations (mg/m3) of
As were 0.071 (packing room);
0.254 (drying room); 0.373
(sieving room); 0.696 (near
kibbler operation); urine sam-
ples and hair samples of exposed
workers had significantly higher
As levels than controls; pigmen-
tation and keratosls
Exposure to arsenical Insec-
ticide containing 4.3 to 56*
Asj03; As found In urine of
living patients
(continued)
-------
TABLE 8 (continued)
Study
Ott et al.
(1974)
Ott et al.
(1974)
Newman et al.
(1975)
Nelson et al.
(1973)
Type of study,
and period of
observation
Proportionate mor-
tality
•
Retrospective cohort
Clinical
Followup to 1938-
1969
Process
Lead and calcium
arsenate, copper
acetoarsenite,
magnesium arsenate
People exposed oc-
cupationally or
otherwise to lead
arsenate pesticide
spray or residue
Number In cohort
173 deaths with
exposure; 1809
decedents with no
exposure
Cohort of 603 chem-
ical workers with
at least 1 month of
work In As produc-
tion
Findings
Proportionate mortality for
lung cancer significantly
higher among exposed group
(16.2%) than in controls
(5.7%); lymphatic cancer
also significantly higher
in exposed group (3.5% vs.
1.4% expected)
Overall mortality lower than
U.S. white male mortality;
lung cancer mortality (0/E
ratio of 3.45) and cancer of
lymphatic and hematopatic
tissues expect leukemia (0/E
ratio 3.85) were signifi-
cantly higher than expected
1n cohort of exposed workers
Significantly increased lung
cancer among men of Anaconda
and Butte and in women of
Butte as compared with
Montana as a whole
No excess lung cancer
Exposure data
Dose-response shown between As
exposure for respiratory can-
cer mortality
-------
also showed the typical signs of chronic arsenic poisoning. Roth
concluded that they could not be attributed to the increase in
lung cancer in the general population, but must be regarded as
occupational lung cancers caused by arsenic. In only one of the
cases was there a history of heavy, smoking. The latent period he
observed for malignant tumors (of the skin, liver, bile duct,
esophagus, and respiratory tract) was 13 to 25 years.
Since Moselle vintners were thought to have a high consump-
tion of wine, some of which contained high levels of arsenic,
Roth explicitly stated that his findings "cannot be explained by
chronic alcoholism, nor do they provide evidence of combined
poisoning" (i.e., via inhalation and ingestion), but he does not
present a quantitative defense of this reasoning. He concluded
that "the late manifestations of chronic poisoning, particularly
cirrhosis of the liver and multiple neoplasms, are occupational
diseases."
In a second article, Roth (1958) added 20 autopsies of
winegrowers to the earlier 27. Among the total of 47 cases, 30
deaths were attributed to cancer and 3 to malignancies. In 18
cases lung cancer was listed as the cause of death. There were
also six cases of hemangio-sarcomas of the liver, five cases of
cancer of the esophagus, and one of cancer of the bile duct.
Roth also compared the proportional mortality rates of six
rural and urban districts of the Moselle and one district of the
Ahr. Mortality due to lung cancer was higher in winegrowing
areas of the Moselle, where arsenical insecticides had been used,
than in urban and nonwinegrowing areas. In winegrowing areas of
the Ahr, where arsenical insecticides had never been used, the
rate for lung cancer was lower than in the winegrowing areas of
Moselle. Roth took this as added evidence of a causal connection
between arsenical insecticides and lung cancer.
Braun (1958) performed another clinical study in which 16
vintners from the Palatine region of Germany were examined
between 1951 and 1957. These workers had definite histories of
occupational exposure to arsenical pesticides (from about 1925 to
48
-------
1942) and of frequent wine consumption. All 16 showed palmar and
plantar keratoses characteristic of chronic arsenic poisoning; 9
had pigmentation, 9 had inoperable lung cancer, 1 had bile-duct
cancer, and another had a malignant lymph node tumor. Two of the
lung cancer patients also had skin cancer (squamous cell carci-
noma) , and two other lung cancer patients had Bowen's disease.
Only 3 of 16 vintners with arsenical skin lesions had no neo-
plastic changes.
In the whole period from 1939 to 1952 only one vineyard
worker with lung cancer was reported. Braun explains the dra-
matic increase in lung cancer cases among vintners in the years
following 1952 by the long latent period for arsenic-induced
cancer. Unlike Roth, Braun does not refer to the smoking his-
tories.
Galy et al. (1963) reported three cases of lung cancer among
winegrowers in the Beaujolais region of France. The three had
handled arsenic pesticides (lime arsenate and lead arsenate) 20
years earlier. 'All three had keratosis. The authors note that
in spite of the ingestion of a "nonnegligible quantity of arse-
nic, . . . ingestion of arsenic seems to be secondary to inhala-
tion."
Nelson et al. (1973) did an epidemiologic study of the
effects of exposure to arsenical pesticides in a followup of an
earlier morbidity study among 1231 residents of the Wenatchee
Valley, Washington, some of whom were exposed to lead arsenate
spray. The original Public Health Service study in 1938 to 1939
divided the study group into three exposure groups:, "orchard-
ists, who had prepared and used lead arsenate sprays during
1938; "consumers," who were not exposed to the spray (mainly
women and children), and "intermediates," a mixed group including
former orchardists, warehouse workers, and people with occasional
exposure to lead arsenate spray.
The followup study, which covered the 30-year period from
1938 to 1968, included more than 98 percent of the original
cohort. Standard mortality ratios (SMR's) for different exposure
49
-------
groups and for different lengths of exposure were calculated with
the population of the State of Washington as a standard. The
life-table method was used to calculate expected deaths. The
cohort as a whole had an SMR of 70, indicating mortality rate
that is favorable relative to the Washington average. The
mortality rates for the three exposure groups did not reflect
exposure. "Intermediates" had the highest SMR (78) and "orchard-
ists" had the lowest (65). Mortality rates also failed to
reflect increased length of exposure. No excess deaths were
found to be due to specific causes of death (heart disease,
cancer, stroke).
Nelson et al. point out several limitations in their study:
(1) a small number of workers in the original study; (2) the lack
of adequate exposure data; and (3) the possible loss of the most
susceptible workers, who may either have left orchard work or
died before 1938.
Because results of this study conflicted with evidence of
the effects of chronic .exposure to arsenic, NIOSH (1975) re-
assessed the mortality experience of Wenatchee Valley residents
exposed to lead arsenate, using independent sources of data.
Information on occupation and cause of death in all deaths among
adult white males in the State of Washington over the period
1950 to 1971 showed that the lung cancer rate of orchardists was
19 percent higher than expected. Over the 11-year period (1961
to 1971) lung cancer showed a "statistically significant increase
of 27 percent" (NIOSH, 1975).
These studies are summarized in Table 8.
MEDICINAL USE OF ARSENIC COMPOUNDS
The medical use of arsenic compounds in the treatment of
diseases such as psoriasis, eczema, dermatitis, anemia, asthma,
and epilepsy provides data on the effects of chronic ingestion of
relatively high doses of arsenic compounds. Such effects include
pigmentation, keratoses on the hands and feet (characteristic of
chronic arsenicism), skin cancer, and, possibly, lung cancer.
50
-------
Despite the fact that the first association between arse-
nical treatment and skin cancer was made in 1887 by Hutchinson,
the use of arsenicals in medicine continued until recently (NAS,
1977).
In 1947 Naubauer published an extensive review of the liter-
ature on arsenical cancer. Of the 143 cases of skin cancer
included in his review, nearly all had been treated with arsenic
in the inorganic trivalent form, most often as Fowler's solution
(potassium arsenite). Approximately 90 percent of the patients
had taken Fowler's solution for more than 1 year, and 50 percent
for more than 5 years. On the average, a total quantity of 28 g
of arsenic was ingested by each patient (NAS, 1977).
Ninety percent of the patients treated with Fowler's solu-
tion had keratoses, typically on the hands and feet, and many had
hyperpigmentation. The period from the beginning of treatment to
the appearance of skin cancer averaged 18 years, with a range of
3 to 40 years. The latent period for keratosis averaged 9
years. In these patients treated with arsenic, skin cancer
appeared at a relatively early age (one third were 40 or younger;
70% were 50 or younger). Thirteen of the 143 patients developed
cancers at other sites (NAS, 1977).
There is clinical evidence for the induction of lung cancer,
as well as skin cancer, from treatment with arsenicals. Robson
and Jelliffe (1963) described six cases of lung cancer in patients
who had been treated with arsenic. Four of the six had received
arsenic in the form of Fowler's solution for periods ranging from
3 to 15 years; the remaining two patients had received an un-
specified form of arsenic for several years. All of the patients
showed keratoses; three showed intraepidermal epitheliomas, and
one showed intraepidermal carcinomas. All six had poorly dif-
ferentiated bronchial carcinoma, the same type observed by
Newman et al. (1976) and taken by them to be a "marker" for
arsenic exposure. Two of the patients were moderate cigarette
smokers, one was a light pipe-smoker, and three were nonsmokers.
51
-------
The average latent period before the clinical onset of lung
cancer was 32 years.
Fierz (1964) carried out a followup study of 262 patients
who had received Fowler's solution within the previous 26 years.
Knowledge of the amount of Fowler's solution received by each
patient enabled Fierz to establish a clear dose-response rela-
tionship between the quantity of arsenic ingested and the fre-
quency of keratosis and skin cancer. As a group, the patients
received from 10 to 2600 ml of Fowler's solution. Hyperkeratoses
were present in 106 of the 262 subjects (or in 40%); melanosis
occurred in only 5 cases; and skin cancer occurred in 21 cases of
the subjects (or 8%). The cancers were multiple in 13 of the
21 cases. Fierz calculated an average latency period of 14 years
for skin cancer/ 4 years shorter than that calculated by Neubauer.
As in the cases reviewed by Neubauer (1947), Fierz's skin
cancer cases were relatively young (20 to 21 patients were less
than 60 years old at the time of appearance of skin cancer).
Fierz, however, purposely examined only patients under'65 years
of age to avoid "distortion of the picture by any old-age can-
cer." Since some of the patients had begun arsenic treatment
only 6 years before examination and the' latency period for
arsenic-induced skin cancer is 14 years or longer, Fierz notes
that additional cases of skin cancer might appear (Fierz, 1964).
These studies are summarized in Table 9.
INGESTION OF ARSENIC FROM CONTAMINATED FOOD AND WATER
Clinical and epidemiological studies of accidental ingestion
of arsenic-contaminated food and water are summarized in Table 9.
An early report was that of Reynolds (1901), in which he studied
an unusual number of cases of skin eruptions (including erythema,
keratosis, and pigmentation) in a Manchester infirmary. He
attributed these to arsenic poisoning by beer containing approxi-
mately 2 to 4 ppm of arsenious oxide (As20_) from brewing sugars.
Because of the large number of cases that Reynolds examined and
his detailed descriptions of symptoms, this article has been
52
-------
TABLE 9. SUMMARY OF EPIDEMIOLOGICAL AND CLINICAL STUDIES
OF EFFECTS OF INGESTED ARSENIC
Study
Reynolds
(1901)
-
Tseng
et al.
(1968)
Study
popu-
lation
- —
40,421
90.6%
of the
popula-
tion
were
exam-
ined;
total
popula-
tion at
risk
103,154
Numbers and
types of
disease
500 cases seen
by Reynolds
(13 deaths) ;
most prominent
symptoms
erythromelalgia ,
keratosis, pig-
mentation, neur-
itis, cardiac
failure, cir-
. chosis of the
_.liver — — •
7418 cases of
hyperpigmen-
tation (18.4%)
2860 of kera-
tosis (7%);
428 of skin
cancer (1%) ;
360 of Black-
foot disease
(0.9%)
Degree of
exposure
2-4 ppm in
contamina ted
beer; patients
had drunk 2-16
pints of beer
daily for many
months
• • •
0.01-1.82
ppm in water
from artesian
wells; most well
water had an
arsenic content
around- 0.4-0.6
ppm I
Form of
arsenic
Arsenious
oxide
»
Arsenic in
hair or
urine
_-.
Histology
131 cases of
skin cancer
out of 420
were proved
by biopsy
specimens;
permission
to perform
biopsies was
refused in
remaining
297 patients
but the
lesions had
all the
characteris-
tics of
arsenical
cancer _ r_
Dose-
response
Clear
dose-
response
relation-
ship: the
greater
the arsenic
content of
the water.
the higher
the preval-
ence of
skin can-
cer, kera-
tosis,
hyperpig-
mentation.
and Black-
~ foot
disease
Control
7500 persons^
whose drinking-
water contained
no detectable
arsenic or water
containing
0.001-0.017 ppm
had not a single •
case of melanosis.
keratosis, or skin
cancer
•
Ui
U)
(continued)
-------
TABLE 9 (continued)
Ul
Study
Fieri
(1964)
Regelaon
et al.
(1968)
Study
popu-
lation
262
Numbers and
types of
disease
Hyperkera tosis-
106 or 40*
melanodenna-
S cases;
skin cancer-
21't cases or
8%; S deaths
vfrbm malig-
nant tumors
including 3
from bron-
chial carci-
noma
1 patient with
hemangio-
endothelial
sarcoma of
the liver.
plantar and
palmer kera-
tbses.
Degree of
exposure
10 to 2000 ml
of Fowler's
solution
17 years of
treatment
with
Fowler • s
solution .
k
Form of
arsenic
Potassium
arsenite
As203
in Fowler's
solution
Potassium
arsenite
*
Arsenic in
hair or
urine
Not found
(not
surprising
7 years
after
ingestion)
Histology
All skin
changes were
tested
histologi-
cally
Interpreted
variously as
"cirrhosis
with acute
hepatitis"
and as
"hemangio-
endothelial
sarcoma"
Dose-
response
Both hyper-
keratosis and
skin cancer
showed in-
crease with
increasing
arsenic
dose
Control
(continued)
-------
TABLE 9 (continued)
01
Ul
Study
Hamamoto
(1955)
Neubauer
(1947)
Robson and
Jelliffe
(1963)
Study
popu-
lation
Numbers and
types of
disease
Of 61 babies
hospitalized
98% had mel-
anosis, 26%
had hyper-
keratosis,
100% had
liver
swelling
Keratoses in
90% of
patients who
received
Fowler's
solution;
skin cancer
in all 143
patients
Six cases
of lung
cancer;
all 6 had
keratoses,
3 had intra-
epidermal
epithelioma,
1 had intra-
epidermal
carcinoma
Degree of
exposure
1.5-2.4 rag
As/100 g
of powdered
milk
Average
total
quantity of
28 g;
90% of
patients
received
Fowler's
solution
for <1
year; 50% ,,
for <5
years
All
received
arsenic
for 3 to
15 years
Form of
arsenic
Trivalent
arsenic
•
Nearly
all
received
inorganic
trivalent
arsenic
(most
often
Fowler ' s
solution)
4 had
received
Fowler ' s
solution
Arsenic in
hair or
urine
Hair
3 . 5 ppm
in hair
of 1
patient;
1.5 ppm
in hair
of another
patient
Histology
Half of skin
cancers were
aquamous
carcinomas;
half were
basal cell
epithelio-
ma s
All tumors
were poorly
differen-
tiated
carcinomas
Dose-
response
Control
(continued)
-------
TABLE 9 (continued)
Study
Yell
(1973)
Bergoglio
(1964)
'Bargono
and
Greiber
(1972)
Study
popu-
lation
2355
deaths,
556 from
cancer
and mal-
ignant
tumors
(23.8%)
27,088
children
180 in-
habitants
of Anto-
fagasta
Numbers and
types of
disease
35% of cancer
deaths were
due to res-
piratory
cancer;
35%, diges-
tive tract;
2.3%, skin
12% had cu-
taneous
changes;
1/4 - 1/3
had systemic
symptoms;
11% had
aero-
cyanosis;
80% had
abnormal
skin pig-
mentation;
36% had
hyperkera-
tosis; 30%
had Raynaud's
syndrome
Degree of
exposure
0.8 ppm in
drinking-
water
Form of
arsenic
»
1
Arsenic in
hair or
urine
Histology
303 lesions
in 184
patients
studied
histolog-
ically
Dose-
response
Depart-
ments
with
higher
cancer
mortality
also have
higher
arsenic
content
in drink-
ing water
Control
137,702 deaths
among entire
population
of Cordoba
Province of
1,759,997 inhab-
itants (1949-
1959). 15.3%
of all deaths
were due to
cancer and
malignant tumors
(average for
11-year period)
No cases of
abnormal skin
pigmentation,
hyper keratosis.
or Raynaud's
syndrome
01
-------
referred to as "the definitive medical description of subacute
poisoning with ingested arsenic" (NAS, 1977). Reynolds person-
ally examined 500 patients suffering from arsenical poisoning,
each of whom had each drunk 2 to 16 pints a day of contaminated
beer for many months. He classified the patients into four
groups: (1) those with skin lesions principally; (2) those with
cardiac and hepatic lesions principally; (3) those with paralytic
lesions principally; and (4) those with all symptoms. The major
cause of death was cardiac failure.
