United States
Environmental Protection
Agency
Environmental Criteria and
Assessment Office
Research Triangle Park NC 27711
EPA-600/8-83-014F
August 1984
Final Report
Research and Development
vvEPA
Health Assessment
Document for
Chromium
Final
Report
-------
EPA-600/8-83-014F
August 1984
* i
Health Assessment Document
for Chromium
Final Report
Environmental Criteria and Assessment Office
Office of Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
p-S. Environ-ronta! P
~ . • -i -v- -1 T , . ". •-••--,.
.,, ,,".'-'' ' "'•'" I 'J*'u I ' .
' ~ "-tit, ho on 16'fJ
60604
-------
DISCLAIMER
*
This document has been reviewed in accordance with United States
Environmental Protection Agency policy and approved for publication. Mention of
trade names or commercial products does not constitute endorsement or
recommendation for use.
ii
-------
PREFACE
*
The Office of Health and Environmental Assessment has prepared this health
assessment to serve as a "source document" for EPA use. The health assessment
document was originally developed for use by the Office of Air Quality Planning
and Standards to support decision-making regarding possible regulation of
chromium as a hazardous air pollutant. However, the scope of this document has
since been expanded to address multimedia aspects.
In the development of the assessment document, the scientific literature
has been inventoried, key studies have been evaluated and summary/conclusions
have been prepared so that the chemical's toxicity and related characteristics
are qualitatively identified. Observed effect levels and other measures of
dose-response relationships are discussed, where appropriate, so that the nature
of the adverse health responses are placed in perspective with observed
environmental levels.
iii
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
The Office of Health and Environmental Assessment (OHEA) of U.S. EPA is
responsible for the preparation of this health assessment document.
Dr. Si Duk Lee, as the Project Coordinator and editor, had overall
responsibility for coordination and direction of the document preparation and
production effort. The chapters addressing physical and chemical properties,
sampling and analysis, air quality and toxicity data were prepared by Life and
Environmental Sciences, Syracuse Research Corporation, Syracuse, NY. The
principal authors of these chapters are listed below.
Dr. Dipak Basu
Karen Blackburn
Dr. Bruce Harris
Dr. Michael W. Neal
Frederick W. Stoss
The OHEA Carcinogen Assessment Group (CAG) was responsible for preparation
of the sections on carcinogenicity. Participating members of the CAG are listed
below (principal authors of present carcinogenicity materials are designated
by *).
Roy E. Albert, M.D. (Chairman)
Elizabeth L. Anderson, Ph.D.
Larry D. Anderson, Ph.D.
Steven Bayard, Ph.D.
David L. Bayliss, M.S.
*Chao W. Chen, Ph.D.
Margaret M.L. Chu, Ph.D.
•Herman J. Gibb, B.S., M.P.H.
Bernard H. Haberman, D.V.M., M.S.
•Charalingayya B. Hiremath, Ph.D.
Robert E. McGaughy, Ph.D.
Dharm V. Singh, D.V.M., Ph.D.
Todd W. Thorslund, Sc.D.
iv
-------
The following individuals provided peer review of this document or earlier
drafts of this document:
U.S. Environmental Protection Agency
Joseph Padgett
Office of Air Quality Planning and Standards
U.S. EPA
Karen Blanchard
Office of Air Quality Planning and Standards
U.S. EPA
Mike Waters, Ph.D.
Office of Health Research
Health Effects Research Laboratory
U.S. EPA
Jerry F. Stara, D.V.M.
Office of Health and Environmental Assessment
Environmental Criteria and Assessment Office
U.S. EPA
Debdas Mukerjee, Ph.D.
Office of Health and Environmental Assessment
Environmental Criteria and Assessment Office
U.S. EPA
Lester D. Grant, Ph.D.
Office of Health and Environmental Assessment
Environmental Criteria and Assessment Office
U.S. EPA
K.S. Lavappa, Ph.D.
Office of Health and Environmental Assessment
Reproductive Effects Assessment Group
U.S. EPA
External Peer Reviewers
Dr. Ann Baetjer
John Hopkins School of Hygiene
615 North Wolfe Street
Baltimore, MD 21218
Dr. Derek Hodgson
Professor, Chemistry Department
University of North Carolina
Chapel Hill, NC 27514
-------
Dr. Marshall Johnson
Thomas Jefferson Medical College
Anatomy Department
1020 Locust Street
Philadelphia, PA 19107
Dr. Magnus Piscator
University of Pittsburgh
Graduate School of Public Health
Environmental Epidemiology
Pittsburgh, PA 15261
Dr. Bruce Stuart
Stauffer Chemical Company
Farmington Avenue
Farmington, CT 06032
Dr. Robert Tardiff
1423 Trapline Court
Vienna, VA 22180
Dr. Norman M. Trieff
University of Texas Medical Branch
Department of Pathology, UTMB
Galveston, TX 77550
Dr. Jim Withey
Health Protection Branch
Dept. National Health and Welfare
Tunney's Pasture
Ottawa, Ontario
CANADA K1A 012
EPA Science Advisory Board
The substance of this document was independently peer reviewed in public
sessions of the Environmental Health Committee of EPA's Science Advisory Board.
vi
-------
TABLE OF CONTENTS
Page
DISCLAIMER ii
PREFACE iii
AUTHORS, CONTRIBUTORS, AND REVIEWERS iv
TABLE OF CONTENTS vii
LIST OF TABLES xi
LIST OF FIGURES xiv
1. INTRODUCTION 1-1
2. SUMMARY AND CONCLUSIONS 2-1
2.1 INTRODUCTION 2-1
2.2 FORMS, SOURCES AND CONCENTRATIONS OF CHROMIUM 2-2
2.3 MEASUREMENT METHODS 2-1
2.14 PHARMACOKINETICS AND ESSENTIALITY 2-5
2.U.1 Absorption, Transport and Excretion 2-5
2.1.2 Essentiality of Chromium 2-7
2.5 EFFECTS OF CHROMIUM ON BIOLOGICAL SYSTEMS AND HEALTH 2-7
2.5.1 Toxic Effects in Man and Animals 2-7
2.5.2 Genotoxicity, Carcinogenicity and Assessment of Risk 2-9
3. BACKGROUND INFORMATION 3-1
3.1 CHEMICAL AND PHYSICAL PROPERTIES 3-1
3.2 PRODUCTION, USE, AND RELEASES TO THE ENVIRONMENT 3-6
3.2.1 Production of Chromium Compounds 3-6
3.2.2 Uses of Chromium and Its Compounds 3-8
3.2.3 Releases to the Environment 3-13
3.3 ENVIRONMENTAL FATE AND TRANSPORT 3-17
3.3.1 Air 3-17
3.3.2 Water and Sediments 3-18
3.3.3 Soil 3-20
vii
-------
TABLE OF CONTENTS (cont.)
3.4 LEVELS OF CHROMIUM IN VARIOUS MEDIA 3-20
3.4.1 Ambient Air 3-20
3.4.2 Aquatic Media 3-21
3.4.3 Aquatic Suspended Materials and Suspended 3-28
3.4.4 Soil 3-30
3.4.5 Food 3-30
3.4.6 Cigarettes 3-35
3.5 INDICES OF EXPOSURE 3-36
3.5.1 Chromium in Blood 3-36
3.5.2 Chromium in Urine 3-38
3.5.3 Chromium in Human Hair 3-40
3.6 SUMMARY 3-41
4. SAMPLING AND ANALYSIS 4-1
4.1 SAMPLING AND STORAGE 4-1
4.1.1 Air 4-1
4.1.2 Water 4-3
4.1.3 Soil and Sediments 4-4
4.1.4 Food 4-5
4.1.5 Biological Samples 4-5
4.2 SAMPLE PRETREATMENT 4-5
4.2.1 Wet and Dry Ashing 4-6
4.2.2 Precipitation 4-6
4.2.3 Solvent Extraction 4-8
4.2.4 Chromatographic Method 4-8
4.3 METHODS OF ANALYSIS 4-9
4.3.1 Atomic Absorption Spectrometry (flame) 4-13
4.3.2 Atomic Absorption Spectrometry (flameless) 4-14
4.3*3 Emission Spectroscopy 4-16
4.3.4 Neutron Activation Analysis 4-16
4.3.5 X-ray Fluorescence 4-17
4.3.6 Colorimetric 4-18
4.3.7 Gas Chromatography 4-19
4.3.8 Chemiluminescence 4-19
4.3.9 Polarography 4-20
viii
-------
TABLE OF CONTENTS (cont.)
4.3.10 Mass Spectrometry 4-20
4.3.11 Catalytic Method l»-21
4.3.12 Liquid Chromatography 4-21
4.3.13. Particle Induced X-ray Emission 4-21
4.4 CONSIDERATIONS IN ANALYSIS 4-22
5. CHROMIUM METABOLISM IN MAN AND ANIMALS 5-1
5.1 ROUTES OF CHROMIUM ABSORPTION 5-1
5.1.1 Chromium Absorption and Deposition by Inhalation 5-1
5.1.2 Gastrointestinal Absorption of Chromium 5-4
5.1.3 Chromium Absorption Through the Skin 5-6
5.2 CHROMIUM TRANSPORT, METABOLISM, DISTRIBUTION,
AND ELIMINATION 5-8
5.2.1 Transport and Metabolism ' 5-8
5.2.2 Distribution 5-12
5.2.3 Elimination 5-18
5.3 SUMMARY 5-21
6. CHROMIUM AS AN ESSENTIAL ELEMENT 6-1
6.1 CHROMIUM DEFICIENCY 6-1
6.2 GLUCOSE TOLERANCE FACTOR 6-4
6.3 SUMMARY 6-6
7. CHROMIUM TOXICOLOGY 7-1
7.1 ACUTE EFFECTS OF CHROMIUM EXPOSURE IN MAN AND ANIMALS 7-1
7.1.1 Human Studies 7-1
7.1.2 Animal Studies 7-1
7.1.3. Chromium Hypersensitivity 7-4
7.2 EVALUATION OF THE CARCINOGENICITY OF CHROMIUM 7-13
7.2.1. Animal Studies 7-13
7.2.2. Epidemiologic Studies 7.146
7.2.3. Quantitative Estimation 7-79
7.2.4. Summary 7-102
7.2.5. Conclusions 7-106
ix
-------
TABLE OF CONTENTS (cont.)
Page
7.3 GENOTOXICITY 7_108
7.3.1 In Vitro Mutagenicity 7-108
7.3.2 Effects on DNA and DNA Replication 7-117
7.3.3 Chromium Induced Chromosomal Aberrations and
Cell Transformation 7-121
7.3.1 Summary 7-128
7.4 DEVELOPMENTAL TOXICITY AND OTHER REPRODUCTIVE EFFECTS 7-129
7.1.1 Development Toxicity 7-129
7.1.2 Other Reproductive Effects 7-136
7.1.3 Summary 7-137
#
7.5 OTHER TOXIC EFFECTS OF CHROMIUM 7-137
7.5.1 Respiratory Effects 7-1U1
7.5.2 Renal Effects of Chromium 7-155
7.5.3 Miscellaneous Toxic Effects 7-156
7.6 SUMMARY OF TOXIC EFFECTS OTHER THAN CANCER FOLLOWING
EXPOSURE TO CHROMIUM COMPOUNDS 7-157
8. CURRENT REGULATIONS AND STANDARDS 8-1
8.1 OCCUPATIONAL EXPOSURE 8-1
8.2 EXPOSURE TO CHROMIUM IN AMBIENT WATER 8-1
8.3 EXPOSURE TO CHROMIUM IN AMBIENT AIR 8-5
9. REFERENCES 9>1
-------
LIST OF TABLES
Table Page
3-1 Physical Properties of Selected Trivalent Chromium Compounds 3-3
3-2 Physical Properties of Selected Hexavalent Chromium Compounds .... 3-4
3-3 Manufacturers and Their Production Capacities of Sodium
Chromate and Sodium Dichromate 3-9
3-1 Principal United States Manufacturers of Chromic Acid 3-10
3-5 United States Chromium Consumption Pattern in 1979 3-11
3-6 Sources and Estimates of United States Atmospheric Chromium
Emissions in 1970 3-15
3-7 Regional Distribution of Principal Chromium Emissions 3-16
3-8 Five Forms of Chromium Transported in the Yukon and
Amazon Rivers 3-19
3-9 Total Chromium Concentrations Measured in the Ambient Air of
Selected Sites in the United States During 1977-1980 3-22
3-10 Chromium Levels in a Few Surface Waters and Groundwaters 3-26
3-11 Chromium Concentrations in U.S. Drinking Waters 3-27
3-12 Concentration of Chromium in Sediments 3-29
3-13 Chromium Content in Selected United States' Soils 3-31
3-14 Chromium Content in Various U.S. Foods 3-32
3-15 Concentration of Chromium in a Few Commerical Grade Acidic
Foods 3-31
1-1 Composition and Efficiency of AA Extractant Solutions 4-7
1-2 Analytical Methods for the Determination of Chromium 4-10
6-1 Estimated Adequate and Safe Intake (EASI) for Chromium 6-2
7-1 Inhalation Exposure of Mice to Chromium-Containing Dust 7-14
7-2 Dosage Regimen for Intratracheal Instillation of Sodium
Dichromate and Calcium Chromate 7-19
xi
-------
LIST OF TABLES
Table
7-3 Lung Tumor Incidence in Sprague-Dawley Rats Following
Intratracheal Instillation of Sodium Dichromate or
Calcium Chromate 7-20
7-1 Combined Lung Tumor Incidence in Sprague-Dawley Rats
Following Intratrachael Instillation of Sodium Dichromate and
Calcium Dichromate 7-21
7-5 Carcinomas Produced with Chromium Compounds in Rats 7-22
7-6 Living Tumors Found and Microscopically Confirmed 7-24
7-7 Incidence of Lung Tumors in Rats Following Intrabronchial
Implantation of Various Chromium Compounds 7-27
7-8 Exposure Schedule for Bioassay of Chromium Compounds By
Intrapleural Injection 7-29
7-9 Compounds Reported to Have Been Tested for Carcinogenicity by
Intrapleural Implantation 7-31
7-10 Experimental Conditions Used to Study the Effect of Intrafemoral,
Intraperitoneal, and Intravenous Administration of Chromium 7-34
7-11 Levels of Hexavalent Chromium in Fractionated Residue Dust 7-36
7-12 Compounds Reported to Have Been Tested for Carcinogenicity by
Intramuscular Implantation 7-37
7-13 Carcinogenicity of Chromium Compounds in Experimental Animals 7-11
7-14 Location of Chromate Manufacturing Plants Which Participated
in Epidemiologic Studies and Plants from Which Vital Statistics
Were Obtained for Each Study 7-48
7-15 Observed Number of Deaths, Standardized Mortality Ratios (SMRs),
and 95$ Confidence Limits (95$ CL) for Deaths Due to Cancer of
the Trachea, Bronchus, and Lung and the Number of Reported Deaths
for Which No Certificate Could Be Obtained, by Year of Initial
Employment, Exposure Category, and Total Duration Employed, for
Workers Initially Hired As Hourly Employees 7-58
7-16 Lung Cancer in Workers in the Chromate Pigment Industry 7-69
7-17 Mortality Ratios Resulting from Malignant Tumors Among Workers
in Chromium Ferroalloy Production 7-74
xii
-------
LIST OF TABLES (cont.)
Table
7-18 Age-Specific Lung Cancer Deaths and GradienttExposures to Total
Chromium 7-85
7-19 Combined Age-Specific Lung Cancer Death Rates and Total Chromium
Exposure (in pg/nr) 7.86
7-20 Comparison of Unit Risks (Lifetime Risk Due to 1 pg/m^ of
Hexavalent Chromium in Air) 7-97
7-21 Relative Carcinogenic Potencies Among 51 Chemicals Evaluated
by the Carcinogen Assessment Group As Suspect Human Carcinogens.. 7-99
7-22 The In Vitro Mutagenicity Bioassay of Chromic Compounds 7-109
7-23 Chromium Produced Clastogenic Effects and Cell Transformation ... 7-122
7-24 Teratogenic and Fetotoxic Effects of Chromium 7-133
7-25 Studies Suggesting NOAELS or NOELS 7-139
7-26 Clinical Findings in Workers Employed in Chromium-Plating
Plants 7-143
7-27 Perforation of Nasal Septum in Chromate Workers 7-146
7-28 Perforation of Nasal Septum in Chromate Workers 7-147
7-29 Nasal Medical Findings in a Chromium-Plating Plant 7-148
7-30 Medical Complaints of Workers in "hard" Chromium Electroplating
Plant 7-154
8-1 Recommended Occupational Standards and Recommended Criteria for
Chromium Compounds in the United States 8-2
8-2 Recommended Standards for Chromium in Ambient Waters in the
United States 8-3
8-3 Ambient Water Quality Criteria for the Protection of
Human Health 8-4
8-4 Calculated Ambient Water Quality Criteria for the Protection
of Aquatic Life 8-6
xiii
-------
LIST OF FIGURES
Figure
3-1
5-1
5-2
Simplified flow chart for the production of metallic chromium
and its compounds from chromite
Rate of blood clearance of intravenously injected ^1Cr(III)
from male rats
Whole-body elimination of intravenously administered ^Cr(III)
Page
3-7
5-13
5-19
7-1 Histogram representing the frequency distribution of the potency
indices of 54 suspect carcinogens evaluated by the Carcinogen
Assessment Group 7-98
xiv
-------
1. INTRODUCTION
The 1970 Clean Air Act and its 1977 amendments mandate EPA to regulate,
under Section 112, those pollutants that "may reasonably be anticipated to result
in an increase in mortality or an increase in serious irreversible, or
incapacitating reversible, illness." It also states that EPA must regulate,
under Section 111 (d), those pollutants that "may reasonably be anticipated to
endanger public health or welfare."
For this reason, the Office of Air Quality Planning and Standards has
requested that the Office of Health and Environmental Assessment prepare a scien-
tific assessment for chromium so that it can be determined whether the regulation
would be warranted under these sections of the Clean Air Act, since human expo-
sure to chromium has been a matter of public health concern. Therefore, this
chromium document will serve as a scientific data base for regulatory decision
making by the agency. The health assessment document should represent an inter-
pretive summary of relevant studies rather than a compendium of all available
papers.
The present document represents a comprehensive data base that considers
all sources of chromium in the environment, the likelihood for its exposure to
humans, and the possible consequences to man and lower organisms from its absorp-
tion. This information is integrated into a format that can serve as the basis
for qualitative and quantitative risk assessments, while at the same time identi-
fying gaps in our knowledge that limit accurate health assessment at this time.
Thus, it is expected that this document may serve the information needs of other
government agencies and the private sector that may be involved in decision
making and regulatory activities.
1-1
-------
2. SUMMARY AND CONCLUSIONS
2.1. INTRODUCTION
Trivalent Chromium (Cr III) is considered an essential micro-nutrient at
relatively low levels, largely because chromium deficiency results in a buildup
of glucose in the blood. At much higher levels, certain hexavalent chromium
(CrVI) compounds are known to be carcinogens. Thus, chromium is unique among the
metallic elements, given its paradoxical roles in both nutrition and
carcinogenesis. The seemingly contradictory information on the effects of
chromium is being clarified through increasing understanding of the role of the
differing oxidation states and types of chromium compounds, which apparently
determine the relative risks and benefits to human health of chromium in its
various forms.
In the ambient environment, however, most of the monitoring information has
provided only total elemental chromium levels. Outside of occupational
settings, only limited information exists on the types of chromium compounds to
which the public is exposed, although the trivalent form is known to be
predominant. The assessment which follows focuses on several key areas which
bear on the kind and extent of effects associated with chromium compounds:
sources and concentrations of important chromium compounds (particularly Cr(III)
and Cr(VI)); measurement methods; pharmacokinetics and essentiality; toxic
effects in man and animals; and carcinogenic risks.
2.2. FORMS, SOURCES AND CONCENTRATIONS OF CHROMIUM
Chromium is a metallic element which occurs in nature primarily as the
mineral chrotnite; elemental chromium does not occur naturally. Chromium exists
2-1
-------
in four oxidation states, but only two of them, Cr(III) and Cr(VI), appear to be
important, owing to their predominance and stability in the ambient environment.
All forms are influenced greatly by pH, which affects the solubility and
subsequent reactivity of chromium ions. Trivalent chromium is chemically basic,
while hexavalent chromium is acidic.
Trivalent chromium is the most stable oxidation state, and the most
important chemically. Its foremost characteristics are its ubiquitousness in
the environment as part of the earth's crust, and its tendency to form
kinetically inert hexacoordinate complexes. It reacts with aqueous hydroxides
to form insoluble chromium hydroxide. Hexavalent chromium is the second most
stable state, but the most important toxicologically. It occurs rarely in
nature, apart from man's intervention, because it is readily reduced to Cr(III)
in the presence of organic matter. It is quite soluble, existing in solution as
a complex anion. However, in certain soils and natural waters, it can remain
unchanged for protracted periods of time.
Chromite ore is not mined in the United States, but Cr(VI) chemicals are
produced from imported ores, amounting to 21% of total U.S. chromium consumption.
Metallurgical processes constitute approximately 6058, and refractory uses about
18J. Little direct information exists on the speciation of chromium compounds in
the environment, because of the limitations of existing measurement methods (as
described below). Accordingly, knowledge of chromium chemistry and its sources
must be relied on in estimating the relative ambient contribution of different
species. Direct sources include chemical and refractory plants; indirect
sourc'es include fossil fuel combustion, waste incineration and cement plant
emissions.
Some source categories are likely to emit both trivalent and hexavalent
forms of chromium. These are steel, refractory, chemicals manufacturing, as well
2-2
-------
as sewage sludge and municipal incineration. Cooling towers and chrome plating
facilities emit hexavalent chromium, and chromium ore refining, ferro-chromium
production, cement production, and coal and oil combustion are likely to be
sources of trivalent chromium. Maximum annual average ambient (total) chromium
levels within 20 kilometers of these sources range from approximately 0.01-13.50
Background ambient air concentrations of total chromium have ranged from as
O •}
low as 0.005 ng/nr (at the South Pole) to 1 . 1 ng/nr in other remote areas of the
world. In the United States, recent monitoring of the ambient air in many urban
and non-urban areas has shown total chromium concentrations averaging in the
range of approximately 0.005-0.157 ng/nr. The maximum 24-hour average
concentration found for any one site was 0.684 jig/m in the Baltimore, MD area.
Because Cr(III) is highly stable and Cr(VI) reacts over time to form Cr(III), it
is assumed that most chromium in ambient air occurs in the trivalent state.
Monitoring of both the species and oxidation states of chromium in the ambient
air should be a priority for future research.
The chromium concentration in U.S. waters varies with the type of
surrounding industrial sources and the type of underlying soils. An analysis of
approximately 1,000 tap water samples in representative U.S. cities showed
chromium concentrations ranging from 0.4 to 8 ppb. Chromium levels in soil vary
with soil origin and degree of contamination from anthropogenic sources. Tests
on domestic soil have shown chromium concentrations ranging from an average of
14-70 ppm. Because the amount of chromium in food and food plants is relatively
low, 'and because chromium does not appear to accumulate in mammalian systems,
bioaccumulation in the soil-plant-animal system does not appear to be a
significant exposure source.
2-3
-------
2.3. MEASUREMENT METHODS
One of the main problems in assessing the effects of chromium on human
health is the lack of adequate methods to measure the types and amounts of
chromium compounds. Prior to 1978, urinary chromium levels fell within the range
of 2 to 20 fig/fc. In 1971, radio-tracer experiments indicated that approximately
0.5-1$ of the chromium was absorbed throught the digestive system. Accordingly,
chromium excretion of 10 pg/day would correlate with a chromium intake of 1-2
mg/day. However, few diets contain more than 100 (ig/day chromium; this anomaly
was resolved by showing that the background collection capabilities of the
analytical methods used to measure chromium (atomic absorption) were inadequate
for chromium determinations.
Several methods are available for measuring elemental chromium in both
environmental and biological samples. These include atomic absorption
spectroscopy, instrumental neutron activation analysis, X-ray fluoroescence, and
particle-induced X-ray emissions (PIXE). While these methods are sensitive to
the ppb level, problems in sample collection, preparation and interferences are
shared by all. In biological samples, neutron activation analysis data tend to
be lower than atomic absorption and X-ray fluorescence data. In environmental
samples, neutron activation analysis data are higher. Generally, a comparison of
the results indicates that modified atomic absorption spectroscopy provides
relatively reliable analyses. Another problem in chromium determination is the
lack of adequate reference materials. Ideally, reference materials should match
the samples to be analyzed with respect to chromium levels and each reference
composition. Because the materials are not yet standardized, inter-laboratory
comparisons are difficult.
Techniques for monitoring hexavalent chromium are also subject to
considerable error. For example, although the OSHA colorimetric method is the
-------
most commonly used analytical tool, particularly in occupational settings, low
sample recoveries have been reported in chromium levels of less than 10 |ig.
2.U. PHARMACOKINETICS AND ESSENTIALITY
2.U.1. Absorption, Transport and Excretion. An understanding of the role of
chromium as an essential nutrient and causative agent in toxicity and
carcinogenicity requires knowledge of the rates of absorption, mechanisms of
absorption, transport and organ distribution of the various chromium-containing
compounds. There are three primary routes of entry for chromium into the human
body. For most people, the gastro-intestinal (GI) tract is the primary route of
uptake, while in occupational exposures the airways and skin are the most
important routes of uptake. Rates of uptake in the GI tract depend on a number of
different factors, such as the valence state of chromium in the compound, the
water solubility of the compound and the passage of time through the tract.
Uptake in the airways is also influenced by the particle size distribution of the
inhaled aerosols, and on factors which govern the clearance time of the lung.
Limited work on humans has been carried out on the relationship between
exposure to trivalent chromium compounds and lung uptake and urinary excretion of
chromium. In one study on workers exposed to chromium lignosulfonate, it was
demonstrated that while chromium in the chromium lignin was present in the
trivalent state, it acted pharmacokinetically like water soluble Cr(VI)
compounds. An average of 14 pg/Jl of urine was excreted, at an atmospheric
chromium lignin concentration of 50 (jg chromium/m . One to two percent of the
inhaled chromium was excreted in the urine.
For Cr(VI), the urinary chromium concentrations corresponding to an
airborne concentration of 50 iig/nr Cr(VI) were UQ pg/fc in one study, and 10 to 20
2-5
-------
in another. It was noted that chromium-bearing particles stay longer in the
airways in smokers than in non-smokers.
The established normal levels of chromium in whole blood and in serum have
declined with time, reflecting the changes and improvements in analytical
methods. In the airways and in the GI tract, soluble Cr(VI) compounds are
apparently taken up by epithelial cells by simple diffusion through the plasma
membrane. After entry, Cr(VI) reduction occurs from the action of enzymatically
mobilized electrons, which are available from GSH, NADPH, and NADH. The reducing
capacity inside the cell is limited, so that Cr(VI) and Cr(III) exist
simultaneously inside the cytoplasm; Cr(VI) is then released from the cell by
simple diffusion into the blood stream and taken up into blood cells. In spite
of the refined methods of analysis available, a reliable range of normal blood
chromium concentrations cannot be given with confidence. When using modern
methods for analysis, the whole blood concentration may be suggested to be within
the range of 0.5 to 3 ppb, while the serum level is probably below 0.2 ppb.
The chromium concentration in human tissues has been shown to decrease with
increasing age. In contrast to this, chromium concentrations in the lung have
been shown to increase with age. This increase in chromium content in the lungs
may be due to deposition and retention of insoluble chromium-containing
particles from the inhaled environmental air, as well as from tobacco smoke.
2.4.2. Essentiality of Chromium. Animal studies have demonstrated that
chromium-deficient rodents gain less weight and have a shorter life-span than
animals maintained on a diet containing adequate chromium values. Chromium
deficiency results in glucose intolerance in rats. This intolerance can be
reduced with dietary treatment with Cr(III). In humans, symptoms of chromium
deficiency consist of glucose intolerance, weight loss and confusion. Those
2-6
-------
prone to chromium deficiency include the elderly, diabetics, pregnant women,
malnourished children, offspring or siblings of diabetics and persons with early
coronary heart disease. Although the exact level of chromium needed for good
health is not known, the average American intake of 50 to 200 jig/day is
considered adequate because at such levels symptoms associated with chromium
deficiency are not observed. It should be noted that there is a considerable
difference between the low levels of intake that are associated with nutritional
deficiency and the high levels of exposure which are associated with toxic
effects.
2.5. EFFECTS OF CHROMIUM ON BIOLOGICAL SYSTEMS AND HEALTH
2.5.1. Toxic Effects in Man and Animals. The effects of both Cr(III) and
Cr(VI) have been studied in man and animals. Both long-term and short-term
exposure conditions have been investigated, but most of the long-term exposures
have focused on carcinogenic effects (discussed in Section 2.5.2. below).
The relative chemical inactivity of Cr(III) compared with Cr(VI) correlates
with various acute toxicity studies on chromium salts. Oral LD_rt (dose lethal to
30
50$ of recipients) levels in rats have been reported to range from 135 mg/kg to
11,260 mg/kg for Cr(III). As seen in the previous section on pharmacokinetics,
the relatively high amounts of'Cr(III) which are required to cause death arise
from the relative insolubility and poor intestinal absorption of most Cr(III)
compounds. Unlike the trivalent compounds, those of Cr(VT) tend to cross
biological membranes fairly easily, and are somewhat more readily absorbed
through the gut or through the skin. The strong oxidizing powers of Cr(VI)
compounds explain much of their irritating and toxic properties.
Exposure to Cr(VI) has been associated primarily with renal damage. For
humans no quantitative evidence of acute toxcity through oral ingestion has been
2-7
-------
reported. In various animal species, single injections of 2 mg/kg caused
cellular and structural damage in the kidneys.
The effects of chromium on the skin were recognized over 150 years ago.
Many chromium compounds can damage the skin, but metallic chromium or chromium
alloys are chemically inert and are not harmful. The effects of chromium
compounds on the skin are caused primarily by direct contact. Most of the
effects have occurred in occupational settings, and, as expected, with more men
than women reporting effects. Cr(VI) derivatives can cause ulcers of the hands
and accompanying perforations of the nasal septum. Allergic contact dermatitis
may arise from exposure to either trivalent or hexavalent chromium, although
hexavalent chromium is responsible for most of the reported cases. Cr(VT)
penetrates undamaged skin, and subsequently reduces to Cr(III) which combines
with proteins or other skin components to form a whole skin allergin.
Effects on the upper respiratory tract have been observed in workers in
chromium-related industries. The major effects of chromium on this system
include ulceration of the nasal septum, with subsequent perforation, and chronic
rhinitis and pharangitis. Early studies indicated that approximately one-half
to four-fifths of the workers in chromate plants had perforated nasal septa, at
levels of exposure that approached 1 mg/m . Subsequent work indicated that
•2
chromic acid levels exceeding 0.1 mg/m also caused perforated septums in some
workers.
Limited work has been reported on reproductive effects of chromium. Cr(VI)
and Cr(ITI) have been found to cross the placental barrier in animals (hamsters
and mice) and enter the fetus during mid to late gestation. Fetal uptake of
Cr(VT), however, was much greater than that of Cr(III). Developmental effects
attributed to both Cr(VI) and Cr(IIT) differed between hamsters and mice, and
included such external abnormalities as cleft palate and skeletal defects, and
2-e
-------
(in one study of a Cr(III) compound) neural tube defects. One researcher
concluded that Cr(VI) occurred at sufficiently high fetal concentrations to
cause direct effects on embryonic structures, but also questioned whether all of
the teratogenicity and fetal toxicity associated with exposure to Cr(ITT) might
be attributed to extra-embryonic effects, for example, those on placental
tissues.
2.5.2. Genotoxicity, Carcinogenicity and Assessment of Risk.
2.5.2.1. GENOTOXICITY — In recent years, much evidence has accumulated to
show that compounds of chromium possess the ability to cause transformations and
mutations, as evaluated in a wide variety of in vitro assays such as the reverse
and forward mutation, gene conversion, and DNA modification tests. Genotoxic
effects have been demonstrated primarily for chromium compounds containing the
Cr(VI) species, including effects such as:
— Mutagenic responses in bacterial strains.
— Morphologic changes in mammalian fetal cells.
— Cytogenic effects on mammalian bone marrow cells.
— Increased gene conversion in yeast species.
— Increased transformation frequencies in mammalian cells.
~ Chromosomal damage in cultures of human lymphocytes.
In general, soluble Cr(VI) compounds are less active in the presence of
metabolic activating systems. The reduction of Cr(VI) to Cr(III) by cellular
agents in metabolic activation systems, in part, explains the reduced mutagenic
activity of Cr(VI) in the presence of such activating systems. Some recent
evidence implicating both Cr(VI) and Cr(III) in induced mutagenesis has been
reported in DNA interaction and DNA polymerase infidelity assays, and several
tests with apparently pure Cr(III) samples have found chromosomal aberrations.
2-9
-------
However, with the exception of hexacoordinate Cr(ITI) compounds, and
Cr(III) compounds contaminated with Cr(VI), trivalent chromium is generally
considered to be a relatively inactive genotoxic agent, owing to its inability to
•»
cross cell membranes.
2.5.2.2. CHROMIUM CARCINOGENESIS AND ASSESSMENT OF BISK — The
epidemiologic studies of chromate production workers have demonstrated an
association of exposure to chromium compounds with respiratory cancer. Whether
the association implicates hexavalent chromium (Cr VI) alone, or trivalent (Cr
III) as well, is not definitively addressed by these studies. The strength of
the association is evidenced by the high relative risks of lung cancer (e.g., a
lung cancer mortality ratio of 29 was found in one study of chromate production
workers), the consistency of results by different investigators in different
countries, a dose-response relationship, and the specificity of the tumor site
(i.e., the lung). Results of three epidemiologic studies of chrome pigment
workers also provide suggestive evidence that these workers are at an elevated
risk of lung cancer. One epidemiologic study of chrome pigment workers (Davies
1978, 1979) suggested that zinc chromate was carcinogenic while lead chromate was
not. However, the usefulness of the data on the lead chromate pigment workers is
limited by small sample size. Using the criteria of the International Agency for
Research on Cancer (IARC), the epidemiologic studies of chromate production
workers would be classified as showing sufficient evidence of carcinogenicity.
Several hexavalent chromium compounds have been shown to be carcinogenic in
cancer bioassay studies. Only calcium chromate has consistently produced lung
tumors in rats by several routes of administration. Other chromium compounds—
strontium chromate, zinc chromate, sodium dichromate, lead chromate, lead
chromate oxide, and sintered chromium trioxide—have produced local sarcomas or
2-10
-------
lung tumors in rats at the site of intrabronchial, intratracheal, intramuscular,
subcutaneous and intraperitoneal application. Hexavalent chromium compounds
have not induced lung tumors by inhalation; however, studies have not been
reported in detail. Trivalent chromium compounds have not been reported to be
carcinogenic by any route of administration. Animal cancer bioassay studies
suggest that hexavalent chromium compounds (particularly soluble and sparingly
soluble compounds) are probably the etiologic agent in chromium related human
cancer. Under the TABC criteria, the animal bioassay studies would constitute
sufficient evidence of the carcinogenicity of hexavalent chromium compounds.
Using the IARC classification scheme, the level of carcinogenic evidence
available for the combined animal and human data would place hexavalent chromium
(Cr VI) compounds into Group 1, meaning that there is decisive evidence for the
human carcinogenicity of those compounds.
The lifetime cancer risk due to air containing 1 |ig/m of hexavalent
_2
chromium compounds is estimated to be 1.2 x 10 . This would place hexavalent
chromium (Cr VI) in the first quartile of the 53 compounds evaluated by the CAG
for relative potency.
2-11
-------
3. BACKGROUND INFORMATION
3.1. CHEMICAL AND PHYSICAL PROPERTIES
Chromium was discovered in 1797 by the French chemist, Louis Vanquelin.
Metallic chromium is steel gray in color, melts at 1857 + 20°C, boils at 2672°C,
and has a specific gravity of 7.20 at 28°C (Weast, 1980). As found in nature,
chromium is a mineral mixture of four stable isotopes of mass numbers 50, 52, 53,
and 5M (Gary, 1982).
The inorganic chemistry of chromium and its compounds has been extensively
studied. However, its physical or chemical forms and the mode by which they are
incorporated into biological systems are poorly characterized. Inorganic
chromium compounds occur in valence states ranging from -2 to -1-6.
Chemically, the Cr(III) state is the most stable and important form of
chromium. In neutral and basic solutions, Cr(III) forms binuclear and poly-
nuclear compounds in which adjacent chromium atoms are linked through hydroxy-
(OH) or oxo-(O) bridges. Interestingly, Cr(III) forms stable complexes with
amino acids and peptides (deMeester and Hodgson, 1977; deMeester et al., 1977).
Cr(III) also has a strong tendency to form hexacoordinated octahedral complexes
with ligands, such as water, ammonia, urea, ethylenediamine, halides, sulfates,
and organic acids. These relatively stable complex formations (Cotton and
Wilkinson, 1972; Kiilunen et al., 1983) can prevent precipitation of Cr(III) at
pH values at which it would otherwise precipitate.
Cr(VI) exists in solution as hydrochromate, chromate, and dichromate ionic
species. The proportion of each ion in solution is dependent on pH. In strongly
basic and neutral pHs, the chromate form predominates. As the pH is lowered, the
hydrochromate concentration increases. At very low pHs, the dichromate species
3-1
-------
predominates. In the pH ranges encountered in natural water, the predominant
forms are hydrochromate ions (63-6J) at pH 6.0 to 6.2 and chromate ion (95.7» at
pH 7.8 to 8.5 (Trama and Benoit, 1960). The oxidizing ability of Cr(VI) in
aqueous solution is pH dependent. The oxidation potential of Cr(VI) increases at
lower pHs. The ability of Cr(VI) to oxidize organic materials and the tendency
of the resulting Cr(III) to form stable complexes with available biological
ligands afford a reasonable mechanism by which chromium can interact with the
normal biochemistry of man (Towill et al.t 1978).
The physical properties of several environmentally significant trivalent
chromium compounds are shown in Table 3-1. The same parameters for selected
Cr(VI) compounds are shown in Table 3-2. It should be mentioned that there is
considerable disagreement in the literature concerning the physical parameters
given in Tables 3-1 and 3-2 and therefore these values should be accepted with
reservation. The disagreement in the values is possibly due to the reactions of
these compounds with other substances, namely with moisture and air at high
temperatures, impurities, and structural and compositional changes occurring
during the experimental determinations.
Reactions of Cr(III) in alkaline solution depend on the concentration of OH~
ions. When hydroxyl ions are added to Cr(III) solutions in quantities insuffi-
cient for precipitation, basic ions are formed, which then polymerize with the
formation of hydroxy- or oxy-bridged compounds as shown below:
Cr(NO,), » [Cr(H,0),]3+ -A-2S* [Cr(OH)(H?0),J2+
-> J c. o r
-------
U)
TABLE 3-1
Physical Properties of Selected Trlvalent Chromium Compounds*
Compound
Chromic acetate
Chromic chloride
Chromic chloride,
hexahydrate
Chromic formate,
hexahydrate
Chromic oxide
Chromic phosphate,
hydrated
Chromic sulfate
Chromic sulfate,
hydrated
Density,
Formula g/cmj
Cr(CH3COO)3 • H20
CrCl3 2.76 (15°C)
[Cr(H20),,Cl2]Cl • 2H20 1.76
[Cr(H20)g]Cl3 NR
[Cr(HCOO)3] • 6H20 NR
Cr_0, 5.21
CrPO.. • 2H,0 2.42 (32.5"C)
CrPOJj • 6h|o 2.121 (14«C)
Cr-CSOj)- 3.012
Cr (S0a), • 15H-0 1.867 (17*C)
Cr^soj)^ • 18H20 1.7 (22°C)
Melting
Point, °C
NR
«1150
83
NR
decomposes
above 300
2266
NR
100
NR
100
100(-12H20)
Boiling
Point, °C
NR
1300
(sublimes)
NR
NR
NR
4000
NR
NR
NR
100(-100 H-0)
NR
Solubility
in Hater, g/100 ml
slightly soluble
insoluble
58.5 at 25°C
soluble
soluble
insoluble
slightly soluble
insoluble
insoluble
soluble
120 at 20eC
•Source: Veast, 1980; The Herck Index, 1976
•• * Vot reported
-------
TABLE 3-2
Physical Properties of Selected Hexavalent Chromium Compounds8
Density,1*
Compound Formula g/cnr
Ammonium chromate (M^). CrO,. 1'^112
Ammonium dichromate (NH^) Cr-0_ 2.155-c
Barium chromate BaCrOH 4.498.,.
H 25
Chromium (VI) oxide CrO^ 2'7025
Lead chromate PbCrO^ ***1215
-«=• Mercurous (I) chromate Hg^rO^ NR
Mercuric (II) chromate HgCrO^ NR
Potassium chromate KjCrO^ 2. 732...,
Potassium dichromate KjCr^O- 2.67&25
Sodium chromate Na^rO,. 2.723,c
Sodium dichromate Na.Cr.O- • 2H.O 2.348..
dihydrate
Melting
Point, °C
180
decomposes
180
decomposes
decomposes
197
844
decomposes
decomposes
971
398
792
84.6
( incongr uent )
Boiling
Point, °C
NR
NR
NR
decomposes
decomposes
NR
NR
NR
500
decomposes
NR
400
decomposes
Solubility in
Water, g/100 ml
40.5 at 30°C
30.8 at 15°C
3.4 x 10"11
at 160°C
67.45 at 1000C
5.8 x 10~6 at
25°C
very slightly
soluble
slightly soluble,
decomposes
62.9 at 20*C
4.9 at 0°C
102 at 100°C
87.3 at 30°C
180 at 20°C
aSource: Weast, 1980; Hartford, 1979
The lower figures indicate the temperature (°C) at which the densities were measured.
NR = Not reported
-------
Such ions are of the proper size to cross-link protein fibers and may play an
important part in the chemistry of tanning. The single hydroxyl-bridged rhodo
and erythro binuclear Cr(III) amine complexes also have been extensively studied
(Veal et al., 1973; Cline et al., 1981).