A more recent and more quantitative study of accidental
arsenic ingestion was done by Tseng et al. (1968) , who examined
the incidence of melanosis, keratosis, skin cancer, and Blackfoot
disease (a peripheral vascular disorder resulting in gangrene of
the extremities) in 37 villages on the southwest coast of Taiwan
where the population had been exposed for over 40 years to the
high levels of arsenic in the drinking water. The concentrations
of arsenic in the drinking water ranged from 0.01 to 1.82 ppm.
The arsenic content of most well water was 0.4 to 0.6 ppm.
Among the 40,421 inhabitants studied, the overall incidence
of melanosis was 183.5/1000; the incidences were 71.0/1000 for
keratosis, 10.6/1000 for skin cancer, and 8.9/1000 for Blackfoot
disease. (Incidence of skin cancer among Chinese in Taiwan is
normally low, 2.9 percent.) On the whole, the rate of all four
conditions increased steadily with age, though women over 69
showed a lower rate of cancer and melanosis. Over 10 percent of
the people above age 59 had skin cancer.
The youngest cancer patient was 24, the youngest with
melanosis was 3, and the youngest with keratosis was 4; these
observations were taken to indicate that there is a latency
period even at high dose levels. The association of Blackfoot
disease with melanosis, keratosis, and skin cancer was signifi-
cantly higher than expected, suggesting a causal relationship
between Blackfoot disease and chronic arsenicalism. Among the
428 cases of skin cancer, 89.7 percent had melanosis, and 71.7
percent had keratosis.
57
-------
When villages were grouped by the arsenic content of their
well water ("low," below 0.3 ppm; "mid," 0.3 to 0.6 ppm; and
"high," above 0.6 ppm) the incidence of all four conditions rose
with increased arsenic concentration. This was taken as evidence
of a dose-response relationship. The dose-response data for skin
cancer are shown in Figure 1.
Evaluation of a control population of 7500 persons whose
drinking water contained no detectable arsenic or very low
amounts (0.001 to 0.017 ppm) revealed no cases of melanosis,
keratosis, or skin cancer.
Results of biopsies taken during the study by Tseng et al.
are discussed by Yeh (1973). Yeh also calculated standardized
mortality ratios for the patients with skin cancer and those with
Blackfoot disease and found that both groups had a much higher
mortality rate than the study population, especially in the total
age group below 49 years. The numbers of cases were small: 7
skin cancer patients below age 49; 16 Blackfoot disease patients
•
below age. 49.- The most common cause of death among skin cancer
patients was carcinoma of various sites (skin cancer having the
highest incidence) and that among patients with Blackfoot disease
was cardiovascular disease.
Irgolic (1978, unpublished report supplied by EPA) has
analyzed samples of well water from two locations in Taiwan by
flameless atomic absorption spectrometry and neutron activation
analysis. Both arsenite and arsenate (in a ratio of 1:10 in one
sample and in a ratio of 1:7 in a second sample) were detected,
but the ratio of arsenite to arsenate may have changed from the
time of collection to the time of analysis, since some of the
well waters are known to be anaerobic and since some oxidation
may have occurred during transit.
Recently, fluorescent compounds have been detected in sam-
ples of well water from areas of Taiwan where Blackfoot disease
is endemic (Lu, et al., 1975, 1977a, 1977b). One of these fluo-
rescent compounds appeared to be lysergic acid or alkaline
hydrolysate of ergotamine tartrate, or a related compound. Lu et
58
-------
200 r 1/1 000
150
100
50
0
.
.
M 4 M 1 11
| |«.3 1.5 1.7
AGE 20-39
200
150
100
50
0
1/1000
.
H
98. J
H
n ARSENIC (
MS. 3
r
M U
HH 1
H
H
L
" M
163.4 U
r^
I
H
31.0 M L U
1 rTL£Tl
40-59 60 AND OVER TOTAL
H
MALE
H
110.1
ia.0
AGE 20-39
200
150
100
50
0
1/1000
-
—
iH M L U
M U
«.o 62.»
TiJ
I^L^
40-59 60 AND OVER TOTAL
H
72.C
FEMALE
iio
xt U
MI
M , U
106.7 i 1*7.9
i
i
*
L
Z7.1
H
1?1 L 104
ARSENIC CONCENTRATION (ppm) IN WELL WATER
HIGH 0.60 AND OVER
MED 0.30 - 0.59
LOW 0.00 - 0.29
UNDETERMINED*
AGE 20-39 40-59 60 AND OVER TOTAL
BOTH SEXES
"Undetermined" exposures refer to villages where arsenic
polluted wells were no longer used or where the levels of
arsenic in different wells varies too widely to allow
classification.
Figure 1. Age-specific and sex-specific prevalence rate
(1/1000) for skin cancer by arsenic'concentration in
well water (from Tseng et al., 1968).
59
-------
al. (1977a) reported that one of the fluorescent compounds (as
yet unidentified) produced abnormalities in developing chick
embryos.
Another report relating cancer to the ingestion of arsenic
in water is that of Bergoglio (1964). Arguello et al. (1938) had
reported a high rate for skin cancer in Cordoba Province, Argen-
tina, where well water was known to have contained high levels of
arsenic for at least 40 years, but this report was not defini-
tive. Bergoglio examined the specific cause of death among all
deaths occurring in the 11-year period between 1949 and 1959 in a
region of Cordoba Province with well waters known to have high
arsenic levels. He also classified deaths in Cordoba Province by
geographic department.
During the 11-year period, there were 2355 deaths in the
study region; 556 (or 23.8%) were attributed to cancer and
malignant tumors. Thirty-five percent of all cancers were
respiratory cancer; 35 percent, digestive tract cancer; and 2.3
•
percent, skin cancer. The proportion of deaths due to cancer
(23.8%) was significantly higher than that of the province of
Cordoba as a whole over the 11-year period (15.3%), with no
significance level stated. Respiratory cancer and skin cancer
accounted for a greater-than-expected proportion of the total
cancer deaths among the study population, but corresponding
figures for the control population are not given. Fifty percent
of the deaths from lung cancer occurred among females. The
higher proportion of mortality from cancer and malignant tumors
corresponded to those regions with high levels of arsenic in the
drinking water.
In Antofagasta, Chile, in the early 1960's, dermatological
symptoms, especially among children, drew attention to high
levels of arsenic contamination (0.88 ppm) of the drinking
water. Of 27,088 schoolchildren surveyed in this city, Borgono
and Greiber (1972) found that 12 percent showed skin changes
characteristic of arsenicism; one fourth to one third had sys-
temic symptoms, and 11 percent had acrocyanosis (a circulatory
60
-------
disorder affecting the hands). The authors compared the records
of 180 inhabitants of Antofagasta with those of 98 people of
Iquique, a city with no arsenic in its water supply. Among the
180 Antofagasta residents, most of whom were under 20 years old,
80 percent had abnormal skin pigmentation; 36 percent had hyper-
keratosis; 30 percent had Raynaud's syndrome; 22 percent had
acrocyanosis; and 15 percent had bronchopulmonary disease. There
were no cases of these disorders among the control group. The
authors do not discuss cancer, but since the latent period for
arsenic-induced skin cancer is long, one would not expect to find
an increase of cancer within the period of the study.
In the midfifties, two incidents of arsenic food poisoning
occurred in Japan. In 1955, more than 10,000 infants who had
been fed powdered milk contaminated with pentayalent inorganic
arsenic showed signs of poisoning, and 130 infants died. The
major acute symptoms, reported in the many papers published .
between 1955 and 1975, were fever, diarrhea, vomiting, and
anorexia. Incidence of melanosis was also high. The reports are
summarized in Tsuchiya (1977).
Hamamoto (1955) recorded the symptoms of 61 hospitalized
infants who had been fed the tainted milk. Melanosis was ob-
served in 98 percent, liver swelling in 100 percent, and hyper-
keratosis in 26 percent. Heart palpitations or instability were
noted in 60 percent of the cases. Laboratory tests revealed
anemia and granulocytopenia. No neuritis was observed. Hamamoto
calculated that the infants of 3 months or older developed
symptoms after ingesting from 90 to 140 mg of arsenous acid.
Hamamoto further noted the remarkably rapid disappearance of
symptoms following BAL therapy.
It is unclear from the various reports whether the arsenic
in the powdered milk was in the trivalent or pentavalent form.
Tsuchiya (1977) indicates it was pentavalent, whereas Hamamoto
(1955) and Ohira and Aoyama (1973) indicate that it was triva-
lent.
61
-------
The Japanese Pediatric Society (1973) summarized the find-
ings of regional investigations of the long-term effects among
those who were fed the tainted milk in 1955. They reported high
rates for mental retardation, for epilepsy and other abnormali-
ties suggestive of brain damage, and for skin changes (tnelano-
sis). The Japanese Pediatric Society stressed the need for
epidemiclogic studies of the cohort exposed to the contaminated
milk to assess the relationship between arsenic and long-term
effects. Followup studies on the long-term effects of the
children who were exposed to the contaminated milk are compli-
cated by incomplete records and the difficulty of discovering who
was in the exposed cohort (Tsuchiya, 1977).
In 1956, 417 patients who had eaten arsenic-contaminated soy
sauce were examined. The estimated dosage was 0.1 mg/ml arsenic,
thought to be calcium arsenate. Mizuta et al. (1956) reported on
the acute symptoms of 220 of the patients. The duration of
arsenic ingestion was from 2 to 3 weeks in most cases; symptoms
included edeirfa of the face, disturbances of the gastrointestinal
tract and upper respiratory tract, skin lesions, neuritis, and
swelling of the liver. Laboratory findings included slight
anemia, leukopenia, and relative lymphocytesis. Abnormal EKG
findings were noted in 80 percent of the 220 persons examined.
There were no fatalities in this incident, and no long-term
followup has been reported.
62
-------
SECTION 5
NONCARCINOGENIC TOXIC EFFECTS OF ARSENIC ON HUMANS
This section summarizes the major noncarcinogenic toxic
effects of arsenic.
Cases of acute poisoning with 'solid trivalent inorganic
arsenic have occurred most frequently from the accidental inges-
tion of contaminated food or drink. Acute poisoning from indus-
trial exposure is relatively rare. Arsenic causes an increased
permeability of the capillaries and withdrawal of water from the
body. Initial symptoms are inflammation of the stomach and
intestine, vomiting, profuse and painful diarrhea, and difficulty
in swallowing. Nervous symptoms include vertigo, headache, and
pain in'the limbs, with poor circulation in the extremities.
Later symptoms include syncope, coma, clonic and tonic spasms,
and general paralysis. Death usually occurs from exhaustion due
to the prolonged gastroenteritis (Sollmann, 1964). Skin reac-
tions are rare in cases of acute arsenic poisoning (NAS, 1977).
In cases of subacute, nonfatal poisoning, the prominent
symptoms are inflammation of the mucous membranes of the gastro-
intestinal tract, conjunctivitis, coryza, stomatitis, and
pharyngitis. If the poisoning is prolonged, abnormalities of the
skin and nervous system (including neuritis) appear (Sollman,
1964). Autopsy findings on cases with acute arsenic poisoning
reveal gastroadenitis and cell infiltration (Sollman, 1964).
The symptoms of chronic arsenic poisoning generally fall
into three stages:
1. Weakness, loss of appetite, nausea, and occasional
vomiting, sometimes accompanied by diarrhea and pain in
the stomach.
63
-------
2. Conjunctivitis and catarrhal state of exposed mucous
membranes of the nose, larynx, and respiratory pas-
sages; coryza; hoarseness of voice; and mild tracheitis
or bronchitis. (Perforation of the nasal septum is the
most characteristic lesion of the upper respiratory
tract in workers exposed to arsenical dusts and is more
indicative of local irritation rather than chronic
systemic poisoning.)
3. Peripheral neuritis, mainly of the hands and feet
(Buchanan, 1962).
Table 10 summarizes the major noncarcinogenic toxic effects
of different forms of arsenic. Some of the acute and chronic
effects reported in various case studies of exposure to arsenic
are noted below.
The syndrome known as Ronnskar Disease, a chronic "rhino-
pharyngo-tracheo-bronchitis" (reported at the smelting works at
Ronnskar, Sweden) corresponds to Buchanan's second stage of
chronic arsenic poisoning (Buchanan, 1962). Skin manifestations
(especially keratoses and melanosis), hepatitis, and jaundice have
been reported at this stage.
Arsenic causes nephritis that is at first mainly vascular,
but always involves the epithelium (Sollmann, 1964). Arsenic is
hepatoxic, causing fatty swelling which may compress the bile
ducts, resulting in an increase in the bilirubin content of the
blood and in visible icterus (Sollmann, 1964). Cases of cir-
rhosis of the liver have been reported following ingestion or
medicinal administration of arsenic. Acute hepatitis has been
reported in patients undergoing arsphenamine therapy (Buchanan,
1962; Hine et al., 1977).
Systemic arsenic poisoning has been reported to affect
production and survival time of red and white blood cells.
Specific abnormalities include normochromic anemia, neutropenia,
a relative eosinophilia, and thrombocytopenia (Hine et al.,
1977).
In the 41 cases of neuropathy described by Heyman et al.
(1956) , recovery of sensory and motor functions was slow, and the
course of recovery was not improved by treatment with BAL. The
64
-------
TABLE 10. TOXIC EFFECTS OF ARSENIC
Compound
Route of entry
Acute effects
Trivalent inorganic
arsenic
Arsphenamine
Arsine
Ingestion
IV injection
Inhalation
en
Inorganic arsenic
Ingestion
Inhalation
Throat constriction, difficulty swallowing, violent
abdominal pain, vomiting, muscular cramp, possible
death within 1-4 days
Fatty infiltration of cells (especially liver cells);
direct toxic action on cardiac muscle; ECG abnormal-
ities
Stomatitis, albuminuria, jaundice, blood dyscrasias,
dermatitis, severe cerebral symptoms, acute hepatitis,
ascites
Enlargement of the liver, hemoglobinuria, jaundice,
hemolytic anemia, abdominal pain, vomiting; acute
uremia is a common cause of death (from 4th day
onwards); fatty degeneration of cells, particularly
those of the liver, kidneys, and cardiac musculature
Chronic effects
Nausea, vomiting, conjunctivitis and catarrhal state
of exposed mucous membranes of the nose, larynx and
respiratory passages, keratoses, melanosis
Perforation of the nasal septum; Ronnskar Disease
(chronic "rhino-pharyngo-tracheo-bronchitis")
hepatitis, jaundice, peripheral neuritis
Source: Buchanan (1962).
-------
degree of recovery was an inverse function of the severity of the
neuropathy. Biopsies taken from seven of the patients showed
various degrees of degeneration, the severity increasing with
increasing duration of disease. Unfortunately, no data are
available on whether neurological sequellae follow low-level
chronic exposures, and no dose-response information is available.
Electrocardiographic changes have been reported in cases of
both acute and chronic arsenic exposure. Specific effects are:
broadening of Q-R-S intervals, prolongation of Q-T interval, S-T
depression, and flattening of T-waves (Hine et al., 1977; Glazener
et al., 1968). Similar electrocardiographic abnormalities have
been noted in acute cases (Barry and Herndon, 1962). Abnormal-
ities have not been associated with disturbances of the serum
electrolytes, but rather with a direct toxic effect on the
myocardium. The clinical reports are that the myocardial effect
is reversible and recovery is accelerated with administration of
BAL.
A variety of symptoms (including bronchopneumonia, hyper-
keratosis, peripheral vascular disorders) and some deaths were
reported among children in Antofagasta, Chile, where the water
supply contained 0.8 ppm arsenic (Borgono and Greiber, 1972). As
a whole, inhabitants of Antofagasta had an increased incidence of
bronchopulmonary disease, pigmentation, hyperkeratosis, chronic
coryza, abdominal pain, Raynaud's syndrome, and acrocyanosis
(Borgono and Greiber, 1972) .
A high incidence of Blackfoot disease was found in the
population of southwestern Taiwan where drinking water contained
high levels of arsenic (Tseng et al., 1968).
Various sensory and motor neurologic abnormalities have been
observed in cases of acute and chronic arsenic poisoning. These
have been evidenced by measurements of nerve conduction, chron-
axie, and electromyographic activity, which serve as sensitive
indicators of neurological changes. An earlier report by Heyman
66
-------
et al. (1956) describes peripheral neuropathy caused by arsenical
intoxication.
Arsine gas, which is evolved when trivalent arsenic comes in
contact with nascent hydrogen, is extremely toxic. It is a
powerful hemolytic agent and can cause oliguric renal failure
(Fowler and Weissberg, 1974). Acute symptoms include abdominal
cramp, nausea, vomiting, jaundice, and anemia. In fatal cases
there may be delirium followed by coma. Death usually results
from myocardial failure a few days after the onset of anuria.
The symptoms of chronic exposure to arsine gas include vomiting,
dark urine, severe anemia, jaundice, and peripheral neuritis
(Browning, 1969).
EFFECTS OF LOW-LEVEL ARSENIC EXPOSURE
Bencko and Symon (1977) reported changes in the hearing
among 56 10-year-old Czechoslovakian children living near a power
plant burning coal of high arsenic content. Hearing losses were
detected in both air and bone conduction, especially at low
frequencies in the exposed group as compared with a control group
of 51 children of the same age living outside the polluted area.
The differences between the two groups in tests at low frequen-
cies had a high statistical significance (p<0.01). The authors
concluded that the observed losses may be due to toxic neural
damage caused by arsenic.
Milham and Strong (1974) noted that children living near the
Tacoma smelter had elevated levels of arsenic in the hair and
urine, but Milham (1977) reported that their school attendance,
hearing, and hematologic indices were within normal limits. No
adverse effects were associated with chronic, low-level exposure
to arsenic.
Attendance at the Ruston School, less than 100 yards from
the smelter, was similar to attendance at six other Tacoma
elementary schools in the period 1969 to 1974. The hearing
functions of 556 Ruston schoolchildren was favorable compared
with those of 17,623 Tacoma schoolchildren. Blood counts of 33
67
-------
Ruston schoolchildren were similar to those of children at a
school located 8 miles from the smelter (no p-values are given).
ORGANIC ARSENIC
Few studies have dealt with the effects of long-term low-
level exposure to organic arsenic compounds such as the herbi-
cides cacodylic acid, monosodium methanearsonate (MSMA) and
disodium methanearsonate (DSMA). Tarrant and Allard (1972),
studying forest workers who applied cacodylic acid and MSMA,
found significantly elevated (at the 5% level of probability)
urinary arsenic levels in exposed workers as compared with con-
trols. Urinary arsenic content was highest on Fridays and in
most cases had returned to normal by Monday. There was no
indication of a continuing increase over a 9-week study period.
In all but one of the 15 men applying cacodylic acid and
MSMA the urinary arsenic levels were above 0.3 ppm at least once
during the period of exposure. The highest level recorded was
2.5 ppm. No health problems were noted among these workers. The
amount of arsenic absorbed by workers could not be correlated
with different arsenic compounds or different methods of appli-
cation. The authors note that the urinary arsenic level at which
"concern for matters of health should begin is not well known."
Another study (Wagner and Weswig, 1974) of forest workers
exposed to cacodylic acid during a 2-month period also found that
urinary arsenic levels increased in the course of the work week.
Removal from exposure led to a rapid drop in urinary arsenic. No
workers in this study showed a sign of arsenic poisoning. The
authors note the lack of studies of long-term low-level exposure
to organic arsenic compounds and the need for animal studies as a
means of obtaining dose-response data.
Chronic toxicity resulting from reasonable levels of expo-
sure to MSMA and cacodylic acid should not be expected for the
following reasons:
68
-------
1. Their toxicity is 1/100 that of arsenite.
2. They are rapidly excreted in the urine.
3. The dog and cow methylate both arsenate and arsenite,
thus detoxifying them (NAS Arsenic, 1977).
4. Creciluis (1977) showed that the arsenate ingested by
man by drinking well water is rapidly converted to
cacodylic acid and excreted in the urine.
69
-------
SECTION 6
ASSESSMENT OF THE HEALTH EFFECTS OF ARSENIC
INTRODUCTION
A quantitative assessment of the health effects of exposure
to arsenic and its compounds is difficult for many scientific
reasons. Nonetheless, despite the limitations on interpretation
of the data baseband particularly on the extrapolation of a
dose-response relationship, the overwhelming weight of the epide-
miological and clinical evidence, when taken together, is that
exposure to inorganic arsenic compounds, particularly trivalent
arsenic, is related to an increased risk for the.development of
skin cancer, lung cancer,-and various skin disorders; some
diseases of the cardiovascular system; and peripheral neuropathy
and other nervous system manifestations. If these diseases
occur, they are generally serious and irreversible. Such dis-
eases have been associated with high levels of arsenic, and there
is no evidence that they arise from low-level environmental
exposures.
It must be noted, however, that the dose-response relation-
ship between lung cancer and arsenic alone, observed in the
epidemiological studies already cited, can technically be con-
sidered only highly suggestive since other contaminants, usually
sulfur dioxide, have also been present. Such contaminants must
thus be considered possible factors or cofactors in the disease
etiology. This is particularly true in view of the observation
by Lee and Fraumeni (1967) of an apparent dose-response rela-
tionship between lung cancer and concomitant exposure to sulfur
dioxide and arsenic. The ability of sulfur dioxide to act
70
-------
synergistically with other carcinogens such as benzo(a)pyrene has
been demonstrated experimentally (Laskin et al., 1970).
Perhaps the major limitation in assessment of the carcino-
genicity of arsenic is the lack of a consistent experimental
model. Although many experiments have been performed, princi-
pally with rats and mice, few have met with success. Those
studies that have shown positive effects (e.g., Knoth, 1966;
Kanisawa and Schroeder, 1969; Osswald and Goerttler, 1971) have
not been replicated or followed to completion. Difficulties in
experimental design and differences in metabolism and in the skin
of humans and animal species may account in part for the lack of
success in inducing arsenical cancer in animals under experi-
mental conditions. Also, storage of arsenic in the red blood
cells of the rat may render it an unsuitable species for modeling
arsenic carcinogenesis.
The lack of consistent positive evidence of animal carcino-
genicity attributable to arsenic is offset somewhat by experi-
mental evidence of mutagenicity, as discussed earlier, as well as
evidence of chromosomal abnormalities in human populations, where
both in vivo and in vitro studies have shown chromosomal aberra-
tions among people with histories of exposure to arsenic.
Sodium arsenite has been shown to inhibit postreplication DNA
repair in E. coli. These reports suggest that arsenic interferes
with fundamental genetic processes, possibly by binding to the
sulfhydryl groups of enzymes involved in genetic repair. The
effects on health of such interference are not .clear, but Jung
and Trachsel (1969) hypothesize that prolonged inhibition of DNA
repair by arsenic could lead to lesions that "may constitute the
starting point of carcinogenic changes." Since there is high
correlation between mutagenicity and carcinogenicity (Hollaender,
1976), these results lend credence to the idea that the inability
to demonstrate experimentally that arsenic is a carcinogen may be
a function of experimental design.
In addition to data on mutagenicity and carcinogenicity
there is a body of evidence demonstrating that inorganic arsenic
71
-------
is teratogenic. It is known that arsenic can cross the human
placenta and that the fetus can absorb and distribute it (Lugo
and Palmisano, 1969). Here, too, the health ramifications are
not understood, and there are no dosage data, or indeed any data
at all, on the effects of low-level human exposure on reproduc-
tion. It should be noted that no adverse reproductive effects
were reported in the studies of human acute ingestion summarized
in this document.
The distribution of arsenic and its metabolism in animals
and in humans are also incompletely understood. Distribution and
storage vary with different animal species and with route of
entry. Studies indicate, however, that arsenic inhibits a large
number of enzyme systems by binding to enzymes containing sulf-
hydryl groups, as discussed previously. The implications of
these findings for human health are not clear.
In addition to incomplete knowledge of the chemical and
biological action of arsenic, we have little information with
which to assess the effects on^humans of low-level chronic
exposure to arsenic. Only effects resulting from relatively high
levels of exposure have been reported. There are virtually no
studies of morbidity associated with low^-level exposure to
arsenic except for those by Benko, which have not been confirmed
by others.
.INGESTION OF ARSENIC
Several reports of the effects of ingesting high levels of
inorganic arsenic have been described. These reports leave
little room for doubt that ingestion of inorganic arsenic pro-
duces a variety of skin disorders, including skin cancer. In a
well-documented epidemiological investigation (Tseng et al.,
1968) established a clear dose-response relationship between the
amoung of arsenic ingested and the occurrence of skin cancer, as
noted previously.
Figure 1, taken from this study, shpws some of the observed
relationships. It also shows that the levels of arsenic in this
72
-------
study are extremely high. The report does not give enough data
to enable one to estimate the range of the arsenic dosage re-
quired to produce skin cancer.
The characteristic sequence of melanosis, keratosis, and
skin cancer reported by Tseng has also been observed in widely
different situations and different areas of the world, among
populations exposed to contaminated drinking water and in pa-
tients receiving Fowler's solution during medical treatment.
Some rough approximations of dosage can be made. As previously
noted, Neubauer's (1947) review of skin cancer in 143 patients
undergoing medicinal treatment with arsenical preparations (most
often Fowler's solution) affords the estimate that on the average
each patient ingested a total quantity of 28 g of arsenic (NAS,
1977), with an average induction period of about 18 years. The
effects of similar quantities of inhaled and ingested arsenic
cannot be easily equated since it has been demonstrated that
absorption of arsenic from the gastrointestinal tract is far less
*
effective than from the respiratory tract (Dutkiewicz, 1977;
Stevens et al., 1977).- Data on rats administered radio-labeled
arsenic are given in Table 3. Despite the ability of the rat to
store arsenic in the red blood cells, tfiese data are still useful
for demonstrating the differences in absorption by different
routes. That there is a difference is, of course, not surprising,
since a basic parameter of the toxicity of heavy metals (and
other substances) is route of entry. The toxicity of airborne
mercury vapor, for example, is markedly greater than that of
ingested mercury in humans (Browning, 1969).
We note that others have assumed that ingested and inhaled
arsenic are toxicologically equivalent (NIOSH, 1975; Smith et
al., 1977). NIOSH has recommended that the concentration of all
airborne particles, respirable and irrespirable, is to be deter-
mined for compliance with the occupational exposure standard.
Irrespirable particles may become trapped in the mucociliary
defense system and swallowed and then enter the gastrointestinal
73
-------
tract; these particles do not enter the bloodstream as effec-
tively as inhaled respirable particles.
OCCUPATIONAL EXPOSURE
Unlike the reports relating to ingestion of high levels of
arsenic, the literature on populations occupationally exposed to
airborne arsenic clearly shows, that arsenic is related to devel-
opment of lung cancer. Unfortunately, many of the available
epidemiological studies that give disease rates among the exposed
populations use the proportional mortality ratio (PMR) to calcu-
late disease rates. In a PMR study the number of deaths in a
population is expressed relative to the number of all deaths,
rather than to the population of the group. As noted by McMahon
(1970), a proportional rate determined in this way is only sug-
gestive of differences that may exist between a study population
and a comparison population, but no direct comparison can be
made. The PMR does not express the risk of members of the ex-
posed population and "until rates can be computed against a
population base, it will not be known whether the [observed]
differences relate to differences in the sizes of the numerators
or the denominators of the compared rates." Thus the absence of
quantifiable risk in such PMR studies presents a major stumbling
block to a health assessment and to the calculation of a dose-
response relationship, since a dose-response relationship is
defined as an increase in disease risk with an increase in amount
of exposure (McMahon, 1970).
As discussed earlier and summarized in Tables 3 and 4, many
of the studies of populations of arsenic-exposed workers were PMR
studies. The report of Hill and Faning (1948) of sheep-dip
workers should still be considered highly suggestive that lung
cancer and arsenic exposure are related, despite the limitations
in design. The air and urine levels measured by Perry et al.
(1948) and the evident signs of arsenicism make it clear that
this population was heavily exposed to arsenic. Unfortunately,
no dose-response data can be derived from this investigation.
74
-------
Ott et al. (1974) calculated a dose-response relationship
for arsenic and lung cancer. This study involves exposure to
arsenicals without concomitant sulfur dioxide exposure. Certain
problems in the study design, however, limit the applicability of
the dose-response relationship derived.
First, entry into the study population was based on employ-
ment for only 1 or more days in the arsenical work area of the
plant. Further, 138 of the 173 decedents in the study were
exposed to the pesticides for less than 1 year, and 16 of the 28
observed deaths due to respiratory cancer were among this group.
Another of these 28 decedents was a worker with occupational
asbestos exposure, and one of the decedents with more than 1 year
arsenic exposure was also occupationally exposed to asbestos.
Serious questions can be raised about inclusion of workers with
such short-term exposure in the study population. The study
population also included only those workers who had died while
employed by the company or had retired from the company. The
retrospective study also reported about one-third of the cohort
had been lost to followup. It is not unreasonable to expect the
same loss of workers in the PMR study, and it is difficult to
estimate the direction of the bias introduced. Since the work
with arsenical exposure was an entry-level job with often dis-
agreeable conditions, authors believe it likely that job transfer
could be expected, but whether the workers transferred to another
job or simply left the company is not known. This loss of
workers is, of course, the—reason that the authors resorted to a
PMR evaluation.
A more serious drawback is that to predict the expected
number of deaths as a function of age and year of death, Ott et
al. derived and used the following regression equation:
P(x1/x2) = -0.069 + 0.0037x1 - O.OOOSSx.^ - 0.014ex2 + 0.00051xlex2
- O.OOOOOSOx 2ex2
where x = age at death and x~ = (year of death - 1,940)/10.