When a sufficient amount of a base is added to Cr(III) salt solution, a
hydrous oxide of indefinite composition, Cr203»xH20, is precipitated. On addi-
tion of more base to the hydrous oxide, the precipitate redissolves, probably due
•3 „
to formation of complex ions of the type [Cr(OH)x]:>~ (e.g., [CrtOH)^]",
[Cr(OH)6]3").
Cr(III) compounds are reduced to Cr(II) compounds by hypophosphites,
electrolysis, or reducing metals, such as Zn, Mg, and Al in acid solution
(although it should be noted that Cr(II) compounds are stable only in the absence
2»
of air). In basic solution, Cr(III) is readily oxidized to CrOjj by
hypochlorite, hypobromite, peroxide, and oxygen under pressure at high
temperature. Heating of chromium compounds in air in the presence of alkalies
also yields chromate. In acid solution, Cr(III) is harder to oxidize and needs
strong oxidizing agents, such as concentrated HCIO^, sodium bismuthate, and
permanganate.
All Cr(VI) compounds except CrF, are oxo-compounds. Cr(VI) rarely occurs in
nature, apart from anthropogenic sources, because it is readily reduced by
oxidizable organic matter. However, after it is introduced into water, Cr(VI)
frequently remains unchanged in many natural water sources because of low
concentration of reducing matter. Cr(VI) occurs most commonly in the form of
chromate or dichromate, both of which are produced on a large scale in industry.
The dissociation equilibrium of chromic acid solution indicates the weaker
acidity of HpCrOj.. The dissociation of H^Cr 0 appears to be that of a strong
acid. Acid solutions of dichromate are powerful oxidizing agents:
3-5
-------
Cr20?2" + 1i»H* + 6e" »• 2Cr1"3 + 7H20, E° = 1.33V
Bichromate salts are the leading commercial form of Cr(VI).
3.2. PRODUCTION, USE, AND RELEASES TO THE ENVIRONMENT
The purpose of this document is to present available information relevant to
human health effects that could be caused by this substance.
Any information regarding sources, emissions, ambient air concentrations,
and public exposure has been included only to give the reader a preliminary
indication of the potential presence of this substance in the ambient air. While
the available information is presented as accurately as possible, it is
acknowledged to be limited and dependent in many instances on assumption rather
than specific data. This information is not intended, nor should it be used, to
support any conclusions regarding risks to public health.
If a review of the health information indicates that the Agency should
consider regulatory action for this substance, a considerable effort will be
undertaken to obtain appropriate information regarding sources, emissions and
ambient air concentrations. Such data will provide additional information for
drawing regulatory conclusions regarding the extent and significance of public
exposure to this substance.
3.2.1. Production of Chromium Compounds. The industrial processes for the
production of chromium metal and the various compounds have been described by
Hartford (1979). A simplified flow chart depicting these processes appears in
Figure 3-1.
There has been no mining of chromite ore in the United States since 1961.
Chromium ore and ferrochrome alloys are imported mainly from the Soviet Union,
South Africa, Turkey, and Zimbabwe.
3-6
-------
Chromitt
-------
The two primary industrial compounds of chromium made directly from chrome
ores are sodium chromate and sodium dichromate. Secondary chromium compounds
produced in substantial quantities include potassium chromate and potassium
dichromate, ammonium dichromate, chromic acid, and various formulations of basic
chromic sulfate used principally for leather tanning. The United Stated manufac-
turers of the three important chromium compounds, their production capacity, and
the amount produced are discussed below.
The manufacturers of both sodium chromate and sodium dichromate and their
annual production capacity in 1982 are given in Table 3-3. The estimated produc-
tion capacity is based on a 100$ sodium dichromate weight.
The manufacturers of chromic acid and their annual production capacity in
1982 are given in Table 3-**.
3.2.2. Uses of Chromium and Its Compounds. The United States consumption
pattern of chromium and its compounds for the year 1979 is shown in Table 3-5.
It can be seen from Table 3-5 that metallurgical and chemical usages
constituted 82% of the total United States chromium consumption in 1979.
Metallurgical grade chromite ore is usually converted into one of several types
of ferrochromium or other chromium metal that are alloyed with iron or other
elements, such as nickel and cobalt. A great variety of useful steels are
produced from these alloys. Because of their high melting points and chemical
inertness, chromite ore and chrome alloys are used by the refractory industry in
furnaces as linings, in the manufacture of furnace bricks, and as coating
materials to close pores and to join bricks within the furnace. Other uses of
chromite refractories include nonferrous alloy refining, glass making, and
cement processing. The pattern of chromium consumption in the United States has
been consistent over the last 20 years. However, the use of chromite and chrome
3-8
-------
TABLE 3-3
Manufacturers and Their Production Capacities of
Sodium Chromate and Sodium Dichromate3
Annual Production Capacity
Manufacturers in 1982, 103 metric tons
Allied Chemical Corp. 65
Baltimore, MD
American Chrome and Chens., Inc. 45
Corpus Christi, TX
Diamond Shamrock 94
Castle Hayne, NC
TOTAL 201
aSource: SRI International, 1982; 1982 U.S. Industrial Outlook,
Chemical Marketing Reporter, 1982.
Diamond Shamrock will increase capacity to 118,000 tons in January
1983.
3-9
-------
TABLE 3-1
Principal United States Manufacturers of Chromic Acid3'
Annual Capacity in 1982,
Manufacturers 10^ metric tons0
Allied Chemical Corp. 21
Baltimore, MD
Diamond Shamrock Corp. 24
Castle Haynes, NC
Source: SRI International, 1982; Hartford, 1979
Data on actual production of chromic acid are held to be
confidential to avoid disclosing proprietary
information on individual companies.
Q
The estimates for production capacity are based on a
100$ chromic anhydride (CrO ) basis.
3-10
-------
TABLE 3-5
United States Chromium Consumption Pattern in 1979a
Quantity Consumed 'c % Fraction of
10^ metric tons U.S. Consumption
Metallurgical
wrought stainless and heat 235 44
resisting steels
tool steels 6 1.1
wrought alloy steels 44 8.1
cast alloy steels 15 2.7
alloy cast irons 8 1.4
nonferrous alloys 15 .2.7
other 6 1.1
Total 329 61.0
Refractories
chrome and chrome-magnesite 16 3*0
magnesite-chrome brick 23 4.2
granular chrome-bearing 42 7.8
granular chromite 16 3.0
Total 97 18.0
Chemicals
pigments 29 5.4
metal finishing 24 4.4
leather tanning 18 3.3
drilling muds 5 0.9
wood treatment 7 1.3
water treatment 7 1.3
chemical manufacture 9 1.7
textiles 4 0.7
catalysts <2 0.3
other 9 1.7
Total 114 21.0
Grand Total 540 100
aSource: Hartford, 1979, Mineral Commodity Summaries, 1980.
Exclusive of scrap.
Columns may not total exactly due to rounding.
3-11
-------
alloys in the refractory industry is beginning to decline as open hearth furnaces
are replaced by basic-oxygen furnaces (Hartford, 1979). In the future, growth in
chromium usage is expected in the metallurgical and chemical sectors. A recent
study (Morning, 1977) has projected a 3.%% growth annually in the United States
chromium consumption, leading to a total chromium demand of 1 million tons
(900,000 metric tons) in the year 2000.
The consumption pattern of imported chromium for metallurgical, refractory,
and chemical usage is shown in Table 3-5. Chromium-containing pigments can be
primarily classified into chromate color pigments based on lead chromate,
Cr(III) oxide greens, and corrosion inhibiting pigments based primarily on zinc
chromate. In metal finishing, chromic acid is used in chromium plating of metal
surfaces. The chrome tanning of leather is one step in a complicated series of
operations leading from the rawhide to the finished product. The annual consump-
tion of hides by the leather industry is decreasing (Hartford, 1979), and the use
of Cr(VI) compounds for tanning purposes may be on the decline. Chromium
chemicals, such as chromium lignosulfonates, are used in drilling muds during the
drilling of wells to combat fatigue corrosion cracking of drill strings
(Hartford, 1979). Chromated copper arsenate is widely used as a wood preserva-
tive, especially in treating building lumber, and wood foundations. Chroraates
are used to inhibit metal corrosion in recirculating water systems, such as
cooling towers, locomotives, and automobiles. Sodium dichromate and various
chromic salts are employed in the textile industry to improve washfastness and to
oxidize the dyed textile. A large number of chromium compounds are used as
catalysts in various chemical reactions. Barium and calcium chromates are used
as activators and dipolarizers in fused salt batteries. Chromium dioxide (CrO )
is used as a ferromagnetic material in high-fidelity magnetic tapes.
3-12
-------
3.2.3. Releases to the Environment. Although the chromium industry in the
United States has adopted various pollution control measures, some release of
chromium compounds into the environment is occurring. Chromium compounds from
industrial operations enter the environmental air, water, and soil from several
sources. Kilns, smelting furnaces, boilers, leaching tanks, open boiling
vessels, plating tanks, and other installations emit dusts and mists containing
chromium to the atmosphere.
Chromium is a trace component of coal and oil and is released to the
atmosphere upon combustion of these fuels. Fly ash emitted from coal-fired power
plants contain 10 to 600 ppm chromium, depending on the type of boiler firing
(Block and Dams, 1976; Hock and Lichtman, 1982). For power plants equipped with
electrostatic precipitators (ESP) to control particulate emissions, the total
chromium concentration in the emission can be reduced (NAS, 197*0. Rinaldi et
al. (1980) have shown that the chromium concentration of particulates from
controlled coal combustion may be higher than the chromium concentration of
particulates from uncontrolled coal combustion, due to preferential chromium
enrichment on small particles that escape the ESP device.
Wood contains chromium and it is likely that the burning of wood in fire-
places and campfires may contribute small amounts of chromium into the atmos-
phere. Forest fires would therefore be a potential non-anthropogenic source of
atmospheric chromium (NAS, 197U). No estimate of the amount of chromium emitted
from forest fires could be obtained. However, Lantzy and MacKenzie (1979)
estimated chromium flux in the atmosphere from anthropogenic (industrial and
fossil fuel) and non-anthropogenic (continental dust, volcanic dust, and
volcanic gas) sources. The ratio of anthropogenic to non-anthropogenic atmos-
pheric flux of chromium was estimated to be 1.61. Incineration of municipal
3-13
-------
refuse and sewage sludge is also expected to contribute small amounts of chromium
into the atmosphere (Binaldi et al., 1980; Fiscus et al., 1978).
It is now known that asbestos contains as much as 0.15$ chromium (Towill et
v
al., 1978). Thus asbestos mining and the wearing of vehicular brake linings
represent potential sources of chromium in the atmosphere. Catalytic emission
control systems in automobiles using copper chromite reduction catalysts
represent another source of chromium emissions to the atmosphere (TABC, 1980).
Also, when chromate chemicals are used as corrosion inhibitors in recirculating
cooling waters, some chromate is lost to the atmosphere as mist.
According to a Radian Corporation (1984) report, United States industrial
and inadvertent sources of chromium emissions into the atmosphere under existing
controlled operations amounted to about 5,000 metric tons. Estimated amounts of
atmospheric chromium emissions are shown in Table 3-6.
The geographical distribution of atmospheric chromium emissions in the
United States is presented in Table 3-7. It is evident from Table 3-7 that the
Great Lakes area, the Southeast and the East coast south from New York constitute
the bulk of atmospheric chromium emissions in the United States.
Chromium compounds occur in a variety of industrial wastewaters and poten-
tially may enter surface water and groundwater supplies. Wastewaters from
electroplating operations, leather tanning, and textile manufacturing represent
the types of chromium-containing streams that may ultimately enter surface and
groundwaters (Hartford, 1979). It has been estimated that 220 metric tons/year
are discharged in Southern California coastal waters (Schafer, 1977). Ottinger
et al. (1973) estimated that 6200 metric tons of chromium are lost annually in
the sludge of solvent-based paints and another 437 metric tons are discharged as
paint residues.
-------
TABLE 3-6
Sources and Estimates of United States Atmospheric Chromium Emissions*
Source Category Chromium Emissions, Metric Tons/Year
Chrome Ore Refining 3
Ferrochromium Production 13
Chromium Chemicals Production M50-900
Refractory Production 90
Sewage Sludge Incineration 25
30
Steel Production 2870
Utility Cooling Towers 5
Cement Production 16
Combustion of Coal and Oil
boilers 737
process heaters 556
Total 4525-5275
•Source: Radian Corporation, 1984
3-15
-------
TABLE 3-7
Regional Distribution of Principal Chromium Emissions'
EPA Region/States
I/CT,ME,MA,NHtRI,VT
II/NJ,NY,PR,VI
III/DE,MD,PA,VA,WV,DC
IV/AL,FL,GA,KY,MS,NC,SC,TN
V/IL,IN,MI,MN,OH,WI
VI/AR,LA,NM,OK,TX
VII/IA,KS,MO,NB
VIII/CO,MT,ND,SD,UT,WY
IX/AZ,CA,NV,HA
X/AK,ID,OR,WA
TOTAL
Annual Chromium Emissions
From Sources in this Region,
metric tons (Mg)
103
3,140
2,800
3,000
4,830
74
177
107
174
898
15,300
Percent of Total
U.S. Chromium
Emissions
0.6
19.0
17.3
18.5
29.0
0.5
1.1
0.7
1.1
5.4
93.2
Source: OCA Corporation, 1973
Sources include ferrochrome production, refractory production, cement production,
chrome steel production, and coal and oil combustion.
3-16
-------
Solid waste streams containing Cr(VI) constitute the primary problem area
involving chromium solid wastes (Hartford, 1979). Wastes resulting from the
roasting and leaching steps in the chromate manufacturing process traditionally
contain residual Cr(VI). If landfilled, the residual Cr(VI) can slowly leach
into surrounding waters via desorption and disproportionation (Hartford, 1979).
An estimate of the total amount of chromium released into soil and groundwater as
a result of the leaching of chromium-containing solid wastes is not available.
3.3. ENVIRONMENTAL FATE AND TRANSPORT
3.3.1. Air. Little information exists in the literature regarding the nature of
the chemical species present in the atmosphere away from obvious sources of
pollution. Under normal conditions, Cr(III) and Cr(0) are relatively unreactive
in the atmosphere (Towill et al., 1978). Cr(VI) in air may react with
particulate matter or gaseous pollutants to form Cr(III) (NAS, 1974). However,
these atmospheric reactions have not been extensively studied.
Low concentrations of chromium enter the atmosphere as a result of indus-
trial activities and soil-derived aerosols (Towill et al., 1978). Chromium is
removed from air through wet and dry depositions. The total yearly deposition of
2 2
chromium in urban areas may vary from 0.12 pg/m to 3 pg/m (Towill et al.,
1978). In general, urban areas have higher total deposition than rural areas.
Chromium concentration in a wet deposition may vary from 0.004 to 0.060 pg/m&
and 0.0006 to 0.034 pg/fc for urban and rural areas, respectively (Towill et al.,
1978). The precipitated chromium from the air enters the surface water or soil.
Chromium particles of aerodynamic equivalent diameter <20 pm may remain
airborne for long periods and may be transported great distances by wind currents
3-17
-------
and diffusion forces (Sehmel and Hodgson, 1976). Therefore, atmospheric
conditions play an important role in determining the chromium concentration
around emission sites; however, no data relating atmospheric chromium content to
atmospheric or meteorological conditions could be found in the literat\ire.
3.3.2. Water and Sediments. Surface runoff, deposition from air, and release of
municipal and industrial wastewaters are the sources of chromium in surface
waters. Chromium may be transported in five forms (see Table 3-8) in surface
waters.
It can be seen from Table 3-8 that most of the chromium in surface waters
may be present in particulate form as sediment. Some of the particulate chromium
would remain as suspended matter and ultimately be deposited in sediments.
DeGroot and Allersma (1973) found a chromium ratio in water to suspended matter
to be 1 to 2.3 for the Rhine River. In the heavily polluted Qishon-Gadura River
in Israel, chromium concentrations in water were <10 ppb, while sediment
contained from 220 to 610 ppm chromium (Towill et al., 1978).
The exact chemical forms of chromium in surface waters are not well defined.
Although most of the soluble chromium in surface water may be present as Cr(VI)
(Towill et al., 1978), a small amount may be present as Cr(III) organic complexes
(DeGroot and Allersma, 1978; Fukai, 1967). Schroeder and Lee (1975) studied the
transformation between Cr(III) and Cr(VI) in natural waters. They found that
only 3$ of Cr(III) was oxidized by 0 in 30 days at ambient temperature. Cr(VI)
is the major stable form of chromium in seawater (Fukai, 1967); however, Cr(VI)
may be reduced to Cr(III) by organic matters present in water and may eventually
deposit in sediments. Lu and Chen (1976) found that chromium was not signifi-
cantly released from sediments into seawater under either oxidizing or reducing
conditions. Eisenreich (1982), however, noted that in some waterbodies, such as
3-18
-------
TABLE 3-8
Five Forms of Chromium Transported in the Yukon and Amazon Rivers*
Physical Form
In solution and organic complexes
Adsorbed
Precipitated and co-precipitated
In organic solids
In sediments
Percent Present
In
Amazon River Yukon River
10.4
3.5
2.9
7.6
75.6
12.6
2.3
7.2
13.2
61.5
•Source: Towill et al., 1978
3-19
-------
Lake Superior, chromium occurred in trace amounts from in-lake processes. He
estimated an atmospheric chromium flux of 0.9 Hg/m /day on the surface film.
3.3.3. Soil. Most soil chromium is in mineral, absorbed, or precipitated form.
Chromium probably occurs as the insoluble Cr(III) oxide (Cr 0-«nH_0) in soil, as
the organic matter in soil is expected to reduce any soluble chromate to
insoluble Cr20-. Chromium in soil can be transported to the atmosphere by way of
aerosol formation (John et al., 1973; Zoller et al., 1971). Chromium is also
transported from soil through runoff and leaching of water. Runoff could remove
both chromium ions and bulk precipitates of chromium with final deposition on
either a different land area or a water body. In addition, flooding of soils and
the subsequent anaerobic decomposition of plant matters may increase dissolution
of Cr(III) oxides in the soil (Towill et al., 1978).
3.4. LEVELS OF CHROMIUM IN VARIOUS MEDIA
3.1.1. Ambient Air. Background concentrations of chromium in various media have
been reported by Cary (1982). Citing the work of Maenhaut and Zoller (1977), he
noted chromium concentrations at the South Pole were approximately 5 pg/nr (0.005
ng/m ). He noted further that Duce and Zoller (1975) found that chromium levels
over the Atlantic Ocean ranged from 0.007 to 1.1 ng/nr. The authors believed
that this trace amount of chromium originated from a crustal source.
The chemical form of chromium in air depends on the source of emission. The
majority of chromium in the atmosphere, originating from such sources as
metallurgical production, coal and oil combustion and cement production is
usually in the Cr(III) or Cr(0) state. Chrome production, chrome plating, and
cooling tower drifts are primary examples of the sources of Cr(VI) in the
3-20
-------
atmosphere (Towill et al., 1978). The mass median diameter of chromium in air
particulate matter is in the range of 1.5 to 1.9 nm (Cawse, 1974; Lee and von
Lehmden, 1973). Chromate salts are often used in the cooling tower water as a
corrosion inhibitor. Therefore, cooling tower drift consisting of water
droplets formed mechanically within the towers and carried by wind into the
surrounding area may be a source of high Cr(VI) concentrations in air. Air
concentrations of chromium near a cooling tower at the Oak Ridge Gaseous
Diffusion Plant, Oak Ridge, Tennessee, were about 50 ng/nr from distances up to
660 feet (200 m) from the tower (Alkezweeny et al., 1975). Hourly chromium
2 2
deposition was about 1 mg/m at 100 feet (30 m) and about 0.01 mg/m at 330 feet
(1000 m) from the tower. Therefore, substantial amounts of chromium may be
present in the respirable particulate fraction and can be deposited in the
respiratory tract.
The concentrations of total chromium measured in the ambient air of many
urban and nonurban areas of the United States during 1977 to 1980 are given in
Table 3-9. The data in Table 3-9 were obtained from the U.S. EPA's National
Aerometric Data Bank, which is maintained by the Agency's Monitoring and Data
Analysis Division (MDAD) at Research Triangle Park, North Carolina. None of the
data in Table 3-9 have been previously published. The ambient chromium data have
been collected by monitoring networks operated by various State and local air
pollution control agencies as required by the Clean Air Act. After passing
editorial and validation checks which are performed by EPA regional offices, the
data are forwarded to MDAD for incorporation into the National Aerometric Data
Bank.
During the 1977-1980 period, the mean chromium concentrations measured in
the United States (given in Table 3-9) ranged from 0.0052 |ig/nr (24-hour average
3-21
-------
TABLE 3-9
Total Chromium Concentrations Measured in the Ambient Air
of Selected Sites in the United States During 1977-19803
Site
Grand Canyon, National Park, AZd
Los Angeles, CA
Waterbury, CT
Atlanta, GA
Hawaii County, HI
Kansas City, KS
Iberville Parish, LA
Acadia National Park, MEC'd
Baltimore, MD
Worcester, MA
Bayonne, NJ
Newark, NJ
Niagara Falls, NY
Year
1977
1977
1978
1979
1977
1980
1977
1977
1978
1980
1977
1978
1980
1977
1977
1979
1977
1978
1979
1977
1978
1980
1978
1979
1980
1979
1980
Total Chromium
Arithmetic
Mean
0.0058
0.0188
0.0342
0.0326
0.0089
0.0062
0.0063
0.0167
0.0276
0.0191
0.0063
0.0059
0.0052
0.0052
0.1568
0.0935
0.0063
0.0099
0.0067
0.0105
0.0149
0.0123
0.0181
0.0129
0.0091
0.0389
0.0144
Concentration, pg/m
Maximum Observed
Value0
0.0134
0.0666
0.2178
0.1396
0.0441
0.0194
0.0216
0.0413
0.0724
0.0358
0.0159
0.0128
0.0052
0.0052
0.2470®
0.4589
0.0167
0.0239
0.0166
0.0253
0.0324
0.0508
0.0301
0.0333
0.0369
0.5590
0.0603
3-22
-------
TABLE 3-9 (cont.)