75
-------
Despite its complicated six terms, the predictor equation ex-
plains only 57 percent of the variability of the data. The
authors provide no further statistical analysis of the equation,
such as a calculation of the least squares residuals, which might
indicate those/-ages at which the predictor equation fits the data
well and those at which it fits poorly. Furthermore, the stan-
dard deviation of the predicted proportion of deaths from respi-
ratory cancer among the control population is between about 25
and 33 percent of the predicted value, a wide error limit.
Because of the relatively poor fit of the predictor equa-
tion, it has limited usefulness for providing the expected
numbers of deaths for calculation of an observed/expected ratio.
Hence, quantitative estimates of a "response" for a dose-response
relationship cannot readily be made from this paper. The estima-
tions of dosages also are of limited value because they are
highly approximate. Ott and coworkers relied on 15 breathing-
zone samples taken in 1952. The samples ranges from 1.7 to 40.8
mg As/m in the drum dryer area and 0.26 to 7.5 mg As/m in the
packaging area. These measurements, which cover a range greater
than one order of magnitude, were used to derive time-weighted
averages, which were then extrapolated according to employee work
histories.
In summary, the study design, the predictor equation, and
the dosage estimates are such that the report cannot be used as
the basis for extrapolating a dose-response relationship. The
work of Ott et al. is, however, highly suggestive of a causal
relationship between lung cancer and arsenic exposure and is in
accord with other available evidence.
SMELTER WORKER MORTALITY RATE STUDIES
The strongest evidence that arsenic is a lung carcinogen,
possibly in conjunction with sulfur dioxide, comes from the
mortality studies of smelter workers. Lee and Fraumeni (1969);
Pinto et al. (1977); Pinto et al. (unpublished); Tokudome and
Kuratsune (1976); and Rencher and Carter (1977), for example,
76
-------
present similar results, derived from different smelter cohorts,
indicating an excess risk associated with exposure to arsenic
(and to sulfur dioxide). These studies further confirm that the
risk for lung cancer increases with exposure; that is, they
provide additional support for a dose-response relationship.
Blot and Faumeni (1975) obtained evidence of an increased rate of
lung cancer when they calculated an increased risk for lung
cancer among residents of counties in which copper smelters or
refineries are located.
The results of these smelter-worker studies are analyzed in
some detail here to develop an estimate of the extent of the risk
observed and the dosages with which these risks can be-~corre-
lated. The estimates of dosages also involve an analysis of
serious deficiencies in both environmental and biological moni-
toring of arsenic exposure.
The rates calculated by several authors for respiratory
cancer are given in Table 11. They are all significantly ele-
vated. When rates are calculated as a function of duration and
degree of exposure in Table 12, a definite gradient is observed.
All the authors state that a similar gradient is observable with
respect to concurrent sulfur dioxide exposure. It is important
to note that the gradients shown in Table 12 should not be com-
pared, since the classifications "heavy," "medium," and "light"
do not refer to the same degrees of arsenic exposure. They are
purely relative indices taken from the original sources.
Once a dose-response gradient is observed, it is important
to quantify the dose leading to the response. Accurate quantifi-
cation depends on accurate environmental and/or biological
assays. For this reason, it is appropriate to discuss relevant
parameters of such monitoring before doses are extrapolated.
BIOLOGICAL MONITORING
Inherent in the assumption that one can extrapolate an index
of exposure from urinary arsenic levels to ambient arsenic is
reliance on a correlation between urinary arsenic levels and
77
-------
TABLE 11. STANDARDIZED RESPIRATORY CANCER MORTALITY RATES
OBSERVED AMONG SEVERAL SMELTER-WORKER COHORTS
oo
Tokudome and Kuratsune
'(1976)
Lee and Fraumeni (1969)
Pinto et al. (1977)
Rencher and Carter
(1976)
No. of cases
29
61 (>15 yr
before 1930)
37 (>15 yr
1938-63)
10 (10-14 yr)
15 (5-9 yr)
24 (1-4 yr)
32
17
SMR
1189a
469b
W
370b
23 £
268°
20311
305
10.1/10,000°
Utah
3.3/10,000
l
Comments
Dose-response relationship obtained for length of
employment and degree of exposure; no relationship
observed between latency period and dose
Dose-response relationship obtained for length of
employment and degree of exposure
Dose-response relationship obtained for length of
employment and degree of exposure; only deaths among
retirees included
Cohort included all workers
Significant at 5% level.
Significant at 1% level.
Statistically significant level not stated.
-------
TABLE 12. RESPIRATORY CANCER SMR's AS A FUNCTION OF
DURATION AND DEGREE OF EXPOSURE
All cohorts:
Tokudome and Kuratsune
(1976)
Lee and Fraumeni
(1969)
Cohorts by length of exposure:
Tokudome and Kuratsune
(1976)
Lee and Fraumeni
(1969)
Pinto et al.
(1977)'
Heavya
1485b
667b
2048b
>15 yr prior
to 1948
800b
>15 yr prior
to 1938
600C
<25 yr
833C
>25 yr
Medium3
1250b
478b
1064b
>15 yr
667b
>15 yr
267
<25 yr
364C
>25 yr
Light3
615b
239b
614b
1-14 yr
250b
1-14 yr
95
<25 yr
278C
>25 yr
-J
VO
Exposure categories are not strictly comparable because they represent different
actual values. They are taken from the authors' estimates and should be considered
relative to each other in each report.
Significant at 1% level.
Significant at 5% level. .
-------
airborne arsenic. The recent epidemic-logical investigations of
Pinto et al. (unpublished) explicitly derive an arsenic exposure
index from urinary arsenic levels. The air/urine correlation
cited in Figure 2 is that of Pinto et al. (1976). Our examina-
tion of the data presented in Pinto et al. (1976) shows, in fact,
a poor correlation between urinary arsenic levels and airborne
arsenic for all exposures less than 60 yg/m , despite the statis-
tically significant correlation reported. Figure 2 illustrates
the correlation presented by Pinto et al.
Although Pinto and coworkers calculated a significance level
of p<0.01, this significance level has little meaning for the
lower end of the regression line, since it does not correlate
well with the data. At these values, the nonrandom scatter of
the points in Figure 2 illustrates this poor fit.
The mean and standard deviation of the data, 170 and 113,
respectively, also show wide variability in the relationship
between airborne arsenic levels and urinary arsenic levels. A
similar conclusion has been reached by others (NIOSH, 1975).
Thus, in order to estimate the exposure indices and dosages of
workers at risk for lung cancer, wide error limits must be
applied to the calculated exposures. For purposes of this
document a 7-fold to 10-fold variation will be assumed, since
that is the approximate range of the urinary arsenic values
corresponding to very small changes in airborne levels shown in
Figure 2.
AIR SAMPLING
Even greater error limits must be applied to the urine/air
correlations to account for errors in the methods of air anal'-
ysis. Lao et al. (1974) clearly demonstrate that the physical
properties of arsenic require a differential between temperatures
of the ambient air and of the collector. If the collector is not
at a lower temperature, volatilization of the As20_ collected (as
As.O..) will occur and. as much as 90 percent of the sample may be
4 o
80
-------
oo
a:
DC
s
a:
180
160
140
120
100
80
60
40
20
CORRELATION 0.530
REGRESSION Y = 0.304X
NOTE: ONE DATA POINT
•(224,295) NOT SHOWN
40 120 200 280 360 440 520
URINARY ARSENIC, ug/liter
Figure 2. Comparison of urinary arsenic excretion and
concentration of inhaled arsenic calculated by Pinto
et al. (1977) .
81
-------
lost. The National Academy of Sciences (1977) warns in its
Recommendations that unless the methods of Lao et al. are fol-
lowed, air sampling measurements will be seriously deficient.
The purpose of this assessment is to extrapolate the health
effects of arsenic in the best way feasible and not simply to
dismiss the data. Thus air sampling data presented here will be
considered to be within an order of magnitude of the correct
value, a justifiable approximation given the constraints pointed
out by Lao and others. Although such precision is of little
value to an analytical chemist charged with determining compli-
ance with ambient air levels, it is certainly .within the degree
of precision with which all the epidemiological and clinical data
presented here can be evaluated.
Thus, taken together, approximately a 70- to 100-fold range
and variability must be considered for urine/air correlations and
the indices derived from them.
EVALUATING THE EXPOSURE
With the constraints of a 70- to 100-fold range for urine/
air correlations and 7- to 10-fold and 10-fold ranges for urine
and air measurements, respectively, extrapolations of dosages can
be made on the basis of the reports cited in Table 12. The
dosages are given in Tables 13 and 14.
Table 13, taken from NIOSH (1975), gives the ranges used by
Lee and Fraumeni (1969) for their classification of exposures.
There is great variability in the data. The calculated standard
deviations for the areas numbered 1, 2a, 2b, and 3a, respectively,
are 3.47, 1.74, 2.54, and.0.48.
Exposure levels are not given by Tokudome and Kuratsune
(1976), and those of Rencher and Carter (1977) are not well
defined, with no units given.
Table 14 is taken from Pinto et al. (1976). The exposure
-index calculated is a relative index constructed from the average
urinary arsenic levels found in workers in 33 departments of the
Tacoma smelter and the employment history for each of the 525
82
-------
TABLE 13. ATMOSPHERIC ARSENIC CONCENTRATIONS
IN 1965 SMELTER SURVEYa
_ _ _ , _ . ._.. .... _ *j _„_ .._
(mg As/m )
"Heavy exposure area" as classified by Lee and Frauiaeni
Arsenic roaster area
0.10
0.10
0.10
0.10
0.10
0.10
0.17
0.20
0.22
0.25
0.35
1.18
5.00
12.66
Mean: 1.47
Median: 0.185
"Medium exposure areas" as classified by Lee and Fraumeni
Reverberatory area
0.03
0.22
0.23
0.36
0.56
0.63
0.66
0.76
0.78
0.78
0.80
0.83
0.93
1.00
1.27
1.60
1.66
1.84
1.94
2.06
2.76
3.40
4.14
8.20
Treater building and arsenic loading
0.10
0.10
0.10
0.11
0.48
0.62
3.26
7.20
Mean: 1.56
Median: 0.88
Mean: 1.50
Median: 0.295
'Light exposure areas" as classified by Lee and Fraumeni
Copper concentrate transfer system
0.25
0.65
1.20
Samples from flue station
0.10
0.24
Reactor building
0.001 0.003
0.002 0.009
0.002 0.010
0.002
Mean:
Median:
Mean:
Median:
Mean:
Median:
0.70
0.65
0.17
0.17
0.004
0.002
Source: National Institute of Occupational Safety and
Health, 1975.
83
-------
TABLE 14. RELATIONSHIP BETWEEN ARSENIC EXPOSURE AND LUNG CANCER
MORTALITY CALCULATED BY PINTO et al. (1977)
Exposure index
(mean index)
<2000(1514)
2000-2999(2513)
3000-5999(4317)
6000-8999(7473)
9000-11,999(10,135)
>12, 000(14, 712)
Respiratory cancer deaths
No. o£ men
36
109
205
109
38
29
Observed
1
4
11
7
4
5
Expected
0.9
2.1
3.9
2.3
0.7
0.6
SMR
111.1
190.5
282. Oa
304. 3a
571. 4a
833. 3a
p <0.05
84
-------
retirees included in the study. Thus the index is a measure of
intensity, not arsenical exposure. Table 15 gives the corres-
ponding mean urinary levels, which appear in Pinto et al. (un-
published) .
Applying the regression fit calculated by Pinto et al.
(1977) to data also provided by Pinto et al. (1977), which
calculates SMR as a function of duration and intensity of ex-
posure, we obtain the air levels shown in Table 16. In order to
be most conservative in the extrapolation and thus ensure the
greatest margin of safety, we shall use the correspondence be-
tween an SMR of 833 and an air level of 106 yg/m for further
calculation.
Assuming that the worker experiencing a mortality rate of
833 is exposed to the workplace environment about 22 percent of
the year (40 hours/week; 3 weeks vacation and leave), the 106
yg/m is equivalent to an air level of 23 yg/m in the environ-
ment for 24 hours per day all year. This figure must be further
tempered because Pinto et al. (1977) reported only on the mor-
tality experience of retirees. An upward bias is thus intro-
duced, because some men who worked at the refinery but left may
also have died of respiratory cancer. It can also be argued that
such reporting pushes the estimate downward, in that smelter
workers probably tend to live in arsenic-polluted environments
and hence continue to be exposed to levels of arsenic greater
than that to which the general population is exposed when not at
work. It is difficult to weigh these biases quantitatively.
If we next consider the Lee and Fraumeni report, we see
that, taking the heaviest exposure and the corresponding SMR, a
median of 0.185 mg As/m and a mean level of 1.47 mg As/m
resulted in an SMR of 800 (significant at the 1% level). Using
the same approximation of' 22 percent (see above) this is equiva-
lent to a 24-hour exposure level between 41 and 323 yg/m .
Considering the approximations, the unreliability of the
measurement methods, the loss of followup of cohort, the vari-
ability in the air levels used by Lee and Fraumeni and Pinto et
85
-------
TABLE 15. MEAN URINARY ARSENIC VALUES USED TO
CONSTRUCT TABLE 14a
Department
Cottrell
Arsenic plant
Roaster
Boiler room
(waste heat)
Janitor
Repair
Steel shop
Reverberatory
Mason
Carpenter
Yard
Crushing plant
Pipe shop
Slimes
Laundry
Acid plant
Electric shop
Lead burners
Converter
Fire
Mobile equipment
Martin mill
Track
Machine shop
Mobile equipment
repair
Anode
Refinery
Office
Warehouse
Sample dept.
Watchman
Refined casting
Number of samples
1948-52
62
222
77
7
10
25
81
36
5
94
85
0
6
27
6
6
10
9
6
0
0
0
0
5
0
4
13
30
2
10
7
7
1973
48
85
103
30
7
41
46
38
20
41
17
16
22
26
3
13
24
8
52
2
23
20
6
16
9
36
157
58
5
16
13
56
Average for 26
departments in which
samples were taken in
both periods.
Concentration in
urine, vq As/liter
1948-52
553
804
556
787
176
328
624
346
270
219
506
1258
127
115
142
354
2144
161
~~*
120
103
103
108
80
215
135
67
400
1973
526
516
414
409
289
288
272
269
260
255
226
222
218
207
201
180
171
166
160
140
119
115
112
10.2
101
98
98
88
82
81
77
58
220
Percent
change
- 5
-36
-26
-48
+64
-12
-56
-22
- 4
+16
-55
-83
+63
+75
+27
-52
-92
0
-15
- 5
- 5
-19
+ 3
-62
-43
-13
-95
a Pinto et al. in press. Values are averages by department.
86
-------
TABLE 16. DERIVATION OF AIR LEVEL EQUIVALENTS FROM URINARY ARSENIC
LEVELS AND CORRESPONDING SMR's AS A FUNCTION OF DURATION
Calculated air
level, vg/m^
15.2 - 60.5
60.8 - 106
>106
Intensity of
exposure,
yg/liter urine
50 - 199
200 - 349
_>350
Duration
<25 years
Obs.
2
4
3
Exp.
2.1
1.5
0.5
SMR
95
267
600
>25 years
Obs.
10
8
5
Exp.
3.2
2.2
0.6
SMR
278
364
833
GO
-J
Using the regression fit y = 0.304X, calculated from Pinto, et al. (1976)
-------
al., as well as all the other constraints already mentioned, the
ratios obtained from the Pinto et al. and the Lee and Fraumeni
studies of air levels per SMR of 23 yg As/m per 833 and 41 to
323 yg As/m per 800, respectively, are still in reasonable
agreement with each other.
Other Cancer Risks
There is evidence that arsenic exposure is related to cancer
of the digestive tract, possibly arising from the swallowing of
irrespirable particles trapped by the muco-ciliary defense sys-
tem. Tokudome and Kuratsune (1976) calculated an SMR of 508 for
digestive tract cancer, significant at the 1 percent level, among
copper smelter workers; Milham (1977) obtained a PMR of 162, for
cancer of the large intestine, significant at the 5 percent
level. Lee and Fraumeni (1969) did not calculate an elevated SMR
(101); however, they did not classify the deaths by exposure and
an effect observable among those with the greatest and longest
exposure may have been obscured. This is plausible since they
did obtain such an exposure/duration gradient for lung cancer.
Pinto et al. (1977) calculated an SMR of 122 for digestive can-
cer, but it was not statistically significant. At this point it
is not possible to calculate an equivalent dose because corres-
ponding air levels are not available.
It is also interesting to note that Rencher and Carter
(1977) and Pinto et al. (unpublished) calculate that nonsmokers
are at greater risk of lung cancer than smokers, as shown in
Table 17. Since the mortality rate from lung cancer of the
smoking retirees is more than 3-fold that of the nonsmoking
retirees, it is still an open question as to whether this is a
true observation or just the result of earlier death among smok-
ing smelter workers with resultant loss to followup.
ASSESSMENT OF THE EFFECTS OF COMMUNITY EXPOSURE
Throughout this assessment we have stressed that one cannot
state categorically that arsenic alone induces lung cancer and
88
-------
TABLE 17. RESPIRATORY CANCER MORTALITY AND
SMOKING AMONG SMELTER WORKERS
Observed respiratory cancer deaths among 377 men alive on
January 1, 1961a
Smoking status
Smokers
Ex-smokers
Nonsmokers
No. of retirees
189
69
119
No. of deaths
15
3
3
SMR
287. 3b
245.1
506. 5b
Percentage of deaths due to lung cancer by location
Nonsmokers
Smokers
Smelter
3.3
9.2
Mine
0.7
-3.3
Other
0.8
3.3
Source: Pinto et al. (in press).
p<0.05.
Source: Rencher, Carter, and McKee (1977) .
89
-------
that the presence of ubiquitous contaminants must also be con-
sidered. The major environmental sources of airborne arsenic,
such as smelters and coal-fired generators, usually do not emit
arsenic alone to the community, but also emit the contaminants
encountered in the workplace environments studied. Thus it can
be expected that the etiological relationships observed in the
workplace between arsenic and lung cancer in the presence of
contaminants will be observed wherever such contaminants occur
together, including the general environment. It is relevant that
high-temperature generation of arsenic and heavy metals, such as
from power plants, produces particles that are in the respirable
size range and hence can effectively penetrate the respiratory
system and enter the bloodstream (Natusch and Wallace, 1974).
Further confirmation of this point is provided by Pinto and
McGill (1953), who state that a "substantial part" of the arsenic
trioxide emitted by smelters is over 5.5 um in diameter. This
means that 77 percent is below 5.5. ym and much of this is in the
respirable range.
Additional evidence of community exposure to arsenic is
found in the elevated levels of urinary arsenic reported in
communities near copper smelters. Milham and Strong (1974) found
that urinary arsenic levels of schoolchildren residing near the
Tacoma smelter were comparable to those of smelter workers. As
noted earlier, however, this study failed to take into account
ingestion of arsenic in seafood or specific gravity of the spot
urine samples. Elevated community levels are also confirmed in a
fashion by Pinto and McGill (1953), who give 0.13 pg/liter
urinary arsenic as the background level among nonsmelter workers
assumed not to be exposed to arsenic; later studies suggest that
this level is indicative of exposure. Apparently these workers
lived in the community near the Tacoma smelter, which may account
for their exposure. Furthermore, high levels of sulfur dioxide
were noted in the area surrounding the Ronnskar smelter works
(Pershagen et al., 1977), and it is likely that communities
90
-------
adjacent to smelters are exposed to significant amounts of sulfur
dioxide in combination with arsenic.
The report by Blot and Fraumeni (1975) supports the con-
clusion that the elevated ambient levels of arsenic and the
elevated urinary levels of arsenic observed in communities near
smelters lead to an excess of lung cancer. Their results, dis-
cussed earlier, are summarized in Table 18.
Nelson (1977) has criticized the findings of Blot and
Fraumeni on the basis that they made no distinction between
smelters and refineries. This is a valid criticism, since
smelters are responsible for the greatest amount of arsenic
pollution. When the data of Blot and Fraumeni are recalculated
leaving out the four counties not cited by Nelson as containing
smelters, the figures in Table 19 are obtained, showing that
removal of counties with refineries does not affect the result.
Nelson also cites the data in Table 18 and derives the data
shown in Table 20 (the mortality rates shown as SMR1s are just
mortality rates and not SMR's) in order to establish that there
is no relationship between arsenic levels and community rates.
No direct extrapolation can be made, however, for several rea-
sons. One is that the use of national mortality rates for com-
parison purposes is not as good as the use of demographically
derived comparison rates. The second is that distribution of the
arsenic in the entire county is a function of the size of the
county, the geographical location of the smelter, and meteoro-
logical factors. For example, the Tacoma smelter, which has the
highest percentage of arsenic in the feed, is located in the
northwest corner of a large county. If we look at the county/
state/SMR breakdowns (Table 20) provided by Blot and Fraumeni in
their original draft, but not published in the final report, the
relative populations become apparent. For these geographical and
demographic reasons it is also inappropriate to calculate the
ambient levels that may correspond to the increased mortality
rates.
91
-------
TABLE 18. LUNG CANCER MORTALITY RATES AMONG WHITE MALES
IN COUNTIES WITH COPPER SMELTERSa
(showing national average)
County
Lung cancer
mortality rate
Deer Lodge, Montana
Gila, Arizona
Pima, Arizona
Cochise, Arizona
National average
Pierce, Washington
El Paso, Texas
Greenlee, Arizona
Pinal, Arizona
Ontonagon, Michigan
Polk, Tennessee
Grant, New Mexico
Salt Lake, Utah
White Pine, Nevada
65.2
46.3
39.7
38.1
38.0
35.8
33.9
32.0
31.7
29.2
28.8
26.3
26.2
20.0
Source: U.S. Cancer Mortality by County:
U.S. DHEW, Publication No. (NIH) 74615.
1950-1969,
92
-------
TABLE 19. DISTRIBUTION OP LUNG CANCER SMR's IN U.S. COUNTIES WITH.
COPPER SMELTERS AND REFINERIES (SR), AND WITH ONLY COPPER SMELTERS'
vo
U)
Male SMR
-90
90 - 110
110+
Total
-90
1
1
0
0
3
3
4
4
90-110
0
0
1
2
1
3
2
5
110+
2
2
1
2
4
5
7
9
Total
3
3
2
4
8
11
14
18
Values for counties with both smelters and refineries are in upper
portion of each box. Table calculated and adapted from Blot and
Fraumeni (1969).
-------
TABLE 20. ARSENIC IN SMELTER FEEDS AND LUNG CANCER RATES
Smelter (company)
Tacoma (Asarco)
Anaconda (Anaconda)
El Paso (Asarco)
Garfield (Kennecott)
Hayden (Asarco)
Hayden (Kennecott)
San Manuel (Magma)
Hurley (Kennecott)
White Pine (Copper Range)
U.S. average
Arsenic in feed, %
5.200
0.96
0.800
0.135
0.040
0.015
0.007
0.005
0.002
Lung cancer /
SMRa — 7W '
Male
35.8
65.2
33.9
26.2
46.3
46.3
31.7
26.3
29.2
37.98
Female
6.4
4.3
7.6
3.5
7.3
7.3
7.8
10.8
2.0
6.29
(sic) taken from Nelson (1977) .
94
-------
Lyon et al. (1977) recently reported on community lung
cancer rates and arsenical pollution. The authors found no
excess lung cancer mortality among residents of Salt Lake County,
Utah. This finding agrees with those of Blot and Fraumeni for
this county (Table 19). There is currently no explanation of the
differences between this smelter county and the others shown in
Table 21.
95
-------
TABLE 21. U.S. COUNTIES ENGAGED IN THE PRIMARY.
SMELTING AND REFINING OF NONFERROUS ORES IN 1963£
County/State
Cochise, Arizona
Gila, Arizona
Greenlee, Arizona
Pima, Arizona
Final, Arizona
Anne Arundel, Maryland
Baltimore City, Maryland
Roughton, Michigan
Ontonagon, Michigan
Cascade, Montana
Deer Lodge, Montana
White Pine, Nevada
Middlesex, New Jersey
Grant, New Mexico
Polk, Tennessee
El Paso, Texas
Salt Lake, Utah
Pierce, Washington
Population
in 1960
55,039
25,745
11,509
265,660
62,673
206,634
939,024
35,654
10,584
73,418
18,640
9^,808
433,856
18,700
12,160
314,070
383,035
321,590
Estimated No.
of smelter-
refiner
workers
in 1963
750
1300
175
175
350
750
750
210
175
1160
925
375
- 4685
210
75
1500
2250
760
1950-69
Lunq cancer SMR
Male
161
154
138
136
106
118
109
105
130
134
246
78
121
114
115
89
75
100
Female
130
145
46
121
156
108
102
132
43
110
85
123
93
238
90
115
58
97
Source: Blot and Fraumeni (1975).
96
-------
SECTION 7
REFERENCES
1. American Conference of Industrial Hygienists. Documentation
of the Threshold Substances Added or Changed Since 1971.
Cincinnati: Amer. Con. of Ind. Hygienists, 1976.
2. Ancel, P. Recherche Experimentale Sur Le Spina Bifida.
Arch. Anat. Mier. Morph. Exp 36:45, 1946.
3. Argentine Association of Dermatology and Syphilology and Its
Affiliates. Argentine Review of Dermatosyphilology.
Buenos Aires: Frascolli & Bindi, 1938. (Trans, for EPA by
Nundus Systems, Washington, D.C.)
4. Arguello, R.A., D.D. Cenget, E.E. Tello. Y Cancer Arseni-
cismo Regional Endemico En Cordoba. Argentine Dermatosifilol
22: 461-487, 1938.
5. Arsenault, R.D. Health Aspects of C.C.A. Wood Preserva-
tives— A Review of Arsenates and Cancer. BWPA Annual
Convention, 1977.
6. Baetjar, A.M., M.L. Levin, A. Lilienfeld. Analysis of
Mortality Experience of Allied Chemical Plant — 2000 Race
Street, Baltimore, Maryland. Unpublished report submitted
to the Allied Chemical Corp., Morristown, New Jersey, July
16, 1974. (Summarized in NIOSH, 1975.)
7. Baron, D., I. Kunick, I. Frischmuth, J. Petres. Further In
Vitro Studies on the Biochemistry of the Inhibition of
Nucleic Acid and Protein Synthesis Induced by Arsenic. Arch
Derm Res 253:15-22, 1975.
8. Baroni, C., G.J. Van Esch, U. Saffiotti. Carcinogenesis
Tests of Two Inorganic Arsenicals, Arch of Environ. Health
7:668-673, 1963.
9. Barry, K.G., E.G. Herndon. Electrocardiographic Changes
Associated with Acute Arsenic Poisoning. Med Ann Dist of
Columbia 31:25-27, 76-66, 1962.
97
-------
10. Beaudoin, A.R. Teratogenicity of Sodium Arsenate in Rats.
Teratology 10:153, 1974.
11. Beckman, G., L. Beckman, I. Nordenson. Chromosome Aberrations
in Workers Exposed to Arsenic. pp. 145-146. In National
Institute of Environmental Health Perspectives, Vol. 19.
Research Triangle Park, North Carolina. U.S. Department of
HEW, Public Health Service, National Institute of Health.
Publication Number (NIH) 77-218, 1977.
12. Bencko, V. Arzen Vo Vlasoch Neprofesionalne Exponovanej
Populacie (Arsenic In Hair of Professionally Not Exposed
Population). Ceskoslovenska Hygiena, 11:539-543, 1966.
13. Bencko, V., K. Nejedly, J. Scmora. Histological Examination
of Some Organs After the Peroral Administration of Arsenic
to Hairless Mice. C.S. Hyg. 13:344, 1968.
14. Bencko, V., A. Dobisova, M. Macaj. Arsenic in the Hair of
a Non-Occupationally Exposed Population. Atmos. Envir.
Pergamon Press 5:275-279, 1971.
15. Bencko, V. Carcinogenic, Teratogenic, and Mutagenic Effects
of Arsenic, pp. 179-182. In National Institute of Environ-
mental Health Sciences (Ed.), Environmental Health Perspec-
tives, Vol. 19. Research Triangle Park, North Carolina.
U.S. Dept. of HEW, Public Health Service, National Institute
of Health. Publication Number (NIH) 77-218, 1977.
16. Bencko, V., K. Symon. Test of Environmental Exposure to
Arsenic and Hearing Changes in Exposed Children, pp. 95-
101. In National Institute of Environmental Health Sciences
(Ed.). Environmental Health Perspectives, Vol. 19. Research
Triangle Park, North Carolina. U.S. Department of HEW,
Public Health Service, National Institute of Health.
Publication Number (NIH) 77-218, 1977.
17. Bergoglio, R.M. Mortality From Cancer in Regions of Arsen-
ical Waters of the Province of Cordoba Argentine Republic.
Pren. Med. Argent. 51:994-998, 1964.
18. Biomedical Studies Group. Review of the Environmental
Effects of Arsenic. Oak Ridge National Laboratory, Oak
Ridge, Tennessee. Operated by Union Carbide Corp., for the
Energy Research and Development Administration, Contract No.
W-7405-Eng.-26, 1976.
98
-------
19. Blejer, H.P., W. Wagner. Case Study 4. Inorganic Arsenic
Ambient Level Approach to the Control of Occupational
Cancerigenic Exposures, pp. 179-186. In Saffiotti, U.,
J.K. Wagoner, Occupational Carcinogenesis. Annals of the
New York Academy of Sciences, Vol. 271. New York, New York,
1976.
20. Blot, W.J., J.F. Fraumeni. Arsenical Air Pollution and Lung
Cancer. Lancet: 142-144, July 26, 1975.
21. Borgono, J.M., R. Greiber. Epidemiological Study of Arseni-
cism in the City of Antofagasta. pp. 13-24. In Hemphill,
D.D. (Ed.). Trace Substances in Environmental Health V.
Proceedings of Univ. of Missouri's 5th Annual Conference on
Trace Substances in Environmental Health. June 29 - July 1,
1971. Columbia Univ. of Missouri, 1972.
22. Borgono, J.M., P. Vicent, H. Venturino, A. Infante. Arsenic
in the Drinking Water of the City of Antofagasta. Epidemio-
logical and Clinical Study Before and After the Installation
of the Treatment Plant, pp. 103-105. In National Institute
of Environmental Health Sciences (Ed.). Environmental
Health Perspectives, Vol. 19. Research Triangle Park, North
Carolina. U.S. Dept. of HEW, Public Health Service, Nat.
Inst. of Health. Pub. No. (NIH) 77-218, 1977.
23. Boutwell, R.K. A Carcinogenicity Evaluation of Potassium
Arsenite and Arsanilic Acid.. Agr. Food Chem. 11:381-385,
1963.
24. Braman, R.S., C.C. Foreback. Methylated Forms of Arsenic in
the Environment. Science 182:1247-1249, 1973.
25. Braun, W. Carcinoma of the Skin and the Internal Organs
Caused by Arsenic. Germ. Med. Monthly 3:321-324, 1958.
26. Browning, E. Toxicity of Industrial Metals. New York.
Appleton-Century-Crofts, 1969.
27. Buchanan, R.J.M. Cases of Arsenical Peripheral Neuritis.
Lancet 170-172, January 19, 1901.
28. Buchanan, W.D. Toxicity of Arsenic Compounds. Amsterdam:
Elsevier Pub. Co., 1962.