Site
Akron, OH
Cincinnati, OH
Steubenville, OH
Black Hills National Forest, SDC
Chattanooga, TN
Norfolk, VA
Tacoma, WA
Year
Total Chromium Concentration, pg/nT
Arithmetic Maximum Observed
Mean
b
Value
1977
1978
1979
1980
1977
1978
1979
1980
1978
1979
1978
1977
1978
1979
1980
1977
1978
1979
1980
1977
1978
1980
0.0126
0.0188
0.0166
0.0201
0.0083
0.0116
0.0451
0.0150
0.0517
0.1212
0.0090
0.0122
0.0140
0.0112
0.0150
0.0067
0.0069
0.0083
0.0119
0.0099
0.0249
0.0104
0.0610
0.0528
0.0389
0.0710
0.0377
0.0294
0.4316
0.0718
0.2602
0.6839
0.0295
0.0453
0.0463
0.0760
0.0705
0.0152
0.0158
0.0291
0.1456
0.0330
0.1425
0.0283
a
Source: Unpublished data in the National Aerometric Data Bank maintained by
the Monitoring and Data Analysis Division of EPA, Research Triangle Park, NC.
Values represent annual average.
CValues represent maximum 24 hour averages.
Background sites; all other sites are determined to be populated urban areas,
Corrected from Maryland State Yearly Air Quality Data Report, Baltimore MD,
March, 1978.
3-23
-------
background level) to 0.1568 ug/m3 (urban annual average). The highest maximum
24-hour average of 0.6839 ug/m3 was recorded in Steubenville, OH in 1979. For
the sites with >2 years worth of data, no discernible upward or downward
concentration trends are evident. For example, in Newark, New Jersey, the mean
chromium concentration dropped from 0.0181 ug/m3 in 1978 to 0.0129 ng/m3 in 1979
and to 0.0091 ug/nr in 1980. However, in Norfolk, Virginia, over the same period
the chromium levels rose from 0.0069 ug/m3 in 1978 to 0.0083 pg/m3 in 1979 and to
0.0119 mg/ar in 1980. In Akron, Ohio, the mean chromium concentration in 1978
was determined to be 0.0188 ug/m3. In 1979, the concentration in Akron dropped
to 0.0116 ug/m3 but in 1980 it rose again to 0.0201 ug/m3. The mean chromium
concentrations in nonurban, background areas such as national parks ranged from
0.0052 ug/m3 to 0.0090 ug/m3 over the 1977 to 1980 period.
Some source categories emit trivalent chromium, some emit hexavalent
chromium and others emit a combination of the two. Sources likely to be
trivalent emitters are refining, cement production and coal and combustion.
Hexavalent sources include chrome plating and cooling towers. Those that emit
both forms include chromium chemicals production, refractory production, steel
manufacturing, and sewage sludge and municipal refuse incineration. The
relative proportions of each form emitted are unknown at this time. Maximum
annual average concentrations predicted, based on dispersion modeling, with 20
km of these source categories range from 0.01 ug/m for refractory production.
3.4.2. Aquatic Media. Naturally occurring chromium concentrations in water
arise from the weathering of minerals, soluble organic chromium, sediment load
and precipitation (Cary, 1982). Of 170 lakes sampled in the Sierra mountains in
California, only two contained up to 5 ug/S, (ppb) chromium; the mean pH was 6.0
(Cary, 1982). Both the Amazon and Yukon rivers, which are considered to
represent unpolluted watershed systems, reportedly contained 2.0 and 2.3 ug/Jl
3-24
-------
(ppb) chromium, respectively (Gibbs, 1977). These background freshwater
chromium levels compare well with estimates given by Bowen (1979) (median of 1
pg/Jl (ppb) chromium within a range of 0.1 to 6 pg/Jl (ppb)). Morevoer, Gary
(1982) notes that over 35% of the stream and river water sampled in Canada for
chromium contained less that 10 pg/Jl (ppb). The chromium levels detected in a
few surface waters and groundwaters are presented in Table 3-10. The amounts of
chromium found in these waters are usually related by the authors to
anthropogenic input. For example, it has been shown by Kopp and Kroner (1967)
that water from Lake Michigan near industrial discharge points contained a higher
level of chromium than lake water contaminated with lesser industrial input (5 to
19 ppb, compared to 2 to 1 ppb).
The valence of chromium in surface water can be either VI or III. Although
Cr(VI) is the more stable species in seawater, Fukai (1967) provided data to show
that both Cr(III) and Cr(VI) were present. Surface seawater contained Cr(III)
and Cr(VI) in the range of 0.02 to 0.11 ppb and 0.28 to 0.36 ppb, respectively,
while seawater at depths of 5, 500, and 1000 m contained about the same level
(0.2 ppb) of Cr(III) and Cr(VI). Cranston and Murphy (1978) found that the
ratios increased at depths greater than 100 m. In areas of the Pacific Ocean,
chromium levels averaged 0.12 pg/Jl (ppb). Above 100 m, 83$ occurred as Cr(VI);
below 100 m chromium averaged 0.16 pg/Jl (ppb), and CR(VI) accounted for
approximately 90$.
In a survey of 14 groundwater and 69 surface water supplies in 83 United
States cities in Illinois, Indiana, Michigan, Minnesota, Ohio, and Wisconsin,
the chromium level in the raw waters used for drinking waters were found to range
between <5.0 and 17.0 ppb (see Table 3-11) (U.S. EPA, 1975). Groundwater
contamination of 220 pg/Jl (ppb) Cr(VI) was reported by Robertson (1975) in ground
water of Paradise Valley, AZ.
3-25
-------
TABLE 3-10
Chromium Levels in a Few Surface Waters and Groundwaters4
Source
Lake Tahoe
Colorado River
Columbia River
Mississippi River
Missouri River
Ohio River
U.S. surface waters
U.S. surface waters
Natural water, Oak Ridge, TN
Water near cooling tower,
Oak Ridge, TN
Uncontaminated stream, NY
Contaminated stream, NY
Uncontaminated well, NY
Contaminated well, NY
Illinois River
Spring water, CA
Well water, CA
Stream water, CA
Seawater, CA
Frequency of
Detection %
NR
12
87
23
10
20
25
25
NR
NR
NR
NR
NR
NR
NR
6
5
0
0
Cone, (ppb
with Detec
Average
<0.62
NR
NR
NR
NR
NR
<1
9.7
NR
NR
<10
1250
<10
6000
21
NR
NR
NR
NR
or pg/fc) in Samples
table Chromium Level
Range *
<0.07 to <0.91
10 to 30
1 to 10
3 to 20
8 to 10
i| to 16
<1 to 19
1 to 112
50 to 120
2500 to 2790
NR
NR
NR
NR
5 to 38
0 to 21
0 to 13
NR
NR
aSource:" Towill et al., 1978
NR = Not recorded
3-26
-------
TABLE 3-11
Chromium Concentrations in U.S. Drinking Waters'
Concentration, pg/& or ppb
Water
Tap water, Dallas, TX
100 largest cities, U.S. (1962)
380 finished waters, U.S.
(1962-1967)
3834 tap waters, U.S.
(1974-1975)°
83 Midwestern cities, U.S.6
115 Canadian municipalities
(1976-1977)
r
Median
4b
0.4°
K
7.5b
Range
1 to 20
0.2 to 35
1 to 29
1.8°
NR
<2.0
0.4 to 8S
<5.0 to 17.0
<2.0 to 4.1
aSource: NAS, 1977
Average value; sampling date unavailable
cMedian value
Greathouse and Craun, 1978
eU.S. EPA, 1975; sampling date unavailable
fMeranger et al., 1981
g28$ of areas had detectable levels
NR = Not recorded
3-27
-------
The chromium concentration in various United States drinking water supplies
is presented in Table 3-11. In a survey of 2595 water samples from 969 water
supplies in the United States, only four samples showed chromium levels above the
detection limit of 50 ppb (McCabe et al., 1970). The maximum chromium concen-
tration in water detected in this survey was 80 ppb.
In a more recent survey (197*1 to 1975) with an analytical method of better
sensitivity, 3834 tap waters from 35 geographical locations representative of
the U.S. population were monitored for metal content (Greathouse and Craun,
1978). The detection limit for Cr in this survey was 0.1 ppb. The results of
this survey, presented in Table 3-11, also indicated that 28J of the area
surveyed had Cr levels above the detection limit. It should be mentioned that
the mean value of Cr in this study may be a little higher than reported since the
tap waters were not adequately flushed before collection.
3.4.3. Aquatic Suspended Materials and Sediments. The concentration of chromium
in suspended materials in several United States rivers was found to range between
37 and 460 ppm on a dry weight basis (Turekian and Scott, 1967). Chen et al.
(1974) determined the concentration in dry season suspended silts of Southern
California waters to be =500 ppm in "natural" areas and 2000 ppm in urbanized
areas.
Chromium concentrations determined for a variety of bottom sediments are
shown in Table 3-12. For the areas in the United States sampled, sediment
chromium levels ranged from about 1 to 450 ppm. An examination of Table 3-12
shows that the chromium concentration in sediments from several Wisconsin lakes
and Southern Lake Michigan does not significantly decrease with depth. A similar
finding concerning sediment chromium content was made by Bruland et al. (1974)
upon the analysis of chromium levels in sediments of different depths from the
3-28
-------
TABLE 3-12
Concentration of Chromium in Sediments*
Area
Delaware Bay
New York City Bight:
Background
Sludge dumping area
Puget Sound
Houston ship channel, TX
Neches River, TX
Sabine River (low in
industrial activity)
TX-LA Border
Southern Lake Michigan:
surface sediments
sediments from >15 to
100 cm depth
estimated background
Illinois River
Non-industrial stream, IL
Buzzard Bay, MA
Wisconsin Lakes:
surface sediment
•Source: Towill et al., 1978
NR s Not recorded
Chromium Concentration (ppm)
Median Range
NR
33 to
106
a 105
NR
, TX NR
NR
NR
y)
an:
77
to 52
nd NR
17
m, IL 6
33
NR
cm depth NR
2 to 310
50 to 209
43 to 154
39 to 254
8 to 288
41 to 89
35 to 165
32 to 68
20 to 40
2 to 87
3 to 7
NR
1 to 49
0.8 to 35
3-29
-------
Southern California basin. These studies of sediment chromium content versus
depth appear to indicate that natural background chromium levels contribute
heavily to the chromium levels found in surface sediments.
The concentration of chromium in an incinerated sewage sludge" ash was
determined to be 5280 pg/g (ppm). The concentrations of metals in ash residue
after incineration is «4 times those present in dried sludge (Fraser and Lum,
1983). Therefore, based on this one analysis, sewage sludges may contain high
concentrations of chromium.
3.4.4. Soil. The concentration of chromium in soil varies in accordance with
soil origin. Soils derived from nonserpentine areas can contain from traces to
300 ppm chromium. Soils derived from serpentine areas can contain up to 2% by
weight of chromium. The chromium concentrations in selected soils from various
parts of the U.S. are shown in Table 3-13. Mean chromium levels in Canadian
soils were 13 ppm (McKeague and Wolynetz, (1980). Most chromium in soils is
apparently insoluble. Extraction of soils with 2.5% acetic acid, ammonium
acetate at pH 4.8, and even with 0.1 N HC1 have shown that only 0.01 to 4$ of the
total soil chromium is extractable (Towill et al., 1978). Generally, chromium
concentrations in soils correlate well with chromium concentrations in parent
rock (Gary, 1982).
3.^.5. Food. The chromium content of a variety of foods is presented in
Table 3-14. The values given in Table 3-14 represent the average of several food
items- in each category. It can be seen from this table that the chromium
concentrations in different categories of food determined by Schroeder et al.
(1962) were lower than the values determined by other investigators. It is not
known whether the observed discrepancies are due to geographical and seasonal
3-30
-------
TABLE 3-13
Chromium Content in Selected United States' Soils*
Location
Pennsylvania
Peninsular Florida
Florida
Missouri
New Jersey
Michigan
Soil
Characteristic
agricultural surface
and subsoil
surface and subsoil
surface and subsoil
on and off road soil
various soils
various surface soils
•Source: Towill et al.f 1978
NR = Not recorded
Chromium content
ppm or pg/g
Range Median
NR
<1 to 1000
<1 to 500
NR
29 to 75
3.2 to 17.6
50
NR
71
NR
NR
3-31
-------
TABLE 3-1*1
Chromium Content in Various U.S. Foods
Sample
Mean Concentration
(ppm or pg/g)
Reference
Fresh vegetables
Fresh vegetables
Fresh vegetables
Frozen vegetables
Canned vegetables
Fresh fruits
Fresh fruits
Fruits
Canned fruits
Dairy products
Whole fish
Meat and fish
Meat and fish
Sea foods
Sea foods
Grains and cereals
Grains and cereals
Fruit juices
0.14
0.03 to 0.05
0.14
0.23
0.23
0.09
0.19
0.02
0.51
0.10
0.05 to 0.08
0.23
0.11
0.12
0.47
0.04
0.22
0.09
Thomas et al. , 1974
Schroeder et al., 1962
Toepfer et al., 1973
Thomas et al., 1974
Thomas et al., 1974
Toepfer et al., 1973
Thomas et al. , 1974
Schroeder et al., 1962
Thomas et al. , 1974
Schroeder et al., 1962
Okuno et al., 1978
Toepfer et al., 1973
Schroeder et al., 1962
Zook et al. , 1976
Meranger and Somers, 1968
Schroeder et al., 1962
Toepfer et al. , 1973
Meranger, 1970
3-32
-------
variations in trace element content of foods or due to errors in analytical
determinations.
The chromium content in acidic foods is often higher than other categories
of foods. The values of Cr content in a few commercial acidic foods which had
been in contact with stainless steel surfaces during harvesting, processing, or
preparation for market are shown in Table 3-15 (Stoewsand et al., 1979).
Jennings and Howard (1980) reported slightly higher levels of Cr in British
commercial alcoholic beverages than the Cr content in U.S. wines as given in
Table 3-15. The chromium contents in wines, beers, and spirits were reported to
be 0.45, 0.30, and 0.135 mg/Jl (ppm), respectively. However, it is difficult to
assess the daily chromium intake in the United States from the Cr content in
individual foods. The chromium contents in a selected diet composite were
determined by Kumpulainen et al. (1979). This study is the most recent diet
study conducted by FDA-USDA (personal communication with Dr. W.R. Wolf, USDA).
The chromium intake from typical American diets containing 43? fat was determined
to be 62 + 28 pg/day. The corresponding intake from typical American diets
containing 25% fat was determined to be 89 ± 56 pg/day.
3.4.5.1. BIOCONCENTRATION IN FOOD CHAINS — Several authors have found that
chromium concentrations decrease in higher trophic level organisms in aquatic
ecosystems (Towill et al., 1978). For example, Mathis and Cummings (1973)
detected =10 ppm chromium in worms, -5 ppm in clams, =1.2 ppm in omnivorous fish,
and =1 ppm in carnivorous fish. Lack of assimilation of chromium is probably the
major reason that the organisms of the higher trophic levels contain lesser
amounts of chromium.
A bioconcentration factor (BCF) is the ratio of the concentration of a
chemical in aquatic species to the concentration in the water in which they live.
3-33
-------
TABLE 3-15
Concentration of Chromium in a Few Commercial Grade Acidic Foods3
Commodity
Red cabbage
Red cabbage brine
Sauerkraut
Sauerkraut brine
Sauerkraut
Sauerkraut brine
Different brand
of honey
Vinegar
Hard cider
Cheese whey
Wine, Catawba
Wine, red concord
Wine, red concord
Container
glass
glass
can
can
flexible
pouch
flexible
pouch
glass
bulk
bulkb
bulk
lastiglass
(1 year)
redwood
(8 months)
stainless
steel
(7 months)
Sample
pH
—
3.4
—
3.5
—
3.5
3.5 to 3.7
2.8
3.5
1.8
3.1
3.2
3.2
Cr Concentration,
ppm or pg/g (wet wt.)
2.6
10.3
0.3
0.3
0.2
0.2
O.OM to 0.18
0.01
0.001*
0.01
0.02
0.07
0.02
a
Source: Stoewsand et al., 1979
'Before acidification
3-31
-------
An appropriate BCF can be used with data concerning food intake to calculate the
amount of chromium which might be ingested from the consumption of fish and
shellfish. Residue data for a variety of inorganic compounds indicate that
bioconcentration factors for the edible portion of most aquatic animals is
similar, except that for some compounds bivalve molluscs (clams, oysters,
scallops, and mussels) should be considered a separate group. An analysis (U.S.
EPA, 1980b) of data from a food survey was used to estimate that the per capita
consumption of freshwater and estuarine fish and shellfish is 6.5 g/day
(Stephan, 1980). The per capita consumption of bivalve molluscs is 0.8 g/day and
that of all other freshwater and estuarine fish and shellfish is 5.7 g/day.
The BCF for Cr(VI) in fish muscle appears to be less than 1.0 (Buhler et
al., 1977; Fremm and Stokes, 1962), but values of 125 and 192 were obtained for
oyster and blue mussel, respectively (U.S. EPA, 1980c). For Cr(III), BCF values
of 316, 153, and 86 were obtained with the American oyster (Shuster and
Pringle, 1969), soft shell clam, and blue mussel (Cappuzzo and Sasner, 1977),
respectively. It appears that the two valence states of chromiumdll and VI)
have about the same BCF values, and that the geometric mean of 130 can be used for
•bivalve molluscs. If the values of 0.5 (BCF for fish and mussels) and 130 (BCF
for bivalve molluscs) are used with the consumption data, the weighted average
bioconcentration factor for chromium in the edible portion of all freshwater and
estuarine aquatic organisms consumed by Americans can be calculated to be 16 on
the basis of per capita consumption of 0.8 g/day and 5.7 g/day for bivalve
molluscs, and fish and shellfish, respectively (U.S. EPA, 1980b).
3*4.6. Cigarettes. Chromium has been determined to be a component of cigarette
tobacco. Tobaccos grown in the United States have been found to have a chromium
content ranging from 0.24 to 6.3 mg/kg (ppm) (IARC, 1980).
3-35
-------
3.5. INDICES OF EXPOSURE
Past exposure to low levels of chromium may be associated with higher than
normal levels of chromium in the blood, urine and hair. In hair samples, the
relationships are tenuous in young children and women, as a result of variation
in chromium levels related to age and pregnancy (Creason et al., 1975). In blood
and urine, marked variations have been reported in the linearity between exposure
levels and the levels in body fluids as a result of sequestering and release of
chromium from body depots.
3.5.1. Chromium in Blood. Chromium is absorbed through both the respiratory and
gastrointestinal tracts (U.S. EPA, 1978). In the respiratory tract, water and
serum soluble chromates are absorbed into the blood system, whereas insoluble
Cr(III) particles and the inert oxides and hydroxides of Cr(III) remain in lung
tissue (U.S. EPA, 1978). In the blood stream, chromium compounds are bound by
proteins (Gray and Sterling, 1950). It has been shown that ionic Cr(VI)
(injected intravenously) passes through the membrane of red blood cells and binds
to the globin moiety of hemoglobin. Once inside the erythrocyte, Cr(VI)
compounds are rapidly reduced to Cr(III) and are unable to pass through the cell
membrane (Aaseth et al., 1982; Yamaguchi et al., 1983). In healthy red cells,
Cr(III) is partially bound to hemoglobin and partially to small molecular weight
substances.
Chromium disappears quickly from the blood and is taken up by other tissues
in the body, where it is concentrated much more heavily (by a factor of 10 to 100)
than in the blood. Therefore, blood levels of chromium may not be a usable
indicator of chromium nutritional status (Mertz, 1969; Mertz and Roginski,
1971).
A wide range of values for chromium content in blood has been reported.
Schroeder et al. (1962) reported chromium levels in serum ">f 0.52 and 0.17 ppm,
3-36
-------
whereas Doisy et al. (1969, 1971) found a chromium concentration of 2 ppb in
serum. Other chromium values reported have ranged from 0.11 to 55 ppb in human
plasma, and from 5 to 51* ppb in red blood cells (Underwood, 1971). Imbus et al.
(1963), working with United States subjects, found blood chromium levels ranging
from 13 to 55 ppb with a median of 27 ppb, while Hamilton et al. (1973), studying
subjects from the United Kingdom, reported a blood level of 70 ppb chromium.
However, researchers have discovered, through the use of new technology in
flameless AAS analyses, that the amount of chromium in normal blood and urine was
an order of magnitude lower than previous measurements had shown. It was
considered by Guthrie et al. (1978), Kayne et al. (1978) and Anderson (1981) that
measurements made before 1978 are probably high as the result of inadequate
background correction for non-atomic absorption. Thus, measurements of absolute
values of chromium in normal body fluids made before 1978 are unreliable (Toxic
Material News, 1982), although they were useful as pioneering attempts to
elucidate chromium metabolism.