29. Buechley, R.W. Bibliography on Health Effects of Arsenic.
U.S. Department of Health, Education, and Welfare, 411 West
Chapel Hill Street, Durham, North Carolina 27701. Unpub-
lished.
99
-------
30. Burgdorf, W.H. Sister Chromatid Exchanges in Patients with
Chronic Arsenic Use. Unpublished Masters Thesis, 1977.
University of Minnesota.
31. Butzengeiger, K.H. Uber Die Chronische Arsenvergiftung.
Deut Arch Fur Klinische Medizin 194:1-16, 1947.
32. Byron, W.R., G.W. Bierbower, J.B. Brouwer, W.H. Hansen.
Pathologic Changes in Rats and Dogs From Two-Year Feeding of
Sodium Arsenite or Sodium Arsenate. Tox. and App. Phar.
10:132-147, 1967.
33. Carcinogen Assessment Group, Assessment of Cacodylic Acid.
(Unpublished report supplied by EPA.)
34. Carcinogen Assessment Group, Preliminary Report on Inorganic
Arsenic. (Unpublished report supplied by EPA.)
35. Carnow, B.W., ed. Health Effects of Occupational Lead and
Arsenic Exposure. U.S. Department of Health, Education, and
Welfare. Washington, D.C. Superintendent of Documents,
U.S. Government Printing Office, 1976.
36. Clay, J.E., I. Dale, J.D. Cross. Arsenic Absorption in
Steel Bronze Workers. J. Soc. Occup. Med. 27:102-104, 1977.
37. Colucci, A.V. Epidemiology of Pesticide and Metal Residues.
pp. 1043-1051. In Comm. of the European Communities; WHO;
EPA (Eds.). International Symposium Proceedings. Recent
Advances in the Assessment of the Health Effects of Environ-
mental Pollution, Vol. 2. Paris: Comm. of the European
Communities; WHO; EPA, 1974.
38. Crecelius, E.A., M.H. Bothner, R. Carpenter. Geochemistries
of Arsenic, Antimony, Mercury, and Related Elements in
Sediments of Puget Sound. Environ. Sc. & Tech. 9:325-333,
1975.
39. Crecelius, E.A. Arsenite and Arsenate Levels in Wine.
Bull. Environ. Contamination and Toxicology, 1977.
40. Crecelius, E.A. Changes in the Chemical Speciation of
Arsenic Following Ingestion by Man. pp. 147-150. In
National Institute of Environmental Health Sciences (Ed.).
Environmental Health Perspectives, Vol. 19. Research
Triangle Park, North Carolina. U.S. Dept. of HEW, Public
Health Service, Nat. Inst. of Health. Pub. No. (NIH) 77-
218, 1977.
100
-------
41. Crema, A. Distribution et elimination de I1arsenic 76 chez
la souris normale et cance'reuse. Arch. int. pharmacodyn.
103: 57-70, 1955.
42. Currie, A.N. The Role of Arsenic in Carcinogenesis. Brit.
Med. Bull. 4: 402-405, 1947.
43. Dubois, L., T. Teichman, J.L. Monkman. The Normal Value of
Arsenic in Human Hair. Proc. Canadian Soc. of Forensic
Science 4:217-231, 1965.
44. Ducoff, H.S., W.B. Neal, R.L. Straube, L.O. Jacobson, A.M.
Brues. Biological Studies With Arsenic. II. Excretion and
Tissue Localization. Proc. Soc. Exp. Biol. Med. 69:548-554,
1948.
45. DuPont, 0., I. Ariel, S.L. Warren. The Distribution of
Radioactive Arsenic in the Normal and Tumor-bearing (Brown-
pearce) Rabbit. Am. J. Syphilis. 26:96-118, 1942.
46. Dutkiewicz, T. Experimental Studies on Arsenic Absorption
Routes in Rats. pp. 173-177. In National Institute of
Environmental Health Sciences (Ed.). Environmental Health
Perspectives, Vol. 19. Research Triangle Park, North
Carolina. U.S. Dept. of HEW, Public Health Service, Nat.
Inst. of Health. Pub. No. (NIH) 77-218, 1977.
47. Eiji, H. Funnyu Ni Yoru Nyuji Hiso Chudokusho (Infant
Arsenic Poisoning by Powdered Milk). Nihon Iji Shimpo Japan
Med. J. 1649:3-12, 1955. Translated for EPA by Leo Kanner
Assoc. TR105-74.
48. Enterline, P.E., S.S. Pinto, V. Henderson. Cancer Among
Arsenic Exposed Workers in a Copper Smelter.
49. Fairhall, L., J.W. Miller. A Study of the Relative Toxicity
of the Molecular Components of Lead Arsenate. Publ. Hlth.
Report, 56:1610-1625, 1942.
50. Falk, H.L., P. Kotin. An Assessment of Factors Concerned
With the Carcinogenic Properties of Air Pollutants. Nat.
Cancer Inst. Mon. 9:81-89, 1961.
51. Feinglass, E.J. Arsenic Intoxication From Wellwater in the
United States. New Eng. J. of Med. 288:828-830, 1973.
52. Ferm, V.H., S.J. Carpenter. Malformations Induced by
Sodium Arsenate. J. Reprod. Fertil. 17:199, 1968.
101
-------
53. Perm, V.H., A. Saxon, B.M. Smith. The Teratogenic Profile
of Sodium Arsenate in the Golden Hamster. Arch. Environ.
Health 22:557-560, 1971.
54. Perm, V.H. Arsenic as a Teratogenic Agent. pp. 215-217. In
National Institute of Environmental Health Sciences (Ed.).
Environmental Health Perspectives, Vol. 19. Research
Triangle Park, North Carolina. U.S. Dept. of HEW, Public
Health Service, Nat. Inst. of Health. Pub. No. (NIH) 77-
218, 1977.
55. Fierz, L. Katamnestische Untersuchungen Uber Die Neben-
wirkungen Der Therapie Mit Anorganischem Arsen Bei Hautkrank-
heiten (Catamnestic Investigations of the Side Effects of
Therapy of Skin Diseases With Inorganic Arsenic). Schweiz
Ges Derm Verner, 46th Annual Meeting, Lausanne, 1964.
Dermatologica 131:41-58, 1965. Translated for EPA by SCITRAN
- TR251-75.
56. Fowler, B.A. International Conference on Environmental
Arsenic. An Overview. pp. 239-242. In National Institute
of Environmental Health Sciences (Ed.). Environmental
Health Perspectives, Vol. 19. Rsearch Triangle Park, North
Carolina. U.S. Dept. of HEW, Public Health Service, Nat.
Inst. of Health. Pub. No. (NIH) 77-218, 1977.
57. Fowler, B.A., J.B. Weissberg. Arsine Poisoning. New Eng.
J. Med. 291:1171-1174, 1974.
58. Fowler, B.A., J.S. Woods, C.M. Schiller. Ultrastructural
and Biochemical Effects of Prolonged Oral Arsenic Exposure
on Liver Mitochondria of Rats. pp. 197-204. In National
Institute of Environmental Health Sciences (Ed.). Environ-
mental Health Perspectives, Vol. 19. Research Triangle
Park, North Carolina. U.S. Dept. of HEW, Public Health
Service, Nat. Inst. of Health. Pub. No. (NIH) 77-218, 1977.
59. Franseen, C.C., G.W. Taylor. Arsenical Keratoses and Car-
cinomas. Am. J. Cancer 22:287-307, 1934.
60. Fraumeni, J.F. Respiratory Carcinogenesis: An Epidemic-
logic Appraisal. J. Natl. Cancer Inst. 55:1039-1046, 1975.
61. Friedrich, E.G. Vulvar Carcinoma In Situ in Identical
Twins— An Occupational Hazard. Obstetrics and Gynecology
39:837-841, 1972.
62. Frohn, W. Occupational Arsenic Poisoning of Wine Growers.
Munchener Medizinishe Wochenschrift 42: 1630-1635, 1938.
102
-------
63. Frost, D.V. Arsenicals in Biology - Retrospect and Prospect,
Fed. Proc. 26:194-208, 1967.
64. Gainer, J.H. Effects of Arsenicals on Interferon Formation
and Action. Am. J. Vet. Res. 33:2579-2586, 1972.
65. Gainer, J.H., T.W. Pry. Effects of Arsenicals on Viral
Infections in Mice. Am. J. Vet. Res. 33:2299-2307, 1972A.
66. Galy, P., R. Touraine, J. Brune, P. Roudier, P. Gallois. Le
Cancer Pulmonaire D'Origine Arsenicale Des Vignerons Du
Beaujolais. J. Fran De Med Et Chir Thorac 17:303-311, 1963.
67. Garb, L.G., C.H. Hine. Arsenical Neuropathy: Residual
Effects Following Acute Industrial Exposure. J. Occ. Med.
19:567-568, 1977.
68. Ginsburg, J.M. Renal Mechanism for Excretion and Transforma-
tion of Arsenic in the Dog. Amer. J. Physiol. 208:832-840,
1965.
"69. Glazener, F.S., J.G. Ellis, P.K. Johnson. Electrocardio-
graphic Findings with Arsenic Poisoning. Cal. Med. 109:158-
162, 1968.
70. Hamamoto, E. Funnyu Ni Yoru Nyuji Hiso Chudokusho (Infant •
Arsenic Poisoning by Powdered Milk. Ninon Iji Shimpo (Japan
Med. J.) 1649:3-13, 1955. Trans, for EPA by Leo Kanner
Assoc. — TR 105-74.
71. Hamilton, A., H.L. Hardy. Industrial Toxicology, 2nd Ed.
New York. Paul B. Hoeber, Inc., 1949.
72. Harkins, W.D., R.E. Swain. Papers on Smelter Smoke (First
Paper). J. Amer. Chem. Soc. 29:970-998, 1907.
73. Harkins, W.D., R.E. Swain. The Chronic Arsenical Poisoning
of Herbivorous Animals. (Papers on Smelter Smoke, Third
Paper.) J. Amer. Chem. Soc. 30:928-946, 1908.
74. Harrisson, J.W.E., E.W. Packman, D.D. Abbott. Acute Oral
Toxicity and Chemical and Physical Properties of Arsenic
Trioxides. Ama. Arch. Ind. Health 17:118-123, 1958.
75. Haywood, J.K. Injury to Vegetation and Animal Life by
Smelter Fumes. J. Amer. Chem. Soc. 29:998-1009, 1907.
76. Heyman, A., J.B. Pfeiffer, R.W. Willett, H.M. Taylor.
Peripheral Neuropathy Caused by Arsenic Intoxication. New
Eng. J. of Med. 254:401-409, 1956.
103
-------
77. Hideo, T., K. Kazuo, S. Tsutomu, S. Hideaki, S. Heiichiro,
S. Chukichi, T. Yoshiro, H. Shigeru, W. Giigichi, H. Kazuo,
O. Tatsuo, S. Chukichi. Mansei Hiso Chudokusho No Rinshoteki
Kansatsu Dai-I-Po (Clinical Observations of Chronic Toxicosis
by Arsenic). Nihon Rinsho Jap. J. of Clin. Med. 18:118-127,
1960. Translated for EPA for Leo Kanner Assoc. - TR106-74.
78. Hill, A.B., E.L. Faning. Studies in the Incidence of Cancer
in a Factory Handling Compounds of Arsenic: Part I, Mortality
Experience in the Factory. Brit J. Indust. Med. 5:1-6,
1948.
79. Hine, C.H., S.S. Pinto, K.W. Nelson.' Medical Problems
Associated With Arsenic Exposure. J. Occ. Med. 19:391-396,
1977.
80. Hodge, H.C., J.W. Embree. Estimation of the Mutagenicity of
Arsenic Compounds Utilizing Dominant Lethal Mutations of
Mice. San Fran. Toxicology Research Laboratory, University
of California, 1977. Contract No. SERA-HI-2, 1977.
81. Holland, R.H., M.S. McCall, H.C. Lanz. A Study of Inhaled
Arsenic74 in Man. Cancer Res. 19:1154-1156, 1959.
82. Holland, R.H., A.R. Acevedo. Carcinogenicity of Inhaled
Arsine and Triphenylarsine in Rabbits, Abstract From Proc.
Am. Assoc. Cancer Res. 4:28, 1964.
83. Hollaender, A. Chemical Mutagens, Vol. IV. Plenum Press
1976.
84. Holmovist, I. Investigations of the Absorption of Some
Metals Among People in the Surrounding Area of a Smelting
Plant, pp. 613-629. In International Symposium Proceedings:
Recent Advances in the Assessment of the Health Effects of
Environmental Pollution, Vol. 2. Paris: Comm. of the Euro-
pean Communities, WHO, EPA, 1974.
85. Hood, R.D., S.L. Bishop. Teratogenic Effects of Sodium
Arsenate in Mice. Arch Environ. Health 24:62-65, 1972.
86. Hood, R.D., S.L. Bishop. Teratogenic Effects of Sodium
Arsenate in Mice. Arch Environ. Health 6:235, 1972. . ,
87. Hood, R.D., G.T. Thacker, B.L. Patterson. Effects in the
Mouse and Rate of Prenatal Exposure to Arsenic, pp. 219-
222. In National Institute of Environmental Health Sciences
(Ed.). Environmental Health Perspectives, Vol. 19. Research
Triangle Park, North Carolina. U.S. Dept. of HEW, Public
Health Service, Nat. Inst. of Health. Pub. No. (NIH) 77-
218/1977.
104
-------
88. Hueper, W.C. Experimental Studies in Metal Carcinogenesis.
V. Tissue Reactions in Rats and Rabbits After Parenteral
Introduction of Arsenic, Beryllium, or Asbestos in Lanolin.
J.N.C.I., 15: 113-124, 1954.
89. Hueper, W.C., W.W. Payne. Experimental Studies in Metal
Carcinogenesis. Chromium, Nickel, Iron, Arsenic. Arch
Environ. Health 5:455-564, 1962.
90. Hunter, F.T., A.F. Kip, J.W. Irvine, Jr. Radioactive
Tracer Studies on Arsenic.Injected as Potassium Arsenite.
J. Pharmacol Exp Ther 76:207-220, 1942.
91. International Agency for Research on Cancer. IARC Monographs
on the Evaluation of Carcinogenic Risk of Chemicals to Man.
Vol. 2. Some Inorganic and Organometallic Compounds. Lyon,
France: WHC, IARC, 1973.
92. Ireland, F.A. Reactions Following the Administration of the
Arsphenamines and Methods of Prevention. Am. J. Syphilis
16:22-35, 1932.
93. Irgolic, K.J. Speciation of Arsenic in Water Supplies. EPA
report, 1978.
94. Irvine, H.G., D'.C. Turnacliff. Study of a Group of Handlers
of Arsenic Trioxide. Arch Derm. Syph. 33:306-315, 1936.
95. Ishinishi, N., Y. Kodama, K. Nobutomo, A. Hisanaga. Pre-
liminary Experimental Study of Arsenic Poisoning in Rat
Lung. pp. 191-196. In National Institute of Environmental
Health Sciences (Ed.). Environmental Health Perspectives,
Vol. 19. Research Triangle Park, North Carolina. U.S.
Dept. of HEW, Public Health Service, Nat. Inst. of Health.
Pub. No. (NIH) 77-218, 1977.
96. Ishinishi, N., Y. Kodama, K. Nobutomo, T. Inamasu, E. Kunitake,
Y. Suenage. Outbreak of Chronic Arsenic Poisoning Among
Retired Workers From an Arsenic Mine in Japan, pp. 121-125.
In National Institute of Environmental Health Sciences
(Ed.). Environmental Health Perspectives, Vol. 19. Research
Triangle Park, North Carolina. U.S. Dept. of HEW, Public
Health Service, Nat. Inst. of Health. Pub. No. (NIH) 77-
218, 1977.