Kayne et al. (1978) used flameless AAS with a tungsten halogen light source
for background correction to determine the serum chromium levels in 8 normal male
subjects. The tungsten halogen light greatly improved the background correction
in the near UV region where chromium is detected as compared to the standard D
light source. With the elimination of non-atomic absorption, the mean level of
serum chromium was determined to be 0.1^1 ug/Jl (ppb). These results were in
agreement with those of Versieck et al. (1978) in which serum chromium was
determined by neutron activation analysis. Serum obtained from 20 healthy
subjects (duplicate samples from 11 subjects) had a range of chromium levels from
0.0382 to 0.351 pg/£ (ppb) with a mean value of 0.16 pg/Jl (ppb). Using this data,
the authors concluded that normal human chromium levels in serum are in the
sub-ppb range. Plasma or whole blood chromium levels have ranged from U-70 ppb,
with a mean value of 30 ppb (total chromium) (Lin, 1983).
3-37
-------
3.5.2. Chromium in Urine. A wide range of values for chromium content in urine
has been reported. Hambidge (1971) reported chromium levels in urine of 8.4 ppb
for adults and 5.5 ppb for children over a 24-hour period. Imbus et al. (1963)
reported median urinary concentrations of chromium for adult males of^3.77 pg/fc
(ppb). Renal excretion is the major pathway of chromium elimination, with >BO%
of injected chromium excreted in this manner (Mertz, 1969).
As discussed with chromium levels in serum, difficulties in determining the
low levels of chromium in urine were not resolved until 1978, and presently,
normal human levels of chromium are considered to be in the sub-ppb range
(Anderson, 1981). Using flameless AAS with background correction by either a
tungsten halogen lamp or a continuum source, echelle, wavelength modulated AA
system (CEW-AA), normal human urinary chromium levels were determined to be >1
pg/A (ppb) (Guthrie et al., 1978), >0.9 pg/& (ppb) for most of 66 samples (Kayne
et al., 1978), and between 0.05 to 0.58 pg/£, (ppb) for 48 males and 28 females
(Anderson et al., 1982). Veillon et al. (1979) obtained excellent agreement
between chromium determinations of pooled urine samples using CEWM-AA and
chromium determinations using stable isotope dilution methods measured by GC/MS.
The mean values for the respective methods were 0.34 +0.1 and 0.32 + 0.02 pg/fc
(ppb). Increases in urinary chromium in humans receiving a daily supplement of
200 pg Cr as CrCl- ranged from mean levels of 0.2 pg/fc (ppb) prior to treatment to
1.02 and 1.13pg/& (ppb) after 2 and 3 months. Although Anderson (1981) cautions
against accepting absolute values for chromium from earlier studies, comparisons
and trends determined in a given study may be valid.
•Franchini et al. (1975) and Borghetti et al. (1977) reporting about workers
exposed to chromium in the chromium-plating industries showed that urinary
excretion and renal clearance of diffusible chromium are biological indices to
evaluate the degree of current exposure and the body burden of the compound.
3-38
-------
Franchini et al. (1978) confirmed their earlier results with an experimental
investigation using rats.
Other authors have demonstrated a close relationship between the amount of
Cr(VI) in the air and urinary excretion (Gylseth et al., 1977) or urina'ry excre-
tion of the metal corrected for creatinine (Tola et al., 1977)* Gylseth et al.
(1977) reported in an abstract that welders exposed to a concentration of
0.05 mg/m (measured as chromium) had a urinary chromium concentration of
-40 pg/fc, measured after work. Sjogren et al. (1983) showed a linear
relationship between worker exposure to stainless steel welding fumes and
urinary chromium levels.
Krishna et al. (1976) studied 30 chrome workers who had nasal perforation.
The atmospheric concentration of chrome ranged from 0.21 to 0.80 mg/m . Urine
samples from the workers were collected at the beginning of the day's shift and
again at the end of the day's shift. Before exposure, eight workers had a
concentration of chrome in the urine of >0.20 pg/Jl (ppm), whereas after exposure,
20 workers had such values. In an unexposed control group, there was no change
in the urine chrome values. While all the workers tested had nasal perforation,
66$ of them (20 of 30) had a urinary chrome concentration of >0.20 pg/mjl (ppm).
Mutti et al. (1979) studied 22 welders who worked with high chromium alloyed
electrodes. The concentration of the breathing zone levels of chromium, deter-
mined during the 1-month exposure monitoring period, ranged from 0.017 to
1.000 mg/m for total chromium, and from 0.002 to 0.350 mg/m as hydrosoluble
chromium. Urine samples from the workers were collected at the beginning and at
the end of the experiment. These results suggest that the urinary concentration
of chromium at the end of a working period is affected by recent exposure to the
compound. At the same airborne concentration, the greater chromium body burden
is associated with greater excretion levels. Even after a week following the
3-39
-------
last exposure, urinary chromium levels provide useful indications on body
burden.
Tandon et al. (1977) have reported urinary excretion of chromium among
electroplaters and polishers in an industrial setting. A total of 12 subjects
were examined. The range in duration of exposure to chromium in electroplating
processes ranged from 2 to 20 years (mean duration for 12 workers = 11.1 years).
Chromium levels in workers' urine samples taken before starting work ranged from
91 to 1116 pg/Jl (mean value = 326.5 pg/A (ppb)). Urinary chromium levels did not
necessarily correlate to duration of exposure, but a slight trend was indicated.
Subjective complaints included coughing, breathing difficulty, dermal itching,
depression, indigestion, body ache, and edema of the lower limbs. Mean urinary
chromium level in appropriate controls was reported to be 38.1 pg/£ (ppb) (range:
0 to 78 pg/&). Subjective complaints did not appear to correlate either with
urinary chromium levels or with duration of exposure. Urine samples collected at
the end of the work day, however, would perhaps have provided a better
correlation between exposure and complaints of illness.
3.5.3. Chromium in Human Hair. Schroeder and Nason (1969) reported a mean
chromium concentration of 0.69 + 0.063 ppm for women. Hambidge et al. (1972)
measured chromium concentrations at various distances from the hair root. They
reported that variation in the concentrations were due to past fluctuations in
chromium nutritional status. Hambidge and Rodgerson (1969) reported higher
levels of chromium in the hair of nonpregnant women (0.2 to 2.81 ppm) than in the
hair of pregnant women (0.04 to 1.11 ppm). However, a later study by Hambidge
and Droegnueller (1974) found changes in hair chromium levels due to pregnancy
not to be statistically significant. Hambidge and Rodgerson (1969) reported that
hair chromium levels in 3- to 8-month-old infants were significantly higher than
3-40
-------
in those of 2- to 3-year-old children. Chromium is obtained through breast milk
during nursing. By the second year of life, mean chromium levels in hair
approached values present in older humans.
3.6. SUMMARY
Chromium is a metallic element which, when found in nature, is a stable
mixture of four separate isotopes. Inorganic chromium compounds occur in valence
states ranging from -2 to 4-6; however, in the environment the Cr(III) and Cr(VI)
states are the most stable. Chemically, the Cr(III) state is the most stable and
important form of inorganic chromium complexes. Cr(VI) compounds comprise the
most commercially important form of chromium, and they also appear to be the most
significant chromium compounds from an environmental standpoint. Because Cr(VI)
is readily reduced in the presence of organic material, it is rarely found in
nature apart from deposition by anthropogenic sources.
Although chromite ore is not currently mined in the United States, several
chromium chemicals are domestically produced from imported ores. Sodium chro-
mate, sodium dichromate, and chromic acid are three of the more important commer-
cial chromium compounds produced in the United States. Metallurgical uses
constitute about 60% of the largest market demand for chromium. Chemical uses
are the second largest consumption sector at 21$, followed by refractory uses at
18*.
Chromium emissions are released into the air, water, and land environments
from a variety of industrial source categories including fossil fuel combustion,
cement production, incineration, cooling towers, refractory production, leather
tanneries, steel and alloy production, electroplating, and chromite ore
refining. The largest chromium emission sources to the air are coal and oil
3-11
-------
combustion, chromium chemicals production, refractory plants, and steel and
alloy plants. Principal sources of chromium in water systems include
electroplating operations, leather tanneries, and textile manufacturing
operations. Significant sources of chromium-containing solid wastes'that are
land disposed include chromite ore refining operations and chromium chemical
production plants.
Recent monitoring of the air in many urban and rural areas of the country
has shown annual average chromium concentrations to be in the range of 0.0052 to
0.1568 ng/nr; in remote parts of the world, concentrations as low as 0.005 ng/m^
have been reported. The maximum concentration determined during any one 24-hour
measurement was about 0.6839 |ig/m . The chromium concentration in U.S. waters
varies with the type of surrounding industrial sources and the type of underlying
soils. An analysis of 3834 tap waters in representative U.S. cities showed a
chromium concentration range of from 0.4 to 8 ppb. Chromium levels in soils vary
with soil origin and the degree of contamination from anthropogenic chromium
sources. Tests on domestic soils have shown chromium concentrations ranging from
1 to 1000 ppm, with the average concentration ranging from 14 to about 70 ppm.
3-42
-------
4. SAMPLING AND ANALYSIS
Chromium occurs in trace amounts and throughout the environment and is an
essential micronutrient for man. While numerous methods have been used to
measure chromium in various media, only recent methods have provided the accuracy
to measure this analytically troublesome element. Moreover, most of the more
accurate methods measure elemental chromium; few provide usable information on
the oxidation states for trace amounts.
The analysis of chromium in a certain medium usually involves three distinct
steps, namely, sampling and storage, sample pretreatment, and analysis. These
steps, and the media to which they apply, are discussed below.
4.1. SAMPLING AND STORAGE
4.1.1. Air.
4.1.1.1. AMBIENT AIR — Dusts and fumes of chromium compounds in the
ambient air are usually collected by high volume sampler at a flow rate of about
20 to 30 nr hr~ (Demuynck et al. , 1976). Typical filter media include cellu-
lose, polyethylene, polystyrene, PVC, and glass-filter. The suitability of the
filtering media primarily depends on their background impurity level, particle
retention efficiency, and tendency to become clogged. Dams et al. (1972)
evaluated different filter materials and concluded that polyethylene filters are
most suitable for the collection of chromium particles in atmospheric air.
When size fractionation is required, multistage cascade impactors are used
most often (Broekaert et al., 1982; Winchester et al., 1981; Kowakzyk et al.,
4-1
-------
1982). At a nominal sampling rate of 80 £ rnin" , chromium particles of
aerodynamic diameter in the range of 0.04 to 25 pm have been separated by this
method (Broekaert et al., 1982). The collection surfaces usually consist of
polycarbonate films, accompanied by Nucleopore backup filters (Kowakzyk et al.,
1982). However, according to Jervis et al. (1983), Nucleopore filters contain
chromium impurities. Chromium found in aerosols from the burning of coal, and
chromate from manufacturing plants, occur as small particles, less than 2 urn and
less than 0.4 urn, respectively (Gary 1982).
Although no specific information is available, it is reasonable to assume
that the filter paper protected from external contamination can be stored for an
indefinite period prior to chromium analysis.
4.1.1.2. OCCUPATIONAL AIR — Chromium in occupational air is collected in
a way similar to ambient air. A known volume of air is drawn through a poly vinyl
chloride (PVC) or cellulose filter. The sampling rate is maintained at
1.5 H min" (NIOSH, 1977). However, it has been determined by Kneebone and
Freiser (1975) that although the PVC filter can be stored for at least 10 days,
cellulose filters cannot be stored for more than several hours without some loss
of Cr(VI).
4.1.1.3. STATIONARY SOURCE — The sampling of chromium from stationary
sources, for example from the stack gasses of refuse incineration and fossil fuel
burning facilities, can be done by collecting the sample isokinetically (Block
and Dams, 1976; Greenberg et al., 1978). The sampling train typically consists
of a copper probe, a rotameter, a manometer, and a pump with a capacity of
•3 _i
20 nr hr . The diameter of the nozzle is adjusted to the stack gas velocity to
achieve isokinetic sampling conditions (Block and Dams, 1976). Depending on gas
4-2
-------
temperatures, glass-fiber, Teflon-fiber, and Whatman 41 filters have been used
(Block and Dams, 1976; Greenberg et al., 1978). However, the use of glass-fiber
filters may result in a high blank value for chromium (Greenberg et al., 1978).
For studying distribution of particle size, the isokinetic sampling is
typically done in combination with a multi-stage cascade impactor
(e.g., Andersen cascade impactor) (Greenberg et al., 1978; Block and Dams,
1976). The different stages of the cascade impactor can be used to determine the
cutoff aerodynamic diameters of the collected particles.
The same method used for storing the filters from ambient air samples can be
used for the filters collected from the sampling of stationary sources.
4.1.2. Water.
4.1.2.1. DRINKING WATER — To collect distributed water samples from
household taps, the taps are typically run to waste at their maximum flow rate
for 5 minutes to clear the lines of overnight standing water (Meranger et al.,
1981). If suspended solids are suspected in the water, filtration is typically
done on location by passing the water samples through a 0.45 pm Millipore mem-
brane filter (Meranger et al., 1979). In order to get representative samples,
some investigators (Greathouse and Craun, 1978) have collected monthly samples
of finished water for a period of 1 year instead of one grab sample. The water
samples are collected in clean linear polyethylene screw-cap collection bottles.
Concentrated nitric acid (1 mi/100 mJl water sample) is added to each bottle and
they are filled to the brim to avoid any air space. The filled and capped bottles
are transported in heavy plastic coolers containing gel-type freezer packs.
Immediately upon receipt in the laboratory, samples should be refrigerated at 4°C
(Meranger et al., 1981).
4-3
-------
Recently, a novel method for preconcentration of trace metals in water
samples has been proposed by Smits and Grieken (1981). In this method, a
cylindrical plexiglass filtration unit consisting of a Nucleopore 0.4 pm
pore-size membrane and a 2,2'-diaminodiethylamine (DEN)-cellulose filter has
been used. The Nucleopore filter was used to separate the suspended particles in
water and the DEN-cellulose filter allowed straightforward preconcentration of
trace cations by a simple filtration step. The collection efficiency for Cr(III)
by this method has been claimed to be 90-100$.
4.1.2.2. RAW AND SURFACE WATER — The sample collection, preservation,
transportation, and storage of raw and surface water samples are similar to those
for drinking water samples (Cranston and Murray, 1978; Pankow et al., 1977). A
more representative sample can be obtained by collecting several small samples
from different parts of the water body than by collecting one large sample at a
single point because of the inhomogeneity of the water body.
4.1.2.3. WASTEWATER — The sample collection, preservation, transporta-
tion, and storage of wastewater samples are similar to those for other water
samples. Grab samples (Larochelle and Johnson, 1978) are often collected but
wherever a process change is suspected, a 24-hour composite sampling is the
preferred or even necessary method.
4.1.3. Soil and Sediments. Grab samples are typically collected for the
analysis of chromium in soil and sediment. In order to evaluate the chromium
content in recent deposits, samples are collected from the upper few inches (2 to
3 inches) of sediment or soil (Pankow et al., 1977). The samples are usually
stored in polyethylene bags or bottles (Iwata et al., 1981; Pankow et al., 1977).
4-4
-------
The samples are sieved through nylon sieves (2 mm) to remove gravel and leaves
(Iwata et al., 1981). The collected samples should be stored at 4°C or
refrigerated during transportation and storage to minimize bacterial action
(U.S. EPA, 1979).
4.1.4. Food. No specific information for sample collection and storage of food
samples could be found in the literature. It is reasonable to assume that grab
samples collected in polyethylene bags or bottles and refrigerated during trans-
portation and storage should be an acceptable procedure for sample collection and
storage.
4.1.5. Biological Samples. Blood samples are typically drawn from donors with
either vacutainers or disposable syringes and aluminum needles. Heparin is used
as the anticoagulant. Plasma samples can be obtained by centrifuging freshly
drawn heparinized whole blood. They can be transported at 4°C in polyethylene
tubes and frozen to -20°C during storage. Urine should be collected from donors
in polyethylene bottles, acidified, and stored at 4°C (Davidson and Secrest,
1972). Other solid tissue samples can be collected in polyethylene bags and
transported at 4°C. The tissue samples can be stored at -20°C. Lin (1982) noted
sources of contamination can arise from steel needles used for blood drawing and
from borosilicate glass containers which contain 1-10 pg/Cr/g glass. Moreover,
in urine determinations utilizing atomic absorption, interferences can arise
from absorption by NaCl or organic materials in the urine (Ping et al., 1983).
4.2. SAMPLE PRETREATMENT
Sample pretreatment is often required as a technique to release the metal
from the sample matrix, and for concentration and separation from potential
4-5
-------
interferences. A few of the typical pretreatment methods have been discussed
below. Generally, the pretreatment of samples is dictated by the subsequent
method of analysis.
4.2.1. Wet and Dry Ashing. This pretreatment method is often used for chromium
analysis in air particulate matter, biological samples, foods, soil, and
sediment samples in combination with atomic absorption, atomic emission,
spectrophotometric, and neutron activation analysis. In wet digestion or
extraction methods, the sample is digested with an acid or a mixture of acids
depending on the sample matrix. The composition and efficiency of commonly used
methods are listed in Table 4-1. Acid mixtures which have been used include both
HNO and H2S04 (NIOSH, 1977), HNO /HP (Pankow et al., 1977), HjSO^/H^
(Kumpulainen et al., 1979), HNCyH^ (Abu-Samra et al., 1975), HCIO^/H^
(Davidson and Secrest, 1972), HNO^HCIO^ (Gallorini et al., 1976),
HNO-XH^Ojj/HClOjj (Kuennen et al., 1982; Feldman et al., 1967) and HNC^/H^O^
(Bryson and Goodall, 1981). The two commonly used dry ashing procedures are
graphite furnace ashing (Meranger et al., 1979; Slavin, 1981) and low-
temperature oxygen plasma ashing (Kumpulainen et al., 1979). Some investigators
also have used a combination of dry and wet ashing for the pretreatment of
samples (Kumpulainen et al., 1979). Whichever procedure is used, a particular
ashing method is always optimized to minimize matrix interference and maximize
the chromium yield from the sample.
4.2.2. Precipitation. The direct precipitation of chromium from aqueous solu-
tion by such reagents as hydroxyquinoline and tannic acid is generally not
suitable for environmental samples containing low chromium concentration.
However, the method of co-precipitation of chromium has been successfully used in
4-6
-------
TABLE 4-1
Composition and Efficiency of AA Extractant Solutions*
Efficiency13
58.7
66.6
62.7
67.9
65.8
60.9
93.7
76.1
Nitric
50 (2m)
50
25
24
30
15
Type and Volume (mi) of Acid Mixtures
Hydrochloric Sulfuric Perchloric
d
50 (2 m)d
50
25
24 4
10 10
6 30
References
Day et al. (1975)
Day et al. (1975)
Agemian and Chau (1976a)
Agemian and Chau (1976a)
Chiu and Hilborn (1979)
Chiu and Hilborn (1979)
Chiu and Hilborn (1979)
Agemian and Chau (1976b)
Adapted from Tinsley et al. (1983)
pg of chromium extracted/g of dust sample
'Acids are concentrated
Molarity of solution
-------
recent years. In this method, Cr(III) is co-precipitated with Fe(III) as
hydroxide at a pH of 8-8.5 (Cranston and Murray, 1978; Pik et al., 1981). The
co-precipitation of Cr(VI) has been accomplished by reducing Cr(VI) to Cr(III) by
the addition of Fe(II) and subsequent co-precipitation of hydroxides of Cr(III)
with Fe(II) at pH 8 (Cranston and Murray, 1978), or by co-precipitating Cr(VI) by
Co(II) and ammonium pyrrolidinedithiocarbamate (APDC) addition (Pik et al.,
1981). The co-precipitated chromium is filtered from the solution and analyzed
by x-ray fluorescence or by flameless atomic absorption spectrometry.
1.2.3. Solvent Extraction. In this procedure, Cr(VI) is complexed with APDC at
a pH of 1.8-3.0 and extracted with an organic solvent, typically methyl isobutyl
ketone (MIBK). If Cr(III) is to be extracted, it must first be oxidized to
Cr(VI) with reagents such as silver nitrate and potassium peroxydisulfate or with
potassium permanganate and sodium azide (Towill et al., 1978). In the gas
chromatographic procedure, chromium is chelated with 1,1,l-trifluoro-2,4-
pentanedione or trifluoracetylacetone (HFTA) and the chelate is extracted into
benzene. This method has been used by a number of investigators (Lovett and Lee,
1976; Gosink, 1976).
M.2.4. Chromatographic Method. A number of chromatographic materials including
alumina, cation-exchange resin, anion-exchange resin, and chelating ion-exchange
resin have been used for the cleanup of impurities from chromium samples. In the
anion-exchange procedure, the solution containing Cr(VI) is allowed to pass
through the resin bed at an optimum pH. The Cr(VI) retained on the resin bed is
subsequently extracted with a suitable eluent. The major drawback of this
procedure is that the recovery of Cr(VI) is often poor. . The problem, however,
has been alleviated by using an ascending flow technique or by in situ reduction
14-8
-------
of Cr(VI) by Fe(II) (Pankow and Janauer, 1974). Marino and Ingle (1981) used
medium strength anion-exchange resin to obtain a satisfactory recovery of Cr(VI)
from the resin bed.
In the cation exchange procedure, the resin is used to retain the' cationic
impurities while anionic Cr(VI) passed through the resin bed. This procedure has
been used by Kneebone and Freiser (1975) for the analysis of Cr in occupational
samples.
The chelatory ion-exchange procedure was employed by Leyden et al. (1972)
to absorb quantitatively Cr(III) from a buffered solution. By this procedure,
Cr(VI) was detected by reducing it to Cr(III) with the addition of sodium bisul-
fite.
The alumina procedure has been used for the separation of both Cr(III) and
Cr(VI). Larochelle and Johnson (1978) adsorbed Cr(VI) on alumina column and used
HC1 for the subsequent elution of Cr(VI). Wolf et al. (1972), on the other hand,
used an alumina column for the separation of Cr(III). In this procedure, Cr(III)
is precipitated with other cations by the addition of excess 8-hydroxyquinoline.
The dried precipitate is dissolved in chloroform, diluted with an equal volume of
benzene, and passed through an activated alumina column. Chromium was eluted
with a mixture of chloroform and benzene.
4.3. METHODS OF ANALYSIS
Chromium can be determined by a variety of analytical methods. A few
analytical methods used for the determination of Cr are given in Table 4-2. It
should be emphasized that the detection limit and the percent-relative standard
deviation (% coefficient of variation) values given in Table 4-2 should be taken
as values representative of the specific pretreatment techniques and
4-9
-------
TABLE 4-2
Analytical Methods for the Determination of Chromium
Method Type of Sample
Graphite furnace AA blood, urine, and other
biologic samples9
natural waters
raw, treated, and dis-
tributed water0
-tr
1
_*
O
Flame AA raw, treated, and dis-
tributed water
natural water
blood, urine, and other
biologic samples
Spectrophotometric tissue samples
Preconcentration
none
co-precipitation
with Fe(OH)-
none
APDC-MIBK
extraction
APDC-MIBK
extraction
anlon exchange
MIBK extraction
APDC-MIBK
extraction
Selectivity
total Cr
Cr(III); Cr(VI)
can be reduced
to Cr(III) by
Pe(II)
total Cr
Cr(VI); Cr(III)
can be oxidized
to Cr(VI)
Cr(VI); Cr(III)
can be oxidized
to Cr(VI)
Cr(VI)
Cr(VI)
Cr(VI); Cr(III)
can be oxidized
to Cr(VI)
Detection
Limit
0.1 pg/8.
0.001 (jg/fc
2 PS/*
0.6 pg/i
0.05 pg/4
0.1 pg/l
10 pg/t,
1.2 pg/l
% cv
(at sample
concentration)
5.2%
(10 pg/l)
5%
(0.14 pg/l)
15> (2 pg/l)
20% (2 pg/l)
5% (3 pg/l)
20| (0.1 pg/l)
15* (25 pg/l)
0.4f (800 pg/l)
Interference
Matrix interference
can be avoided by
«et ashing.
none reported
none reported
none reported
none reported
none reported
none reported
Fe, Ni and PO.
may interfere
3-
Catalytic method
air participates8
Neutron activation air particulates
freshwater^
.none
none
none
Cr(VI)
total Cr
total Cr
1 ng
30 ng/n
0.12 pg/l
6% (<100 (ig/t) Pb, Cu, Cr(III),
Fe(III), and V(V)
nay interfere
151
none reported
(1.3 ng/m)
21 % (1.4 pg/l) none reported
-------
TABLE H-2 (cont.)
Analytical Methods for the Determination of Chromium
Method
Gas chroma tography
(electron capture
detection)
Gas chroma tography
(AAS detection)
Gas chromatography
(MS detection)
Liquid chromatography
Type of Sample
natural waters
blood, serum, orchard
leaves
blood"
natural water"
Preconcentration
HTFA-benzene
extraction
HTFA-benzene
extraction
HTFA-benzene
extraction
1 I sample concen-
Selectivity
Cr(III) and Cr(VI)
Cr(III)
Cr(III)
Cr(VI)
Detection
Limit
0.1 pg/t
<1 ng
0.5 pg
0.8 pg/t
% CV
(at sample
concentration)
2.6| (1.9 pg/t)
<6> (1 ng)
9* (10 ng/g)
<2J (90 pg/t)
Interference
none reported
none reported
none reported
so'2, pofl3"
(coulometric detection)
X-ray fluorescence
(energy dispersive)
X-ray fluorescence
(energy dispersive)
Differential pulse
polarography
Emission spectroscopy
(inductively coupled
plasma source)
Mass spectrometry
dried solution deposit
surface water and drinking
water'
natural waters
natural waters
variety of samples
trated to 10 mt
none
chelating ion-
exchange membrane
none
none
none
total Cr
Cr(III)
Cr(VI)
total Cr
total Cr
1.5 pg/gp
0.8 pg/t
10 pg/l
1.8-6
0.05-1 pg
(1 pg/cm )
10-15*
(1 jig/i)
may inerfere
absorption by
matrix and dif-
ference in
particle size
may cause error
excess alkali and
alkaline earth
metals may inter-
fere
34) (61 pg/£) excess Cu(II) and
Fe(III) may inter-
fere
=5J none reported
20) (photo- Any species having
graphic) the same m/e ratio
3J (electrical) as Cr nuclide may
0.5? (isotope interfere
dilution)
-------
TABLE 4-2 (cont.)
Analytical Methods for the Determination of Chromium
Method
Chemiluminescence
Chemiluminescence
(lophlne)
PIXE (Particle
Induced X-ray
Emission)
% CV
Detection (at sample
Type of Sample Preconcentratlon Selectivity Limit concentration) Interference
natural waters and orchard none Cr(III) 0.02 pg/l 20* (at 2.3 Fe(III), Fe(II),
leaves" PPIB) Co(II), S032-.
and HO," may
Interfere
natural waters' ion exchange resin Cr(VI) 0.015 pg/fc Fe(III), Fe(II),
Co(II), Cr(IIJ),
MnOj.~, and
other cations and
anions may interfer
ambient airy; none total Cr <1 ng <35J inadequate sample
biological samples2 preparation; Pe may
interfere
aDavidson and Secrest, 1972
Cranston and Murray, 1978
°Meranger et al., 1981 q
d
VanGrieken et al.,
e
Pankow and Janauer, 1971
Feldman et al., 1967
Bryson and Goodall, 1981
8Kneebone and Freiser, 1975
hDemuynck et al., 1976
1Bhagat et al., 1971
JSalbu et al., 1975
kLovett and Lee, 1976
^olf, 1976
"Wolf et al., 1972
nLarochelle and Johnson, 1978
Camp et al., 1975
pKuhn, 1973
1977
rCrosnun and Mueller, 1975
sQuinby-Hunt, 1978
Boumanns and deBoer, 1972
uAhearn, 1972
vElser, 1976
*Hoyt and Ingle, 1976
xMarino and Ingle, 1981
yMetternich et al., 1981
zBartsoh et al., 1982
-------
instrumental methods used rather than definite data. For example, both the
detection limit and percent CV values obtained by the same instrumental method
may vary several-fold depending on the extent of preconcentration of the sample
>
and the method used for eliminating interferences. These values may also vary
considerably from laboratory to laboratory and even within the same laboratory.
Part of the errors in measurement occur from the absence of low level standard
reference materials. Several methods for analyzing chromium are described
briefly below.
1.3.1. Atomic Absorption Spectrometry (flame). This method has been used by
various investigators for the determination of Cr in surface water, sewage
effluent, and biological samples. The determination of Cr by the air-acetylene
flame is prone to interference by other elements (Thompson and Wagstaff, 1979).
This problem can be avoided by using nitrous oxide-acetylene flame but this
results in a decrease in the detection limit (Thompson and Wagstaff, 1979).
Therefore, most investigators have used air-acetylene flame for the determina-
tion of Cr. However, for samples with low Cr concentration, pretreatment of the
samples providing preconcentration of Cr and the reduction of the interfering
effects from other ions are required. Thompson and Wagstaff (1979) used an
evaporative technique on a hot plate to concentrate the sample 5-fold. A 2%
ammonium perchlorate solution (W/V) was used to suppress the interelement inter-
ference effects. Better methods for pretreatment of samples include
ion-exchange separation (Pankow et al., 1977; Pankow and Janauer, 1971), APDC-
MIBK extraction (Gilbert and Clay, 1973) and MIBK extraction (Feldman et al.f
1967).- The flame AAS technique is not a preferable method for samples with very
low Cr concentration since this method has a much higher detection limit than
flameless AAS (Slavin, 1981).
1-13
-------
1.3.2. Atomic Absorption Spectrometry (flameless). In the flameless AAS
method, the sample is atomized directly in a graphite furnace, carbon rod, or
tantalum filament instead of a flame. It is one of the most attractive methods
for the analysis of solid and liquid biological samples since the method does not
generally require sample preparation. Despite the high sensitivity of the flame
less AAS method, it may suffer from certain disadvantages. For example, matrix
interference, background or nonspecific absorption effects, volatilization of
some Cr during dry ashing, and adsorption of Cr on the walls of crucibles during
dry ashing can cause error in analysis. The matrix interference effect in the
case of water samples can be overcome by chelation and solvent extraction
(Meranger et al., 1979, 1981) or by the co-precipitation technique (Cranston and
Murray, 1978), and in the case of biological samples, by wet disgestion (Davidson
and Secrest, 1972) prior to introduction of the sample into the graphite furnace.
The Zeeman background correction systems for non-specific absorbance are those
most recommended for the analysis of Cr (see handbooks on EAAS by Perkin-Elmer
and Hitachi). Other background correction systems include the dueterium arc
corrector (Cranston and Murray, 1978), and, by continuum source, echelle,
wavelength-modulated, atomic absorption spectrophotometer (CEWM-AA) (Kumpulanen
et al., 1979; Veillon et al., 1982a,b). The volatilization of Cr during dry
ashing at temperatures of 700°C or higher, particularly during the analysis of
complex matrices such as food, may be substantial (Kumpulainen et al., 1979).
Low temperature ashing, for example, with oxygen plasma at 150°C, would eliminate
this problem and the problem of Cr adsorption on crucible walls. However, the
low temperature ashing may not be suitable for some biological samples (bovine
liver), and some biological materials such as brewer's yeast may contain acid
insoluble material that can strongly adsorb chromium (Kumpulainen et al., 1979).
However, these problems have been overcome by utilizing dry ashing at 500°C with
-------
sulfuric acid and hydrogen peroxide as ashing acids (Kumpulainen et al., 1979).
The problems of sorption and volatilization can also be eliminated by the use of
"platform" technique described by Slavin (1981).
Versieck et al. (1978) reported that chromium levels in serum from normal
subjects was 0.16 pg/fc when determined by carefully conducted neutron activation
analysis. In a survey of previous reports of chromium levels determined by
flameless AAS, the range of values was 0.73 to 150 pg/fc, and the levels tended to
become lower as improvements in detection limits occurred. These results
prompted an intensive study of possible sources of artifacts in the flameless AAS
analysis of biological fluids by Kayne et al. (1978) and Guthrie et al. (1978).
Guthrie et al. (1978), in a study of the effects of purge gas, char temperature,
sample volume, and graphite tubes on the determination of low ppb levels of
chromium, demonstrated that all these parameters affected the background non-
atomic absorption, and that there was a direct correlation between background
absorption and the apparent chromium content of the sample. It was suggested
that the background correction was inadequate as the result of the low intensity
of the D- lamp at the near UV wavelength used in chromium analysis. Kayne et al.
(1978) came to a similar conclusion, and modified the background correction by
replacing the usual D_ lamp with a tungsten halogen lamp. This modification
provided adequate background correction, and serum chromium levels determined
for 8 normal male subjects averaged 0.14 pg/H, while most measurements from 66
randomly chosen urine samples were -0.9 pg/fc. Guthrie et al. (1978), Kayne et
al. (1979), and Anderson (1981) noted that earlier reports of chromium levels in
urine and serum should be considered artificially high (by about an order of
magnitude) in light of these findings.
-------
1.3.3. Emission Spectroscopy. In emission spectroscopy, prepared samples are
excited with a flame, arc, spark, or plasma and the resulting light is dispersed
with a monochromator. The characteristic emission lines of the excited elements
are recorded electronically or on a photographic plate. Because of its better
sensitivity, inductively compiled plasma atomic emission spectrometry (ICP-AES)
has been more extensively used in recent years than the other modes of excitation
(Towill et al., 1978; Slavin, 1981). Since the sample is generally introduced
into the plasma source by pneumatic nebulization, all solid and biological
samples require wet digestion prior to analysis. The wet digestion also tends to
minimize the matrix effects. A HF/HNCL digestion at 170°C has been employed for
the simultaneous multi-element analysis of air particulate matters (Broekaert
et al., 1982). Similarly, wet digestion by direct chelation with
1,1,1-trifluoro-2,4-pentaedione (also known as trifluoroacelytacetone) (HTFA) or
hexafluoroacetylacetone (HHFA) has been used for the cleanup and volatilization
of Cr in biological samples (Black and Browner, 1981) prior to ICP-AES analysis.
Recently, a direct atomization technique for raw agricultural crop samples has
been presented (Kuennen et al., 1982). The technique employs a 30-minute pres-
sure dissolution of sample composite with 6M HC1 at 80°C in linear polyethylene
bottles. The dissolved samples after proper filtration by Teflon filters can be
aspirated directly into the plasma source. Combined with real sample matrix
calibration technique, it has been suggested that this method provide comparable
recovery and precision as obtained by the more time-consuming conventional wet
ashing methods.
H.3.4.. Neutron Activation Analysis. Neutron activation analysis is one of the
most sensitive modern analytical techniques for the determination of trace
f
elements. Neutron activation analyses are applicable to many kinds of environ-
4-16
-------
mental samples including air participates, dusts, soils, fresh and marine water,
sediments, biological liquids and solids, and foods. The samples are often
irradiated without prior chemical treatment. A detection limit of 0.12 pg/fc in
freshwater (Salbu et al., 1975) and 0.2 pg/g in biologic materials (Spyrou
et al., 1974) has been reported for samples analyzed without chemical proces-
sing. Lower detection limits generally can be achieved if the samples are
chemically processed to separate and concentrate chromium. For example, in the
chemical processing of samples typically by ion-exchange separation, detection
limits of 0.1 ng/H for river water, 3 ng/8, for seawater, in the ppb range for
biological samples and the ng/nr range for ambient air have been reported
(McClendon, 1974; Robertson and Carpenter, 1974; Kowalczyk et al., 1982; Lin,
1983).
One distinct advantage of this method is the reduced problems arising from
reagent contamination. Even if chemical processing is required, post-irradia-
tion contamination is of no consequence in introducing error to the final result.
In addition to the problem of acquiring a neutron source, this method has another
21
disadvantage. Due to the intense x-ray or bremsstrahlung activity from Na,
oQ ilO C f\ ^2
Cl, K, Mn, and p in many samples, the irradiated samples usually must be
t>
cooled for several weeks before measuring Cr concentration. However, chemically
separating the offending nucleides can reduce the cooling period to about
24 hours (McClendon, 1974).
4.3.5. X-ray Fluorescence. The x-ray fluorescence method is commonly applied
to solid samples. In this technique, the sample is bombarded with high energy
photons, for example, low energy x-ray or gamma photons or with particles such as
protons. The intensity of the characteristic emitted x-ray is currently measured
by two techniques, namely, energy-dispersive analysis and wavelength-dispersive
4-17
-------
analysis. The resolution of wavelength-dispersive analysis is much better than
energy-dispersive analysis which often permits the former method to determine
the oxidation states of an element (Quinby-Hunt, 1978).
This technique is useful for trace analysis in solid samples, namely, air
particulates (collected on filters), sediments, biological specimens, and
filtered suspended solids from aqueous media. However, for aqueous samples,
sample preparation leading to the deposition of the element on a filtering medium
is required. This deposition can be done by a variety of methods. The obvious
and simple method is the evaporative drying of the sample on a Mylar film (Tanaka
et al., 1981). Other methods of sample deposition include retention on chelating
filters (Smits and Van Grieken, 1981), vapor filtration technique (Greathouse
and Craun, 1978) and co-precipitation (Pik et al., 1981). The last three methods
not only allow deposition of Cr on a filtering media, but also permit separation
and concentration of Cr from the aqueous phase during the deposition step.
Sample preparation in the case of solid samples is an important requirement
for this method. In order to avoid absorption and scattering by sample matrix,
the size and shape of the particles on the films should be controlled by pressing
the deposits into thin wafers (Towill et al., 1978).
4.3.6. Colorimetric. The diphenylcarbazide colorimetric method is the most
commonly used for monitoring chromium compounds. Either through the oxidation of
Cr(III) or direct reaction with Cr(VI), this method was proposed by NIOSH (1976)
as a "Draft—subject to review" procedure for measuring Cr(VI) in occupational
atmospheres. While it is sensitive to 10 and 50 pg samples, it has poor
sensitivity for samples of 0.5 and 1 pg. (Bhargava et al., 1983). The method may
not be suitable for use with welding fumes. Instead, a carbonate leaching method
has been described for the determination of Cr in welding fumes and other complex
4-18
-------
matrices (Thomsen and Stern, 1979). Bryson and Goodall (1981) have described a
modified diphenylcarbazide spectrophometric method for the determination of Cr
in biological tissues.
4.3-7. Gas Chromatography.
The gas chromatographic method is suitable for a variety of environmental
samples. The analysis of water samples by this method does not generally require
pretreatment before complexation (Lovett and Lee, 1976). Samples that are solid,
namely, particulate matter, and biological samples are first digested to get Cr
in solution. The Cr is then quantitatively chelated with 1,1,1-trifluoro-2,4-
pentanedione (HFTA) or hexafluoroacetylacetone (HHFA) to form a thermally
stable, volatile Cr(III) complex. To determine Cr(VI), it should be reduced to
Cr(III) by a reducing agent, such as sodium sulfite. The Cr(III)-HFTA or
Cr(III)-HHFA complex is extracted into an organic solvent, usually benzene or
hexane, and an aliquot is injected into the GC column. Recently a direct
chelation (without pretreatment) method has been described for biological
samples (Black and Browner, 1981).
The detection of Cr(III) complex can be accomplished by a variety of detec-
tors. The sensitivity of this method is dependent on the detector used. A
number of detectors including electron capture (Lovett and Lee, 1976), atomic
absorption (Wolf, 1976), mass spectrometric (Wolf et al.f 1972), and ICP-AES
(Black and Browner, 1981) have been used.
The gas chromatography with any of the detection methods described has
excellent sensitivity for Cr determination, but the mass spectrometric method is
extraordinarily sensitive and specific.
4.3.8. Chemiluminescence. Luminal (5-amino-2,3-dihydrophthalazine-1,4-dione)
and lophine (2,4,5-triphenylimidazole) emit light when oxidized by hydrogen
4-19
-------
peroxide. The first oxidation in basic solution is catalyzed by Cr(III) and the
second by Cr(VI). The design of different chemiluminescence instruments for the
determination of Cr vary primarily in the technique for mixing the sample and
reagents. The three types of sample modules which are commonly used are discrete
sampling system, flow system, and centrifugal analyzer (Hoyt and Ingle, 1976).
All these sample modules have been used for the analysis of Cr (Towill et al.,
1978).
The chemiluminescence method has been used primarily for the analysis of Cr
in water and in biological samples (Marino and Ingle, 1981; Hoyt and Ingle,
1976). The method is fast, economical, and has a high sensitivity. The sensi-
tivity can be further increased by preconcentration of the samples (Marino and
Ingle, 1981). However, this method has not been as widely used as some of the
other methods.
4.3.9. Polarography. Two recent variations of the method, namely, single-sweep
polarography and differential pulse polarography, have both been used for the
detection of Cr in natural waters (Whitnack, 1975; Crosmun and Mueller, 1975).
The polarographic method has a comparatively lower sensitivity than other
methods and is not currently popular for Cr analysis.
4.3.10. Mass Spectrometry. The spark-source mass spectrometric method is
applicable to virtually any matrix, but the results are only semiquantitative
(Towill et al., 1978). The precision and accuracy of this method can be greatly
improved by using the isotope dilution technique. This method has been used to
establish the Cr content in NBS bovine liver (Dunstan and Garner, 1977). The
mass spectrometric method is rarely used for the routine determination of
chromium.
4-20
-------
H.3.11. Catalytic Method. This method has been used for the determination of
Cr(VI) in occupational samples (Kneebone and Freiser, 1975). In this method, the
microdetermination of Cr(VI) is done by means of Cr catalyzed oxidation of
o-dianisidine by hydrogen peroxide. The reaction rate is monitored1 spectro-
photometrically. This method has limited application (occupational samples)
because of interferences by various cations, and it is rarely used for the
determination of Cr.
H.3.12. Liquid Chromatography. This method in combination with a coulometric
detection has been used for the determination of Cr in water samples (Larochelle
and Johnson, 1978). In this method Cr(VI) is separated by an alumina column and
the separated solution is passed through a flow-through coulometric detector.
Although the method has been purported to have good sensitivity and precision, it
has found little application in the analysis of chromium.