105
-------
97. Japanese Pediatric Society. Nihon Shonika Gakkai Morinaga
Hiso Miruku Chudoku Chosa Shoiinkai (Tokubetsu I Inkai)
Katsudo Hokoku Yoshi. [Summary of Report of Activities of
the Morinaga Arsenic-Tainted Powdered Milk of Poisoning
Investigation Subcommittee (Special Committee) on the
Japanese Pediatric Society.] May 26, 1973. Translated for
EPA by SCITRAN - TR124-74).
98. Jelinek, C.F., P.E. Corneliussen. Levels of Arsenic in the
United States Food Supply, pp. 83-87. In National Insti-
tute of Environmental Health Sciences (Ed.). Environmental
Health Perspectives, Vol. 19. Research Triangle Park, North
Carolina. U.S. Dept. of HEW, Public Health Service, Nat.
Inst. of Health. Pub. No. (NIH) 77-218, 1977.
99. Jung, E.G., B. Trachsel, H. Immich. Arsenic As An Inhibitor
of The Enzymes Concerned in Cellular Recovery (Dark Repair).
Germ. Med. Mon. 12:614-616, 1969.
100. Jung, E.G., B. Trachsel. Molckularbiologische Untersuchungen
Zur Arsencarcinogeneses. Arch Klin Exp Derm 237:819-826,
1970.
101. Kanisawa, M., H.A. Schroeder. Life Term Studies on the
Effect of Trace Elements on Spontaneous Tumors in Mice and
Rats. Cancer Res. 29:892-895, 1969.
102. Kjeldsberg, C.R., H.P. Ward. Leukemia in Arsenic Poisoning.
Ann. of Int. Med. 77:935-937, 1972.
103. Knoth, W. Arsenbehandlung. Arch. Klin. Exp. Dermatol.
227:228-238, 1966/1967.
104. Konetzke, G.W. Die kanzerogene Wirkung von Arsen und Nickel.
Arch. Geschwulstforsch 44: 16-22, 1974.
105. Kroes, R., M.J. VanLogten, J.M. Berkvens, T. DeVries, G.J.
VanEsch. Study on the Carcinogenicity of Lead Arsenate and
Sodium Arsenate and on the Possible Synergistic Effect of
Diethylnitrosamine. Food Cosmet Toxical 12:671-679, 1974.
106. Kuratsune, M., S. Tokudome, T. Sjoralisa., Toshida, U.
Tokumitsu, T. Hayano, M. Seita. Occupational Lung Cancer
Among Copper Smelters. Int. J. CancerJ13:552-558, 1974.
107. Kyle, R.A., G.L. Pease. Hematologic Aspects of Arsenic
Intoxication. New Eng. J. Med. 273:1823, 1965.
106
-------
108. Lakso, J.U., S.A. Peoples. Methylation of Inorganic Arsenic
by Mammals. J. Agric. Food Chem. 23:674-676, 1975.
109. Lancet: 414, February 9, 1901.
110. Lancet. The Beer-Poisoning Epidemic. 801, March 16, 1901.
111. Lancet. Peripheral Neuritis of Arsenical and Alcoholic
Origin. 340-341, February 2, 1901.
112. Lancet. Arsenical Poisoning by Beer in the County Borough
of Salford. 954, March 30, 1901.
113. Lander, H., P.R. Hodge. Arsenic in the Hair and Nails: Its
Significance in Acute Arsenical Poisoning. J. of Forensic
Med. 12:52-67, 1965.
114. Lander, J.J., R.J. Stanley, H.W. Sumner, D.C. Boswell, and
R.D. Aach. Angiosarcoma of the Liver Associated With Fowler's
Solution (Potassium Arsenite). Gastroenterology 68:1582-
1586, 1975.
115. Lanz, H., P.W. Wallace, J.G. Hamilton. The Metabolism of
Arsenic in Laboratory Animals Using AS-74 As a Trace. Univ.
Calif. Pub. Pharmacol 2:263-282, 1950.
116. Laskin, S., M. Kuschner, R.T. Drew. Studies in Pulmonary
Carcinogenesis, pp. 321-350. In: Hanna, M.G., Jr., P.
Nettlesheim, J.R. Gilbert. Eds. Inhalation Carcinogenesis.
AEC Symposium Series, No. 18. Washington, D.C., U.S.
^Atomic Energy Commission, 1970.
117./Lao, R.C., R.S. Thomas, T. Teichman, L. Dubois. Efficiency
' of Collecting Arsenic Trioxide in High Volume Sampling. The
Science of Total Environ. 2:373-379, 1974.
118. Lee, A.M., J.F. Fraumeni. Arsenic and Respiratory Cancer in
Man: An Occupational Study. J. Natl. Cancer Inst. 42:
1045-1052, 1969.
119. Leitch, A., E.L. Kennaway. Experimental Production of
Cancer by Arsenic. Br. Med. J. 2: 1107-1108, 1922.
120. Lowry, O.H., F.T. Hunter, A.F. Kip, J.W. Irvine, Jr.
Radioactive Tracer Studies on Arsenic Injected as Potassium
Arsenite. Vol. II. Chemical Distribution in Tissues. J.
Pharmacol Exp. Ther. 76:221-225, 19.42.
107
-------
121. Lu, F.J., C.K. Yang, K.H. Lin. Physico-chemical Characteris-
tics of Drinking Water in Blackfoot Endemic Areas in Chia-l
and Tainan Hsiens. J. Formosan Medical Assoc. 74: 596605,
1975.
122. Lu, F.J., M.H. Tsai, K.H. Ling. Studies on Fluorescent
Compound in Drinking Water of Blackfoot Disease Endemic
Areas: 1. The Toxic Effects of Fluorescent Compound on the
Chick Embryos. J. Formosan Medical Assoc. 76: 58-63,
1977a.
123. Lu, F.J., M.H. Tsai, K*H. Ling. Studies on Fluorescent
Compounds in Drinking Water of Blackfoot Endemic Areas: 2.
Isolation and Identification of Fluorescent Compounds. J.
Formosan Medical Assoc. 76: 209-217, 1977b.
124. Lugo, G., G. Cassady, P. Palmisano. Acute Maternal Arsenic
Intoxication With Neonatal Death'. Amer. J. Dis. Child 117,
1969.
125. Luh, M.D., R.A. Baker, D.E. Henley. Arsenic Analysis and
Toxicity — A Review. The Science of the Total Environ.
2:1-12, 1973.
126. Lyon, J.L., J.L. Fillmore, M.R. Kluber. Arsenic Air Pollu-
tion and Lung Cancer. Lancet, p. 869, October 22, 1977.
127. MacMahon, B., T.F. Puch. Epidemiology: Principles And
Methods. Boston, Little, Brown & Co., 1970.
I
128. Mchaffey, K.R., B.A. Fowler. Effects of Concurrent Adminis-
tration of Lead Cadmium, and Arsenic in The Rat. pp. 165-
171. In: National Institute of Environmental Health Sciences
(Ed.). Environmental Health Perspectives, Vol. 19. Research
Triangle Park, North Carolina, NC:U.S. Dept. of HEW, Public
Health Service, Nat. Tnst. of Health. Pub. No. (NIH) 77-
218, 1977.
129. Mealey, J., Jr., G.L. Brownell, W.H. Sweet. Radioarsenic in
Plasma, Urine, Normal Tissues, and Intracranial Neoplasms.
Arch. Neurol Psychiatry 81:310-320, 1959.
130. Milby, T.H., C.H. Hine. A Survey of Mortality Due to
Respiratory Diseases Among Employees of the Kennecott Corp.
Unpublished Report Submitted to the Kennecott Corp., New
York, New York, October 4, 1974. (Summarized in NIOSH,
1975.)
108
-------
131. Milham, S., Jr., T. Strong. Human Arsenic Exposure in
Relation to a Copper Smelter. Envir. Res. 7:176182, 1974.
132. Milham, J., Jr. Studies of Morbidity Near a Copper Smelter.
pp. 131-132. In National Institute of Environmental Health
Sciences (Ed.). Environmental Health Perspectives, Vol. 19.
Research Triangle Park, North Carolina, U.S. Dept. of HEW,
Public Health Service, Nat. Inst. of Health. Pub. No. (NIH)
77-218, 1977.
133. Milham, S., Jr. Occupational Cancer Related to Nonferrous
Metal Smelting. Presented at 1977 International Conference
on Cancer and Environment, October 13, 1977.
134. Miller, R.W. Relationship Between Human Teratogens and
Carcinogens. J. Natl. Cancer Inst. 58:471-474, 1977.
135. Milner, J.E. The Effect of Ingested Arsenic on Methylcho-
1anthrene-Induced Skin Tumors in Mice. Arch. Environ.
Health 18:7-11, 1969.
136. Mizuta, N., M. Mizuta, F. Ito, H. Uchida, Y. Watanabe, H.
Akama, T. Murakami, F. Hayashi, K. Nakamura, T. Yamaguchi,
W. Mizuta, S. Oishi, H. Matsumura. An Outbreak of Acute
Arsenic Poisoning Caused by Arsenic Contaminated Soy Sauce
(SHOYU): A Clinical Report of 220 Cases. Bull. Yamaguchi
Med. Sch. 4:131150, 1956.
137. Musil, J., V. Dejmal. Experimental and Clinical Administra-
tion of Radioarsenic. Casopis lekaru ceskych 96:1543-1546,
1957.
138. National Academy of Sciences. Arsenic. Washington, D.C.,
1977.
139. National Institute For Occupational Safety and Health.
Criteria for a Recommended Standard: Occupational Exposure
to Inorganic Arsenic. U.S. Department of Health, Education,
and Welfare: HEW Pub. No. (NIOSH), pp. 75-149, 1975.
140. Natusch, D.F.S., J.R. Wallace, C.A. Evans, Jr. Toxic Trace
Elements: Preferential Concentration in Respirable Particles,
Science 183:202-204, 1974.
141. Nelson, K.W. Industrial Contributions of Arsenic to the
Environment, pp. 31-34. National Institute of Environ-
mental Health Sciences (ed.). Environmental Health Perspec-
tives, Vol. 19. Research Triangle Park, North Carolina:
U.S. Dept. of HEW, Public Health Service, Nat. Inst. of
Health. Pub. No. (NIH) 77-218, 1977.
109
-------
142. Nelson, W.C., M.H. Lykins, J. Mackey, V.A. Newill, J.F.
Pinklea, D.I. Hammer. Mortality Among Orchard Workers
Exposed to Lead Arsenate Spray: A Cohort Study. J. Chron.
Dis. 26:105-118, 1973.
143. Neubauer, 0. Arsenical Cancer: A Review. Brit. J. Cancer
1:192-251, 1947.
144. Newman, J.A., V.A. Archer, G. Saccamanno, M. Kuschner, O.
Auerbach, R.D. Grondauhl, J.C. Wilson. Histologic Types of
Bronchogenic Carcinoma Among Members of Copper-Mining and
Smelting Communities, pp. 260-268. in Saffiotti, U., J.K.
Wagoner (ed.). Occupational Carcinogenesis. Annals of the
N.Y. Academy of Sciences, Vol. 271.
145. Nieberle, K. Uber Endemischen Krebs Im Siebbein Von Scha-
fen. Zeitschrift Fur Krebsforschung 49:137-141, 1939.
146. Nose, Y. Hiso Konnyu Shoyu Chuddku Jiken No Ekigakuteki
Kansatsu. Tuku No Shoyu Chudoku To Hantei Suru Made No
Keika Ni Oite (Epidemiological Observations of the Poisoning
Incident Caused by Soy Sauce with Admixtures of Arsenic).
Koshu Eisei (Public Health) 21:29-43, 1957. (Trans, for EPA
by trans, consul., Arlington, Virginia. 1964.)
147. Occupational Safety and Health Administration. Proposed
Regulation on Inorganic Arsenic. United States Department
of Labor, Occupational Safety and Health Administration
Office of.Planning, Evaluation, and Research, Room 1162,
1826 M Street, N.W., Washington, D.C. 20210, 1975.
148. Oehme, F.W. Mechanisms of Heavy Metal Toxicities. Clin.
Toxico 5:151-167, 1972.
149. Oppenheim, J.J., W.N. Fischbein. Induction of Chromosome
Breaks In Cultured Normal Human Leukocytes by Potassium
Arsenite, Hydroxyurea, and Related Compounds. Cancer Pres.
25:980, 1965.
150. Osburn, H.S. Cancer of the Lung in Gwanda. Central African
J. Med. 3:215-223, 1957.
151. Osburn, H.S. Lung Cancer in a Mining District in Rhodesia.
S. Afr. Med. Journal. 43:1304-1312, 1969.
152. Osswald, H., K. Boerittler. Leukosen Bei Der Maus Nach
Diaplacentarer Und Postnataler Arsenik-Applikation. Dtsch.
Gesamte. Path. 55:289-293, 1971.
110
-------
153. Ott, M.G., B.B. Holder, H.L. Gordon. Respiratory Cancer and
Occupational Exposure to Arsenicals. Arch. Environ. Health
29:250-255, 1974.
154. Overby, L.R., R.L. Frederickson. Metabolic Stability of
Radioactive Arsenilic Acid in Chickens. J. Agric. Food Chem.
11:378-381, 1963.
155. Paton, G.R., A.C. Allison. Chromosome Damage in Human Cell
Cultures Induced by Metal Salts. Mutation Res., 16:332-336,
1972.
156. Patty, F.A. (ed.). Industrial Hygiene and Toxicology. New
York: Interscience Pub., 1963.
157. Pelfrene, A. Arsenic and Cancer: The Still Unanswered
Question. J. Toxicbl and Environ. Health, I: 1003-1016,
1976.
158. Peoples, S.A. Arsenic Toxicity in Cattle. Ann. N.Y. Acad.
Sci. 111:644-649, 1964.
159. Perry, K., R.G. Bowler, H.M. Buckell, H.A. Druett, R.S.F.
Schilling. Studies in the Incidence of Cancer in a Factory
Handling Inorganic Compounds of Arsenic: Part II, Clinical
and Environmental Investigations. Brit. J. of Ind. Med.
5:6-15, 1948.
160. Pershagen, J., C.G. Elinder, A.M. Bolander. Mortality in a
Region Surrounding an Arsenic Emitting Plant. pp. 133-137.
In National Institute of Environmental Health Sciences
(Ed.). Environmental Health Perspectives, Vol. 19. Research
Triangle Park, North Carolina. U.S. Dept. of HEW, Public
Health Service, Nat. Inst. of Health. Pub. No. (NIH) 77-
218, 1977.
161. Peterkova, R., L. Puzanova. An Effect of Tri- and Pentaval-
ent Arsenic on the Early Development of Chicken Embryos.
Folia Morphol 24:5, 1976.
162. Peters, J., K. Schmidt-Ulrich, U. Wolf. Chromosomen Aberra-
tionen An Menchlichen Lymphocyten Bei Chronischen Arzenschaden,
Deut Med. Wochenschr. 95:79, 1970.
163. Peters, R.A., H.M. Sinclair, R.H.S. Thompson. An Analysis
of the Inhibition of Pyruvate Oxidation by Arsenicals In
Relation to the Enzyme Theory of Vesication. Biochem.
40:516-524, 1948.
Ill
-------
164. Peters, R.A., R.W. Wakelin. Observations Upon the Relation
Between Sulfhydryl Groups and Pyruvate Oxidation in Brain
Tissues. Biochem. 40:513-516, 1946.
165. Petres, J., A. Berger. Zum Eingluss Anorganischem Arzens
Auf Die DNA-Symthese Menschlicher Lymphocyten In Vitro.