H.3.13. Particle Induced X-ray Emission. Particle Induced X-ray Emission (PIXE)
has been used to analyze elemental constituents of particulate matter. PIXE was
first introduced by Johansson and Johansson (1970) and has found wide use in
elemental analysis of atmospheric aerosols and considerably less use for
biological samples. In this method, atoms of interest are bombarded by protons.
The incident particles will interact with the atoms to eject electrons, thereby
ionizing the atoms. The characteristic X-rays emitted are proportional to the
2
atomic number (Z ). PIXE analysis of elemental Cr is sensitive to the nanograra
level; it is a rapid, multielemental, nondestructive technique which requires no
chemical breakdown of the sample. However, elaborate sample preparation and
highly skilled operators are required for successful analyses (Metternich et
al., 1982; Bartsch et al., 1982).
4-21
-------
J4.M. CONSIDERATIONS IN ANALYSIS
The determination of Cr in samples containing trace amounts of the compound
requires special precautionary measures from the initial sample collection step
to the final analytical manipulations of the samples (Zief and Mitchell, 1976).
For example, contamination of samples during collection should be avoided by the
use of Cr-free equipment. This is particularly true for the collection of
biological samples with stainless steel needles, scalpels, trays and utensils
which may contain 18? chromium. Similarly, sample containers should be free from
the possibility of sample contamination. Polyethylene bottles and bags are
particularly suitable as sample containers. Even with polyethylene bottles,
adsorption of Cr on the surface of the container from water samples may be a
serious problem. Therefore, acidification of aqueous samples prior to storage is
necessary to avoid adsorption losses.
The possibility of sample contamination and losses during analytical pre-
treatment of the samples should be avoided. Care should be taken that only
reagents of the highest purity are used. Even so, the quantity added should be
limited to a minimum to avoid unnecessary buildup of contaminants. The use of
Cr-containing grinding or homogenizing equipment can introduce Cr-contamination
into the samples. Grinding such samples with an agate mortar and pestle is a
better procedure. The analysis of Cr, especially in biological samples, is
complicated by extreme matrix effects, possible volatility of some Cr complexes,
and the inherent property of Cr-complexes to bind non-specifically to reaction
vessels, graphite tubes, or other equipment. The methods used to minimize these
problems have been discussed in section 4.3.2.
The problem of developing accurate data from Cr analysis, particularly in
food and biological samples, is amply illustrated by the large variations in the
4-22
-------
interlaboratory comparison data (Towill et al., 1978). The problem is further
aggravated by the non-availability of Standard Reference Materials (SRM). Only
recently has the National Bureau of Standards (NBS) issued materials certified
for Cr, such as brewer's yeast (SRM-1569), bovine liver (SRM-1577), orchard
leaves (SRM-157D, spinach leaves (SRM-1570), pine needles (SRM-1575), and
tomato leaves (SRM-1573). In view of the absence of suitable comparison with
SRMs, the older data should be interpreted with skepticism.
Another problem in dealing with the analytical methods is their ability to
distinguish between Cr(III) and Cr(VI). This is particularly important since
Cr(VI) has been associated with health hazards, while Cr(III) is of substantially
less concern and is in fact necessary (at very low but as yet undetermined
levels) for the maintenance of a normal glucose tolerance factor (Marino and
Ingle, 1981). The problem for the determination of the two oxidation states of
Cr is not critical in foods, sediments, soils, and biological samples, because Cr
is genrally present in the Cr(III) state in these samples. However, Cr may be
present in both oxidation states in ambient and occupational air, and in water
samples. It should also be recognized that Cr(VI) may be present in both water
soluble and water insoluble form in these samples. It is necessary to
distinguish these two forms of Cr(VI), as they may have different genotoxic
properties (Thomsen and Stern, 1979). Some of the methods that distinguish
between Cr(VI) and Cr(III) have already been discussed. The analytical method
for the determination of soluble and insoluble forms of Cr(VI) in occupational
air samples has been described by Thomsen and Stern (1979).
1-23
-------
-------
5. CHROMIUM METABOLISM IN MAN AND ANIMALS
5.1. ROUTES OF CHROMIUM ABSORPTION
5.1.1. Chromium Absorption and Deposition by Inhalation. An important route of
exposure to chromium compounds is through inhalation of chromium containing
aerosols. In general, during inhalation (and exhalation) a portion of the
inhaled aerosol may be deposited by contact with airway surfaces or be
transferred to unexhaled air. The remainder is exhaled. The portion transferred
to unexhaled air may be either deposited by contact with airway surfaces or later
exhaled. These phenomena are complicated by interactions that may occur between
the particles, other gases and the water vapor present in the airways (U.S. EPA,
1982). As such, the deposition and retention patterns for chromium aerosols
depend on the size and solubility of the particular chromium compound.
Chromium aerosols occur mainly in the small particle-size fractions. The
oxidation state of the Cr compound depends on the emission process as well as the
atmospheric interactions. In metallurgy, most of the chromium released is
elemental or trivalent (NAS, 1971*). In the chemical industry, chromate dust from
the processing of chromite ore averaged 0.32 to 0.37 pm in diameter (Gafafer,
1953). Fly ash from coal-fired electric power plants contained Cr in the 1-2 pm
size fractions (Davidson et al., 1974). Combustion processes under alkaline
conditions are conducive to the formation of Cr(VI). The distribution of
chromium in the human lung was investigated by Bartsch et al. (1982), who
examined 35 lung pairs obtained during autopsy from randomly selected patients in
Liege, Belgium, an area containing many steel plants and foundries. Using PIXE
5-1
-------
analysis (Particle Induced X-ray Emission), they found that Cr concentrations in
the lung averaged 2.85 pg/g (ppm) dry matter, with the largest amounts occurring
in the lymph nodes. The remaining chromium was distributed over a gradient
toward the apexes of the lungs, suggesting to the authors a relation to the
normal distribution of inspired air and contaminants during normal breathing.
Langard et al. (1978) exposed rats to zinc chromate(VI) dust at a concen-
tration of 7.35 mg/m . Before exposure, the mean blood chromium concentration
was 0.007 pg/m£ (ppm). Following exposure, values were 0.024, 0.22, and
0.31 pg/mfc (ppm) at 100, 250, and 350 minutes, respectively.
The next portion of the study involved four consecutive daily exposures of
6 hours duration to zinc chromate at a level of 7.35 mg/m . Blood concentrations
appeared to peak after the second exposure session. Mean blood chromium values
from samples taken immediately after each exposure session were: 0.30, 0.56,
0.46, and 0.34 pg/m& (ppm) for days 1 through 4. The mean blood chromium level
following 2 months of exposure for 6.5 hours/day, 5 days/week was quite similar,
0.495 pg/mB, (ppm). There were no significant differences in absorption, as
reflected by blood chromium level, between the sexes or between day and night
exposure regimens. Kamiya et al. (1981) exposed rats to chromite dust for 28
days to determine distribution and retention of total Cr from inhalation
exposure. While data are limited, qualitative conclusions can be drawn. Whole
blood total Cr levels measured at various times throughout the exposures remained
about the same from days 5 through 28. The highest Cr concentrations were found
in the kidney followed in decreasing order by the lung, spleen, liver and blood.
Concentrations in the liver were similar to those in the blood, changing little
from days 5 to 28. Levels in the lung, kidney and spleen were cumulative.
Baetjer et al. (1959a) exposed guinea pigs via intratracheal administration
to 200 pg chromium as sodium chromate(Vl), potassium dichromate(VI), or chromic
5-2
-------
chloride(III), all of which are water soluble salts. For the Cr(VI) compounds 10
minutes post-instillation, 15$ of the dose remained in the lungs, 20% was found
in the blood, and 5% was distributed among various soft tissues. Clearance of
chromium upward from the trachea and subsequent swallowing is assumed to account
for the majority of the dose not found in the blood and tissues. During the first
24 hours, =13$ of the dose was excreted, 11$ remained in the lungs, and 16$ was
found in the blood and other tissues. After 140 days, chromium had been essen-
tially cleared from all tissues except lung and spleen. For chromic
chloride(III) 10 minutes post-instillation, 69$ of the administered dose was
retained in the lungs, with 4$ in the blood and other tissues combined. Twenty-
four hours post-exposure, the lungs retained 45$ of the administered dose; lung
retention after 30 and 60 days was 30 and 12$, respectively.
Of the water soluble salts studied, Cr(III) is absorbed much more slowly
from the lungs than Cr(VI), possibly as a result of binding to extracellular
macromolecules. In addition, analyses of lungs from experimental animals and
human autopsy specimens indicate that water soluble salts undergo conversion to
very insoluble moieties with long residence times in lung tissue (Baetjer et al.,
1959a).
Wada et al. (1983) exposed male SD strain rats to CrCl3 at an atmospheric
concentration of 14.1 mg/m (Cr), and observed that the chromium was associated
with both high and low molecular weight proteins. The chromium which remained in
the lungs was associated with the high molecular weight fraction and this
fraction slowly decreased with time following exposure. The level of chromium
associated with the low molecular weight fraction remained constant for the 5
observation days following treatment; however, chromium associated with this
fraction accumulated with time in the liver. The authors suggested that the low
molecular weight protein was involved in the absorption and transport of chromium
5-3
-------
I ;
following inhalation. Low molecular proteinuria was also noted in rabbits
MM«iving subcutaneous injection of 1.77 mg Cr(VI)/kg body weight (Nomiyama et
ml., 1982).
Visek et al. (1953) also studied the fate of 51CrCl, following intrat-
raoheal instillation. Seven days post-exposure, 55% of the chromium was excreted
IB the feces and 7% in the urine. These data agree with Baetjer et al. (1959a),
in that a large portion of the administered dose appears to have been cleared to
tfee gastrointestinal tract. Tissue concentrations in this study indicated that
«55( of the administered dose was absorbed from the lungs.
No data are available which could be used to accurately estimate total
pulmonary absorption following inhalation exposure. The contribution of gastro-
intestinal absorption to body burden following inhalation or intratracheal expo-
sure is not clear.
5.1.2. Gastrointestinal Absorption of Chromium. Gastrointestinal absorption of
chromium also appears to be dependent on valence state. Donaldson and Barreras
(1966) examined absorption of CrCl-(III) and Na CrOu(VI) in rats and in
3 fc *T
humans and conducted a series of in vitro evaluations to clarify factors
affecting absorption.
Based on fecal excretion, mean absorption of orally administered ^1Cr
compounds in human patients was =0.4$ for CrCl- and =10.6$ for NapCrOj..
Approximately 0.5 and 2.1$ of the doses of CrCl-, respectively, were excreted in
the urine after 24 hours. When CrCl, was administered intraduodenally,
absorption of CrCl- was not appreciably increased; however, intraduodenal
administration of NaCrCv resulted in an estimated 50$ absorption.
In rats, »2$ of the intragastric dose of both CrCl, and Na CrOj. appeared to
be absorbed based on fecal excretion (i.e., it was assumed that any radioactivity
5-4
-------
not recovered in the feces represented the absorbed portion of the dose, although
it should be noted that the percision of these estimates may be questioned).
Jejunal administration increased apparent CrClo absorption to B%, while Na2CrOjj
was increased to =25$. Similar low levels of absorption of less than 1$ were
estimated by determining whole body counts 2 days after administration of radio-
labeled CrCl_ or Na^rOj, (Sayoto et al., 1980).
In vitro studies by Donaldson and Barreras (1966) showed that CrCl, (III)
was bound by both neutralized and acid gastric juice, while NapCrOj. (VI) was
bound by acid gastric juice alone. Binding effectively prevented uptake by
intestinal sections. Acid gastric juice, in addition to binding Na CrOj. (VI),
also was capable of reducing Cr(VI) compound to Cr(III).
These absorption values must be considered rough estimates. Short-term
urinary excretion values cannot take into account either deposition of chromium
into body sinks or the potential contribution of intestinal excretion. Absorp-
tion estimates by difference from fecal excretion also cannot take into account
the potential role of enterohepatic cycling.
MacKenzie et al. (1958) administered either K2Cr04 (VI) or CrCl3 (III) in
the drinking water to rats at 25 mg/S, (ppm). After 1 year of exposure, animals
receiving the hexavalent compound had average tissue levels that were 9 times
higher than those receiving the trivalent salt. This was interpreted to indicate
much greater gastrointestinal absorption of the hexavalent salt. The possible
role of differences in elimination kinetics was not addressed.
Mertz et al. (1965) administered a single dose of 0.15, 1.0, or 10 pg
CrCl,/g body weight by stomach tube to rats. By comparing whole body radio-
activity 4 to 10 days after an intragastric dose with that following an intra-
venous dose, they suggest 2 to 3% as a rough estimate of gastrointestinal
absorption. Absorption was found to be independent of dose level or dietary
5-5
-------
chromium history (deficient versus supplemented). (Doisy (1971) estimated
absorption at 0.5$.
Visek et al. (1953) gave a single dose of CrCl, (III) by gavage to rats.
Tissue distribution indicated <0.5% absorption after 4 days. Urinary excretion
suggested higher absorption estimates, but fecal contamination of the samples
was suspected.
MacKenzie et al. (1959) administered Na^rO^ (VI) as a single dose by
stomach tube. Urinary excretion of 6$ of the dose after 14 days in fasted rats
and 3% in non-fasted rats was observed and used as an estimate of absorption.
Blood chromium concentrations when rats were given Na^CrO^ were 2-fold higher
51
than for animals given CrCl,. This is not surprising, since Cr(III) is cleared
from the serum much more rapidly than Cr(VI) disappears from the erythrocytes.
Urinary excretion data for Cr(III) compound were not reported.
Additional evidence for the poor absorption of CrCl_ (III) was presented by
Mertz et al. (1965). They found that a 100-fold greater oral dose was required
to alter glucose tolerance in rats than the intravenous effective dose.
Ogawa (1976) examined differences in absorption and metabolism via several
routes of exposure using CrCl (III) and Na CrO (VI) administered to rats. They
found gastrointestinal absorption to be 2.4 and 1.4$, respectively, for the two
salts. When animals were fasted for 48 hours, absorption of both salts was
increased to 11$.
Although precise quantitation of absorption efficiency is impossible, a
reasonable estimate from the available literature discussed above is that both
valence states are absorbed at efficiences of <5$.
5.1.3. Chromium Absorption Through the Skin. Percutaneous absorption of
chromium appears to be related to valence state, the particular salt employed,
and the concentration applied.
5-6
-------
Mali (1963) conducted a series of in vitro and in vivo studies to determine
penetration of potassium dichromate(VI) and chromic sulfate(III) through the
skin (the source of skin was not mentioned). They found that neither compound
diffused spontaneously through intact, isolated epidermal membranes. The
diffusion constant for diffusion of the Cr(VI) compounds through dermis
(314 x 10~6 cm2/min) indicated unimpeded absorption; however, the diffusion
ft ")
constant for the trivalent compound was 26.6 x 10~ cm /min.
Significant amounts (30$) of Cr(III) were bound to dermal proteins in vitro,
while only very small amounts (1$) of Cr(VI) were bound (Mali, 1963). These
results were confirmed in human volunteers. It was found that in vivo potassium
dichromate but not chromic sulfate penetrated intact epidermis. In addition,
reduction of Cr(VI) to Cr(III) in the skin tissue was demonstrated. The rate of
this reduction was pH dependent.
Samitz and Katz (1963, 1964) presented additional data which indicate that
Cr(III) binds to skin in vitro, and that binding following exposure to Cr(VI)
salts is dependent upon reduction to Cr(III).
Samitz and Gross (1961) presented preliminary evidence that there is no
difference in absorption of potassium dichromate(VI) as compared to chromium
nitrate(III) in vivo in guinea pigs. Samitz and Shrager (1966) reported addi-
tional evidence that permeability of the skin to Cr(III) is dependent upon which
salt is employed. Their data indicate that absorption of chromic sulfate is
negligible, absorption of chromium nitrate is somewhat greater than chromic
sulfate absorption, and absorption of chromic chloride is as great as potassium
dichromate. Waklberg (1968) noted that percutaneous absorption of sodium
chromate was greater at pH 6.5 and higher compared with pH 5.6 and lower.
Wahlberg (1965) reported that absorption of a given salt is dependent upon
the concentration applied. In this study, application of 0.017 M to 0.239 M
5-7
-------
solutions of Cr(III) resulted in percentages of absorption which were not statis-
tically different from absorption experiments under the same conditions for
Cr(VI). However, at concentrations from 0.261 to 0.398 M, significantly more of
the Cr(VI) compound was absorbed. Absolute absorption rates for chromic'chloride
were maximal with application of 0.239 to 0.261 M solutions, reaching levels of
p
315 to 330 n moles Cr/hour/cm . In contrast, absorption rates of sodium dichro-
p
mate reached a maximum of 690 to 725n moles Cr/hour/cm at a concentration of
0.261 to 0.398 M, the maximum absorption was 4$ for the 0.261 M solution.
In conclusion, percutaneous absorption of chromium is dependent on valence
state, anionic form, the concentration and pH of the applied solution.
5.2. CHROMIUM TRANSPORT, METABOLISM, DISTRIBUTION, AND ELIMINATION
5.2.1. Transport and Metabolism. The mechanism of chromium transport is depen-
dent upon the valence state after reaching the bloodstream. Cr(VI) readily
crosses erythrocyte membranes. Kitagawa et al. (1982) demonstrated in vitro the
inhibition of Cr(VI) penetration into erythrocytes by 4-acetamido-V-isothio-
cyano-stilbene-2,2'-disulfonic acid, an inhibitor of anion transport. After
entry, the Cr(VI) undergoes reduction to Cr(III) and this reduction has been
investigated in other cell systems. In studies with cultured hamster fibro-
blasts, Levis et al. (1978) have shown that only Cr(III) could be identified
intracellularly following treatment with potassium dichromate. The mechanism
for intracellular reduction is not completely understood. Several investigators
have proposed a role for the mixed function oxidases. These proposals are based
upon the reduction of mutagenic activity following addition of microsomal
preparations to Cr(VI) (Gruber and Jennette, 1978; Lofroth, 1978; Petrilli and
DeFlora, 1978a). While rat liver preparations are very effective in reducing
5-8
-------
mutagenic activity, lung preparations show only minimal activity (Petrilli and
DeFlora, 1978b). Addition of reduced cofactors, such as GSH, NADH, or NADPH,
also resulted in reduced activity. Langard (1979) presented evidence that
although reduction of Cr(VI) may take place at any intracellular site where
electron donors are available (which includes the microsomes), the primary site
of reduction is within the mitochondria. In relation to chromium binding to
hemoglobin, Kitagawa et al. (1982) reported that the reduced cofactor GSH was
necessary for the in vitro binding of Cr(VI) to purified hemoglobin, suggesting
that either the chromium or the hemoglobin has been reduced for binding to occur.
Evidence that the reduction of Cr(VI) to Cr(III) may be of biological importance
was provided by Aaseth et al. (1982). Cr(VI) was shown to bind to erythrocytes
under in vitro incubation conditions. When GSH was added to the incubation
media, the binding of chromium to the erythrocytes decreased as a result of the
reduction of Cr(VI) to Cr(III) and the inability of Cr(III) to penetrate the cell
membrane (Aaseth et al., 1982). Reduction of chromium binding was also observed
when intracellular GSH was decreased by the action of diethylmaleate. In this
case, it was proposed that the lower binding resulted from the failure to reduce
Cr(VI) to Cr(III) in the cytosol. Jennette (1982) demonstrated that a Cr(V)
intermediate was formed during the in vitro reduction of Cr(VI) to Cr(III) by rat
liver microsomes and that this intermediate may be the chemically reactive form
of chromium.
Localization in vivo of Cr(III) within the body cells of the rat appears to
be time dependent, with initial high concentrations within the cytosol and
subsequent translocation to mitochondrial and nuclear fractions (Langard, 1979).
In addition, the partitioning of intracellular chromium appears to be dose
related. Tandon et al. (1979) found that chromium nitrate(III) doses of 1, 2 or
3 mg/kg administered intraperitoneally resulted in an increased percentage in
5-9
-------
the mitochondrial fraction with increasing dose. Early hypotheses assumed
intracellular binding of Cr(III) to proteins; however, in vitro data developed by
Sanderson (1976) indicated that intracellular chromic ion exists in the form of
coordination complexes with small organic anions. For Cr(VI), Alexander et al.
(1982) found that the relative chromate uptake in rat liver mitochondria
decreased with increasing chromate dose.
Transport of Cr(III) is facilitated by specific binding with transferrin
(Hopkins and Schwartz, 1964). Following very large doses, binding with other
serum proteins occurs. Red blood cells appear to be essentially impermeable to
Cr(III). Impermeability of all cell membranes to Cr(III) compounds has been
traditionally accepted; however, Levis et al. (1978) have shown uptake of CrCl_
by cultured hamster fibroblasts. In addition, Tandon et al. (1979) have shown
significant intracellular levels of chromium following in vivo exposure to
Cr(III) nitrate.
A number of investigators have demonstrated that in pregnant rodents
exposed to inorganic chromium only a very small fraction of the administered dose
51
is transported to the fetus in utero. Visek et al. (1953) gave Cr, as a single
dose, in the form of sodium chromate(VI) or chromium chloride(III) intravenously
to rats on days 15 to 20 of gestation. Fifty pCi per dose was administered, and
specific activity ranged from 200 to 1250 pCi/mg. Litters were examined 2U hours
post-injection. Regardless of chemical form or time of administration, recovery
of Cr per total litter never exceeded 0.13$ of the administered dose.