Arch. Dermat. Porsche 242:343, 1972.
166. Petres, J., D. Baron, M. Hagedorn. Effects of Arsenic Cell
Metabolism and Cell Proliferation: Cytogenetic and Biochem-
ical Studies. pp. 223-227. In National Institute of
Environmental Health Sciences (Ed.). Environmental Health
Perspectives, Vol. 19. Research Triangle Park, North
Carolina: U.S. Department of HEW, Public Health Service,
Nat. inst. of Health, Pub. No. (NIH), 77-218, 1977.
167. Pinto, S., P.E. Enterline, V. Henderson, M.O. Varner.
Mortality Experience in Relation to a Measured Arsenic
Trioxide Exposure, pp. 127-130. In National Institute of
Environmental Health Sciences (Ed.). Environmental Health
Perspectives, Vol. 19. Research Triangle Park, North
Carolina, U.S. Dept. of HEW, Public Health.Service, Nat.
Inst. of Health. Pub. No. (NIH), 77-218, 1977.
168. Pinto, S.S., C.M. McGill. Arsenic Trioxide Exposure in
Industry. Ind. Med. and Surg. 22:281-287, 1953.
169. Pinto, S.S., B.M. Bennett. Effect of Arsenic Trioxide
Exposure on Mortality. Arch, of Envir. Health 7:583-591,
1963.
170. Pinto, S.S., K.W. Nelson. Arsenic Toxicology and Industrial
Exposure.
171. Pinto, S.S., M.O. Varner, K.W. Nelson, A.L. Labbe, L.D.
White. Arsenic Trioxide Absorption and Excretion in Industry.
J. Occ. Med. 18:677-680, 1976.
172. Pinto, S.S. Arsine Poisoning: Evaluation of the Acute
Phase. J. Occ Med., 18: 633-635, 1976.
173. Pinto, S.S./'V. Henderson, P.E. Enterline. Mortality
Experience of Arsenic Exposed Workers. (Unpublished)
174. Pinto, S.S., K.W.. Nelson. Arsenic Toxicology and Industrial
Exposure. Annal. Rev. Pharm. and Toxico 15, 6:95-100, 1976.
175. Regelson, W., U. Kim, J. Ospina, J.F. Holland, Hemangiden-
dothelial Sarcoma of Liver From Chronic Arsenic Intoxication
by Fowler's Solution. Cancer 21:514-522, 1968.
112
-------
176. Rencher, A.C., M.W. Carter, D.W. McKee. A Retrospective
Epidemiological Study of Mortality at a Large Western Copper
Smelter. J. Occp. Med. 19:754-758, 1977.
177. Reynolds, E.S. An Account of the Epidemic Outbreak of
Arsenical Poisoning Occuring in Beer-Drinkers in the North
of England and the Midland Counties in 1900. Lancet 1:166-
170, 1901.
178. Ridgway, L.P., D.A. Karnofsky. The Effects of Metals on the
Chick Embryos: Toxicity and Production of Abnormalities in
Development. Ann. N.Y. Acad. Sci. 55:203, 1952.
179. Robson, A.O., A.M. Jelliffe. Medicinal Arsenic Poisoning
and Lung Cancer. Brit. Med. J. 2:207-209, 1963.
180. Rosen, P. Theoretical Significance of Arsenic as a Carcinogen,
J. Theor. Biol. 32:425-426, 1971.
181. Rossman, T.G., M.A. Meyn, W. Troll. Effects of Arsenite on
DNA Repair in Escherichia Coli. pp. 229-233. In National
Institute of Environmental Health Sciences (Ed.). Environ-
mental Health Perspectives, Vol. 19. Research Triangle
Park, North Carolina, U.S. Dept. of HEW, Public Health
Service, Nat. Inst. of Health. Pub. No. (NIH), 77-218,
1977.
182. Roth, F. The Sequelae of Chronic Arsenic Poisoning in
Moselle Vintners. Germ. Med. Monthly 2:172-175, 1957.
183. Roth, F. Uber Den Bronchialkrebs Arsengeschadigter Winzer.
Virchows Archiv 331:119-137, 1958.
184. Roy, G.L., A.L. Laferriere, J.O. Edwards. A Comparative
Study of Polyol Complexes of Arsenite, Borate, and Tellurate
Ions. J. Inorg. Nucl. Chem. 4:106-114, 1957.
185. Royal Commission on Arsenical Poisoning. Wednesday, March
6, 1901. Lancet: 828-829, March 16, 1901.
186. Royal Commission on Arsenical Poisoning. Lancet: 672-673,
March 2, 1901.
187. Royal Commission on Arsenical Poisoning. Lancet: 980-981,
March 30, 1901.
188. Royal Medical and Chirurgical Society. Epidemic of Arseni-
cal Poisoning in Beer Drinkers in the North of England
During the year 1900. Lancet: 98-100, January 12, 1901.
113
-------
189. Rozenshtein, I.S. Sanitarno-Toksikologicheskaya Otsenka
Nizkikh Kontsentratsii Myshyakovistogo Angigrida U Atmos-
fernom Vozkukhe (Sanitary Toxicological Assessment of Low
Concentrations of Arsenic Trioxide in the Atmosphere.)
Hyg Sanit 35: 1621, 1970.
190. Salaman, M.H., F.J.C. Roe. Further Tests for Tumor-Initiating
Activity: N,Ndi-(2-chloroethyl)-p-aminophenyl butyric acid
(CB 1348) as an Initiator of Skin Tumor Formation on Mouse
Skin. Brit. J. Cancer, 10: 363378, 1956.
191. Schiller, C.M., B.A. Fowler, J.S. Woods. Effects of Arsenic
on Pyruvate Dehydrogenase Activation, pp. 205-207. In
National Institute of Environmental Health Sciences (Ed.). .
Environmental Health Perspectives, Vol. 19. Research
Triangle Park, North Carolina: U.S. Dept. of Health,
Education, and Welfare, Public Health Service National
Institute of Health Pub. No. (NIH) 77-218, 1977.
192. Schrenk, H.H., L. Schreibeis, Jr. Urinary Arsenic Levels as
an Index of Industrial Exposure. Ind. Hyg. J. :225-228,
1958.
193. Schroeder, H.A., M. Mitchener. Toxic Effects of Trace
Elements on the Reproduction of Mice and Rats. Arch.
Environ. Health 23:102-106, 1971.
o
194. Schroeder, H.A., J.J. Balassa. Abnormal Trace Metals in
Man: Arsenic. J. Chron. Dis. 19:85-106, 1966.
195. Schroeder, H.A., M. Kanisawa, D.V. Frost, M. Mitchener.
Germanium, Tin, and Arsenic in Rats: Effects on Growth,
Survival, Pathological Lesions and Life Span. J. Nutrition
96:37-45, 1968.
196. Shapiro, H.A. Arsenic Content of Human Hair and Nails, Its
Interpretation. J. of Forensic Med. 14:65-71, 1967.
197. Shearer, S.D. Personal communication to the Environmental
Protection Agency.
198. Smith, T.J., E.A. Crecelius, J.D. Reading. Airborne Arsenic
Exposure and Excretion of Methylated Arsenic Compounds, pp.
89-93. In National Institute of Environmental Health
• Sciences (ed.). Environmental Health Perspectives. Vol.
19. Research Triangle Park, North Carolina: U.S. Dept. of
Health, Education and Welfare, Public Health Service,
National Institute of Health. Pub. No. (NIH) 77-218, 1977.
114
-------
199. Snegireff, L.S., O.M. Lombard. Arsenic and Cancer. AMA
Arch. Ind. Hyg. 4:199-205, 1951.
200. Sollmann, T. A Manual of Pharmacology and Its Applications
to Therapeutics and Toxicology. Phil: W.B. Saunders Co.,
1964.
201. Sommers, S.C., R.G. McManus. Multiple Arsenical Cancers of
Skin and Internal Organs. Can 6:347-359, 1953.
202. Sram, R.J., V. Bencko. A Contribution to the Evaluation of
the Genetic Risk of Exposure to Arsenic. CS Hyg. 19:308,
1974.
203. Stevens, J.T., L.L. Hall, J.D. Farmer, L.C. Dipasquale, N.
Chernoff, W.F. Durham. Disposition of C-14 and/or As-74-
Cacodylic Acid in Rats After Intravenous, Intratracheal, or
Peroral Administration. pp. 151-157. In National Institute
of Environmental Health Sciences (Ed.). Environmental
Health Perspectives, Vol. 19. Research Triangle Park, NC:
U.S. Dept. of Health, Education and Welfare, Public Health
Service, National Institute of Health, Pub. No. (NIH) 77-218,
1977.
204. Stocken, L.A., R.H.S. Thompson. British Anti-Lewisite: 1.
Arsenic Derivatives of Thiol Proteins. Biochem 40: 529-535,
1946.
205. Strategies and Air Standards Division, Office of Air Quality
Planning and Standards. Air Pollutant Assessment Report on
Arsenic. Office of Air and Waste Management, EPA. Research
Triangle Park, North Carolina 27711; 1976.
206. Sullivan, R.J. Air Pollution Aspects of Arsenic and Its
Compounds. Bethesda, Maryland: Litton Systems, Inc.,
Environmental Systems Division, 7300 Pearl Street, Bethesda,
Maryland 20014, September 1969.
207. Sunderman, F.W., Jr. A Review of the Carcinogenicities of
Nickel, Chromium, and Arsenic Compounds in Man and Animals.
Prev. Med. 5:279-294, 1976.
208. Sunderman, F.W. Jr. Metal Carcinogenesis. In: Advances in
Modern Toxicology, R.A. Cover and M.A. Mehlman, Eds., 1977.
209. Suta, B.E., R. McGaughy/ A. Pines. Population Exposures to
Arsenic. Stanford Research Institute, Menlo Park, California
94025. Environmental Protection Agency Contract No. 68-OL-
2940, Task 30.
115
-------
210. Swain, R.E., W.D. Harkins. Papers on Smelter Smoke (Second
Paper). J. Amer. Chem. Soc. 30:915-928, 1908.
211. Tarrant, R.F., J. Allard. Arsenic Levels in Urine of Forest
Workers Applying Silvicides. Arch. Environ. Health 24:277-
280, 1972.
212. Thiers, H., D. Colomb, G. Moulin, L. Colin. Arsenical
Cutaneous Cancer of the Vinegrowers of Beaujolais. Ann.
Dermatol. Syphilol. 94: 133-158, 1967.
213. Thompson, R.H.S. The Effect of Arsenical Vesicants on the
Respiration of Skin. Biochem. 40:525-529, 1946.
214. Tokudome, S., M. Kuratsune. A Cohort Study of Mortality
from Cancer and Other Causes Among Workers at a Metal
Refinery. Int. J. Cancer 17:310-317, 1976.
215. University of California, Toxicology Research Laboratory.
Estimation of the Mutagenicity of Arsenic Compounds Utilizing
Dominant Lethal Mutations of Mice. Contract No. SERA-HI-2,
1977.
216. Tseng, W. Effects and Dose-Response Relationship of Skin
Cancer and Blackfoot Disease with Arsenic, pp. 109119. In
National Institute of Environmental Health Sciences (Ed.).
Environmental Health Perspectives, Vol. 19. Research
Triangle Park, North Carolina: US Dept. of Health, Education
and Welfare, Public Health Service, National Institute of
Health. Pub. No. (NIH) 77-218, 1977.
217. Tseng, W.P., H.M. Chu, S.W. How, J.M. Fong, C.S. Lin, S.
Yeh. Prevalence of Skin Cancer in an Endemic Area of
Chronic Arsenicism in Taiwan. J. National Cancer Institute
40:453-463, 1968.
218. Tsuchiya, K. Various Effects of Arsenic in Japan Depending
on Type of Exposure. pp. 3542. In National Institute of
Environmental Health Sciences (Ed.). Environmental Health
Perspectives, Vol. 19. Research Triangle Park, North
Carolina: U.S. Dept. of Health, Education, and Welfare,
Public Health Service, Nat. Institute of Health. Pub. No.
(NIH) 77-218, 1977.
219. Tunnicliffe, F.W., O. Rosenheim. Selenium Compounds as
Factors in the Recent Beer Poisoning Epidemic. Lancet: 318,
February 2, 1901.
116
-------
220. Vallee, B.L., D.D. Ulmer, W.E.C. Wacker. Arsenic Toxicology
and Biochemistry. AMA Arch. Ind. Health 21:132-151, 1960.
221. Von Pein, H. Carcinogenesis in Chronic Arsenic Poisoning.
Archie f. Klinische Medizin, 190: 429443, 1943.
222. Wadkins, C.L. Stimulation of Adenosine Triphosphatase
Activity of Mitochondria and Submitochondrial Particles by
Arsenate. .J. of Bio. Chem. 235: 3300-3303, 1960.
223. Wagner, S.L., P. Weswig. Arsenic in Blood and Urine of
Forest Workers. Arch. Environ. Health 28: 77-79, 1974.
224. Wagner, W.L. Environmental Conditions in U.S. Copper
Smelters. U.S. Dept. of Health, Education and Welfare;
Public Health Service, Center for Disease Control; National
Institute for Occupational Safety and Health; Division of
Technical Services, Salt Lake City, Utah, 1975.
225. Watrous, R.M., M.B. McCaughey. Occupational Exposure to
Arsenic. Indust Med. 14: 639-645, 1945.
226. Webb, J.L. Enzyme and Metabolic Inhibitors, Vol. 2. Malonate,
Analogs, Dehydroacetate, Sulfhydryl Reagent, O-Iodosobenzoate,
Mercurials. New York: Academic Press, 1966.
227. Weissburger, J.H., G. Williams. Metabolism of Chemical
Carcinogens. In Becker, F.F. (Ed.). Cancer. Plenum Press,
1975.
228. Whanger, P.O., P.H. Weswig, J.C. Stoner. Arsenic Levels in
Oregon Waters. pp. 139-143. In National Institute of
Environmental Health Sciences (Ed.). Environmental Health
Perspectives, Vol. 19. Research Triangle Park, North
Carolina: U.S. Dept. of Health, Education, and Welfare,
Public Health Service, National Institute of Health. Pub.
No. (NIH) 77-218, 1977.
229. Willcox, W.H. Toxicological Detection of Arsenic and the
Influence of Selenium on its Tests. Lancet: 778, March 16,
1901.
230. Winkler, W.O. Identification and Estimation of the Arsenic
Residue in Livers of Rats Ingesting Arsenicals. J. Assoc.
of Anal Chem. 45: 80-91, 1962.
231. Yeh, S. Relative Incidence of Skin Cancer in Chinese in
Taiwan: With Special Reference to Arsenical Cancer. Natl.
Cancer Inst. Monogr. 10: 81-107, 1963.
117
-------
232. Yeh, S. Skin Cancer in Chronic Arsenicism. Human Path. 4:
469-485, 1973.
233. Yeh, S., S.W. How, C.S. Lin. Arsenical Cancer of Skin:
Histologic Study With Special Reference to Bowen's Disease.
Cancer 21: 312-339, 1968.
234. Yoshikawa, T., J. Utsumi, T. Okada, M. Moriuchi, K. Ozawa,
T. Kaneko. Concerning the Mass Outbreak of Chronic Arsenic
Toxicosis in Niigata Prefecture. Chiryo 43: 1739-1749,
1960. Translated from the Japanese by Leo Kanner Associates,
October 1973 for the Environmental Protection Agency.
235. Zachariae, H. Arsenik og cancerrisiko. Ugeskr. Laeg. 134:
27202721, 1972.
236. Zettel, H. Der Einfluss Chronischer Arsenschadigung Auf
Herz and Gefasse. Z.F. Klin Medizin 142: 687-703, 1943.
118
------- |