Mertz et al. (1969) administered Cr as chromium acetate(III) to rats.
Single doses were given either intravenously or by gavage at mating. Repeated
doses were given either by daily intubation during gestation or by administration
in the drinking water, at a concentration of 2 mg/£. Five pCi were given intra-
venously and either 5 or 250 pCi by gavage. Specific activity ranged from 30 to
5-10
-------
100 pCi/pg. Pups were sacrificed no later than 3 hours following birth. No Cr
was detected in the young at birth following a single dose by either route to the
dam at mating. Repeated dosing by gavage during gestation resulted in labeling
of the litters at birth with from 0.5 to 1.5$ of the mothers' total body
activity. Exposure of dams via the drinking water did not result in transfer of
51Cr to the litters.
Danielsson et al. (1982) found considerable differences between Cr(III) and
Cr(VI) in their distribution in embryonic and fetal uptake. Forty-two pregnant
mice received single doses (5 pg/kg intravenously) of CrCl_ or Na-Cr^O™ on days 8
to 18 of gestation. On day 13 of gestation, embryonic concentrations were 12%
Cr(VI) and 0.4$ Cr(III). Fetal concentration of both compounds increased with
gestational age, probably binding to fetal calcified bone, which develops on day
14.
lijima et al. (1983) found that radiolabeled Cr levels in fetal tissues
increased, while levels in maternal blood decreased after single intraperitoneal
injections of 9.8 mg Cr (as CrCl,)/kg body weight.
Matsumoto et al. (1976) examined placental transfer following subcutaneous
administration of CrCl-(III) to ICR mice. Six control mice were injected with
saline, 11 mice were injected with CrCl, (as Cr) at a dose of 9-76 mg/kg, and
6 mice were injected with 19.52 mg/kg. Mice were injected 9 times every other
day from the first to the sixteenth day of gestation.
Although there appeared to be a trend of increasing chromium in the pups
with dose, levels were not significantly different from controls for either dose
group. The high dose dams did exhibit placental chromium concentrations that
were significantly elevated above controls.
In contrast to these studies, when Cr was administered to rats in the form
of glucose tolerance factor (a low molecular weight organic complex isolated from
5-11
-------
yeast) by stomach tube during gestation, from 20 to 50$ of the dam's radio-
activity was detected in the litters (Mertz et al., 1969). This is consistent
with the role of chromium as an essential trace element.
In conclusion, the limited data indicate that small amounts of chromium are
transferred from mother to offspring. In one study, embryonic and fetal uptake
of Cr(VI) was approximately 10 times higher than that of Cr(III), when both were
administered at the same doses to pregnant mice. In light of the limited work
that has been done in this area, it is clear that further investigation is called
for, particularly with regard to uptake in the early fetus.
5.2.2. Distribution.
5.2.2.1. BLOOD — Once absorbed, Cr(III) compounds are cleared rapidly
from the blood and more slowly from the tissues. This also indicates that blood-
chromium levels are inadequate indicators of body burden. Clearance of Cr(VI)
from the blood is slower, presumably due to uptake by erythrocytes followed by
reduction to the relatively impermeable Cr(III).
Hopkins (1965) injected 0.1 pg Cr (as chromium chloride)/100 g intra-
venously in male rats. The blood chromium content as a percent of the 15 minute
blood concentration at various time intervals was: 30 minutes, 94$; 1 hour, 87?;
2 hours, 69?; 4 hours, 66$; 8 hours, 47$; 24 hours, 17$; 48 hours, 9$; 96 hours,
5$. Withey (1983) replotted this on semi-log paper and resolved the curve into
initial, intermediate, and terminal phases with respective half-lives of 0.56,
5.53 and 57 hours (Figure 5-1). A correlation coefficient of 0.98 was derived
when these data were fit to a linear curve (Y = a+b In x).
Visek et al. (1953) compared clearance and distribution following intra-
venous injection of several chromium salts in rats. Four days following injec-
tion, NaCrCL (III) blood levels were <0.02$ of the administered dose per gram of
5-12
-------
Q.
2
HI
oc
u
oc
57 HOURS
2-
t% = 0.56 \
HOURS \
I J I I
20 40 60 80
HOURS AFTER INJECTION
100
Figure 5-1. Rate of blood clearance of intravenously injected Cr (III)
from male rats. Data from Hopkins (1965) as replotted by
Whitney (1983).
5-13
-------
blood and CrCl^ (III) levels were 0.05$ of the administered dose per gram of
blood. In contrast, Na^rOjjUl) levels 4 days post-injection represented 0.52$
of the administered dose per gram of blood. Nearly complete blood clearance for
this salt was not achieved until 42 days post-injection compared with only 7 days
for the trivalent salt.
Baetjer et al. (1959a) administered Na^rO^ (VI), K2Cr20? (VI), or CrCl3
(III) intravenously to guinea pigs. For the Cr(VI) salts, values of pg Cr/10 gm
dry tissue/200 pg Cr injected for erythrocytes 1 and 3 days post-injection were
28 and 31, respectively, and for plasma, 4 and 2, respectively. For the Cr(III)
salt, erythrocyte values at the 1 and 3 day time points were 1 and <1, respec-
tively, and plasma values were 5 and 3, respectively.
5.2.2.2. OTHER BODY COMPARTMENTS — Visek et al. (1953) have reported
organ distribution of several chromium salts following intravenous injection in
rats. Sodium chromite (NaCrO.) (VI) was concentrated in large quantities by the
reticuloendothelial system, which in combination with the liver accumulated 90$
of the dose. At 42 days post-exposure, organs with detectable levels were:
spleen > liver > bone marrow > tibia epiphysis > lung > kidney. The liver and
spleen contained 33 and 50$ of their 4 day values, while the lung contained 10$.
The extensive accumulation of chromite in the reticuloendothelial system is
postulated to be the result of the formation of colloids by chromite at
physiological pH.
Chromic chloride concentrated in the liver, spleen and bone marrow; once
deposited, it cleared slowly. At 4 days post-exposure, CrCl, exposed animals had
lower percentages of the total dose in their livers, spleens, and bone- marrow
than those exposed to NaCrO . More CrCl, than NaCr02 accumulated in the kidney,
however. All organs gradually cleared chroraite over the period of the study. In
contrast, the liver in CrCl., -exposed rats was the only organ to clear significant
amounts of chromium over the study period (45 days).
5-14
-------
Cr(VI) was picked up to a much smaller extent than either of the Cr(III)
salts. At the 42-day time point, <1$ of the dose was found in the liver. For
comparison with the trivalent salts, 4-day tissue concentrations in animals
given sodium chromate were (% dose/g tissue): liver, 0.52$; spleen, 0.91?; bone
marrow, 0.56?; tibia epiphysis, 0.38$. At 42 days, corresponding values were:
liver, 0.07$; spleen, 4.8$; bone marrow, 0.16$; tibia epiphysis, <0.02$. The
increase in spleen concentration was attributed to erythrocyte destruction.
Hopkins (1965) examined the kinetics of distribution with trace quantities
of chromic chloride in the rat. Measured concentrations at 15 minutes post-
injection were highest in lung and kidney, followed by heart, pancreas, liver,
bone marrow, spleen, testis, and brain. The heart, lung, pancreas, liver, and
brain showed maximal levels at this 15-minute time point, and subsequently
declined. The spleen reached a maximum level 96 hours post-exposure and the
testis 4 hours post-exposure (300 times the initial level), while the kidney
concentration remained unchanged over 96 hours.
Following intratracheal administration of chromate to rats, highest concen-
trations were found in the lung, followed by liver, kidney, and spleen (Baetjer
et al., 1959a). In contrast to the injection studies, significant concentra-
tions were not found in bone.
Following a single oral dose of chromate administered by gavage, highest
concentrations were found in the liver, followed by the kidney and the spleen
(MacKenzie et al., 1959). When chromate was administered in the drinking water
for 1 year, highest concentrations were found in the spleen, followed by the
kidney, liver, and bone. Year long administration of chromic chloride resulted
in the same type of distribution pattern with lower absolute quantities.
Generalizations regarding the behavior of the two valence states are
difficult to make, since almost all testing has involved chromic chloride(III)
5-15
-------
and sodium chromate(VI) or potassium dichromate(VI). The data of Visek et al.
(1953) showed differences in uptake and distribution between chromic chloride
and chromite; the behavior of other Cr(III) salts is difficult to anticipate.
Data from chromium chloride(III) do indicate that once absorbed, this salt is
quickly cleared from the bloodstream and has a long tissue residence time.
Cr(VI) salts have longer residence times in the blood and are more rapidly
cleared from the tissues. Comparisons of tissue distributions are difficult,
because different authors choose different organs to examine and time between
exposure and analysis varied considerably. In almost all instances, the
following organs accumulated significant amounts of chromium: liver, kidney,
spleen, and bone marrow. In addition, there are some data which indicate that
the testis, heart, lung, and brain accumulate considerable amounts. Tissue
distribution does not appear to be affected by the sex of the animal or by
dietary history in terms of chromium content.
In humans, Teraoka (1981) studied the distribution of 24 elements in 12
Japanese males. The samples obtained at autopsy were from seven workers exposed
to heavy metals and five subjects with no undue exposure. Although absolute
concentrations of chromium varied extensively between individuals, the greatest
concentrations of chromium were found in the Hilar lymph nodes and lungs followed
by spleen, liver, kidney, and heart which all contained approximately the same
level for each individual. The distribution patterns were similar for the seven
men occupationally exposed to metals and the unexposed control group. The
average level of chromium in the Hilar lymph nodes of unexposed men was 8.2 ppm
as compared with 2400 ppm for the two chromium plating workers and 152 ppm for
the three chromate refinery workers. An airplane painter and stone mason who had
possible exposure to chromium had levels of 33 and 4.2 ppm, respectively. The
order of organ distribution appeared to be similar regardless of whether the
5-16
-------
exposure to chromium was from occupational (high levels) or environmental (low
levels) sources.
Similarly, the highest levels of chromium were detected at autopsy in the
lungs of a worker who died of respiratory cancer 10 years after retiring from a
chromate manufacturing plant (Hyodo et al., 1980). The chromium levels in
different portions of the lungs of this worker varied from 616 to 7100 ng Cr/g
wet weight as compared to values of 19-3 to 881 ng Cr/g wet weight of similar
lung sections from five control subjects. The suprarenal gland, brain, and skin
also had relatively high levels of chromium as compared to the control values.
It was reported that Cr(VI) was more prevalent in the chromate worker; however,
the wet oxidation method, using hot nitric acid and perchloric acid, has the
potential of oxidizing Cr(III) to Cr(VI) and makes these results invalid.
Lim et al. (1983) conducted a study of radiolabeled Cr kinetics in three
normal and three hemochromatotic patients used in a previous study. After
intravenous injection of Cr(III), absorption and distribution were determined
with a whole body gamma-ray scintillation scanner, whole body counts and plasma
clearance determinations. Among all body organs studies, the liver and spleen
contained the highest levels. In fact, after 3 months the liver contained half
of the total body burden of chromium. For the liver, spleen and thigh, there
were three major accumulation and clearance components, grouped in ranges of
half-lives of 0.5 to 12 hour (fast), 1 to 1U days (medium) and 3 to 12 months
(slow). More than 50? of the blood plasma of chromium was absorbed within hours
of administration. On the basis of this study, a model was constructed in the
form of a plasma pool in equilibrium with the three compartments. The fast
compartment occurred in adipose and muscle tissues; the middle compartment was
distributed equally between adipose and muscle tissues, and the liver and spleen;
the liver and spleen composed the slow compartment.
5-17
-------
5.2.3. Elimination. Hopkins (1965) examined the kinetics of single doses of 0.1
or 0.01 pg Cr administered intravenously as CrCr.. Urinary excretion was
followed over 4 days, and an inspection of a semi-log plot of these data
indicated at least two elimination components, the first representing rapid
blood clearance of chromium and the second the slower elimination from soft
tissues. Hopkins data also indicate that, although the urine is the primary
route of excretion, the intestines also play a small role.
Mertz et al. (1965) followed the elimination of single doses of intra-
venously administered chromic chloride over a longer time period (72 days).
Whole body retention of chromium as a funtion of time was represented by a
polynomial with three distinct regression terms (Figure 5-2). This evidence
supports a three compartment elimination process. The half-lives for chromium in
the three compartments were estimated from clearance rates to be 0.5 days,
5.9 days, and 83-4 days. At the end of 72 days, 13.5$ of the original dose was
retained.
Onkelinx (1977) also followed elimination of an intravenous dose of chromic
chloride. Their data agree with Mertz et al. (1965), in that the plasma
disappearance curve could be described by the sum of three exponential compo-
nents, representing three compartments. Excretion has been quantified and 3
components have been identified. Urinary excretion represented 51 to 64? of the
total excretion, fecal excretion 5 to 8$, and clearance of chromium into a body
"sink" 31 to 41$. The sink represents compartments with extremely long half-
lives.
In a comparison of Cr(VI) and Cr(III) elimination in rats, Sayoto et al.
(1980) administered radiolabeled Na-CrO^ or CrCl, to animals either by intuba-
tion or intravenous injection. Regardless of the route of administration, the
Cr(VI) compound was excreted more rapidly through the feces and urine. The
5-18
-------
100
UJ
U
cc
10
I V
t% • 5.9 DAYS
t% * 0.5 DAYS
I I I
10 20 30 40 50
DAYS AFTER INJECTION
I
40
I
60
70
I
80
Figure 5-2*. Whole-body elimination of intravenously administered
5 Cr (III) in male rats (Mertz et al.v 1965).
5-19
-------
biologic half life for Wa^rO^ and CrCl, after intubation was 22.24 and 91.79
days, respectively. The authors suggested that these findings indicated that
Cr(III) has a higher affinity for body constituents than Cr(VI).
Yamaguchi et al. (1983) studied the excretion of a Cr(III) and Cr(VI)
following subcutaneous administration to rats of CrCNO^)- and KpCrpO.,. Within 24
hours after treatment 93.6$ of the Cr(VI) was removed from the site of injection
while only 21.3$ of the Cr(III) was adsorbed. In 7 days, 48 and 8.0$ of the
Cr(VI) and Cr(III), respectively, were eliminated in the urine. The biologic
half-time of Cr(VI) in different organs was determined, with half-times for the
terminal component for lungs, liver, kidney, brain, heart, testes and blood
calculated to be 20.9, 15.7, 10.5, 9.6, 13.9, 12.9 and 13.9 days, respectively.
Collins et al. (1961) examined chromium excretion in dogs dosed intra-
venously with chromium chloride or sodium dichromate. Acute exposures (dose not
stated) showed that 25$ of the Cr(III) salt and 9$ of the Cr(VI) salt were
excreted in the urine within 4 hours. Less than 0.5$ appeared in the bile.
In bile fistulated dogs, 4-day excretion values were 50$ in the urine, 0.5$
in the bile, and 3.7$ in the feces for the Cr(III) salt, and 20, 0.9, and 1.2$ of
the dose in urine, bile, and feces, respectively, for the Cr(VI) salt. Excreted
chromium was readily dialyzable indicating that, if the chromium was bound, the
molecule it was combined with was of small size or the binding was reversible.
Tubular reabsorption appeared to represent >63$ of the amount filtered.
Davidson et al. (1974) examined kidney handling of chromium in normal human
subjects and in dogs (not pre-treated with chromium). Their results indicate
that in both dogs and man physiological quantities of chromium have a >99$
reabsorption value. Their data also suggest that there is an active transport
mechanism in the renal tubule for chromium reabsorption. They hypothesize that
the normal filtered load for chromium may be close to the maximal reabsorption
5-20
-------
rate. If this is the case, then any increase in plasma chromium concentration
would result in a significant increase in renal excretion of chromium.
Cikrt and Benco (1979) measured the excretion of Cr in urine, bile and
feces of rats within 24 hours after intravenous injection of TCrCl_ or
NapCrOjjGijmol/kg). Cr excretion in urine (as a fraction of the dose) averaged
22.4 + 2.8$ for Cr(VI); 51Cr excretion in bile averaged 0.5 + 0.1$ for Cr(III)
and 3.5 + 0.7$ for Cr(VI); and Cr excretion in feces (including
gastrointestinal contents) averaged 4.2 + 0.2$ for Cr(III) and 7.4 + 0.4$ for
Cr(VI). These measurements indicate that urinary excretion of CR(III) and Cr(VI)
are equivalent, but biliary excretion of Cr(VI) averages 7-fold that of Cr(III)
under the same conditions.
Norseth et al. (1982) administered 51CrCl3 and Na2 CrO^ to rats by
intravenous injection (1-9 pmol/kg) and identified Cr(III)-glutathione
complexes in bile after the administration of both compounds. They also
demonstrated that biliary excretion of the Cr-glutathione complex is inhibited
by diethylmaleate. This experiment furnishes important evidence that
glutathione is involved in the biliary excretion of chromium. Yamamoto et al.
(1982) isolated a low-molecular-weight, chromium-binding constituent in liver
and kidney cytosol of mice treated with Cr(VI); the low-molecular-weight
chromium constituent was also detected in urine and feces. Alexander et al.
(1982) showed that rat liver mitochondria accumulated Cr(VI) fairly rapidly,
although the relative uptake decreased with increasing dose. Trivalent chromium
was taken up at a much lower rate than hexavalent chromium.
5.3. SUMMARY
There are increasing experimental data available on the pharmacokinetics of
chromium. Absorption by inhalation exposure appears to occur rapidly, although
5-21
-------
it is difficult to quantify the extent of absorption. A preliminary estimate of
pulmonary absorption, following disposition of CrCl- into the lungs by
instillation, indicates that approximately 5% is absorbed. Following oral
exposure, gastrointestinal absorption was also low with estimates that only 5% or
less of chromium was absorbed. In vitro studies indicate that gastrointestinal
juices have the capability to reduce Cr(VI) to Cr(III); however, there are
insufficient data from in vivo studies to demonstrate whether this reduction
process has the capacity to eliminate any differences in absorption between
ingested Cr(VI) and Cr(III) compounds. Percutaneous absorption of chromium
through unbroken skin is variable and dependent on valence as well as the
specific salt.
After absorption, chromium is transported by the blood with Cr(III) trans-
port facilitated by specific binding proteins in the blood. Cr(VI) on entering
the blood stream diffuses into the blood cells where reduction and binding to
cellular components occurs. Both absorbed Cr(III) and (VI) can be transported to
a limited extent to the fetus in utero after exposures of the dams, although the
data do not allow quantitative estimates of fetal exposure. Chromium transported
by the blood is distributed to other organs with greatest retention by the
spleen, liver, and bone marrow. The major deposition site following inhalation
exposure is the lungs, where chromium probably binds to the cellular material
before absorption can occur.
Absorbed chromium is eliminated from the body in a rapid phase representing
clearance from the blood and in a slower phase representing clearance from
tissues. Urinary excretion is the primary route of elimination accounting for
somewhat over 50$ of the eliminated chromium, while fecal excretion accounts for
only 5% of the elimination from the blood. The remaining chromium is deposited
into deep body compartments. Limited work on modeling the absorption and
5-22
-------
deposition of chromium indicates that adipose and muscle tissue retains chromium
at a moderate level (-2 weeks), while the liver and spleen store chromium for up
to 12 months. Estimated half-lives for whole body chromium elimination are 22
and 92 days for Cr(VI) and Cr(III), respectively.
5-23
-------
-------
6. CHROMIUM AS AN ESSENTIAL ELEMENT
6.1. CHROMIUM DEFIENCY
The nutritional need for chromium as an essential element has been reviewed
by Golden and Golden (1981), Anderson (1981), Saner (1980) and Mertz (1974).
Mertz (1974) described the necessary components for considering an element as
essential. These components are: the element must be found in living matter, the
element must interact with living biological systems, and deficiency of the
element must produce a decrement in biological function. Indeed, certain
chromium compounds are found in living matter, it does interact with living
systems, and deficiency syndromes are remedied with its supplementation.
The National Academy of Science (NAS, 1980) has summarized daily intakes of
chromium by persons in the United States as: dietary, a range of 37 to 130 pg/day
with an average of 62 pg/day; air, a range of <0.5 to <4.0 of the dietary intake;
and drinking water, a range of 0 to 224 pg/day with an average of 17 pg/day.
Mertz (1974) provided similar estimates that daily intake of chromium in healthy
humans was between 5 and 100 pg/day and that this intake resulted in blood and
urine levels of chromium of 0.5 to 5 and 5 to 10 pg/fc, respectively. Infants
receive chromium through breast milk during nursing (Kumpulainen et al., (1980).
These levels of chromium intake must be adequate since no serious effects from
chromium deficiency have been observed in the general populace. The Estimated
Adequate and Safe Intake (EASI) values for chromium, provided by the National
Academy of Science (NAS, 1980), are listed in Table 6-1. The lower EASI values
are based on the average United States daily intake from a mixed diet. No
Recommended Daily Allowance (RDA) has been issued.
6-1
-------
TABLE 6-1
Estimated Adequate and Safe Intake (EASI) for Chromium*
Age (years) EASI (mg/day)
Infants
Children
0.0 to 0.5
0.5 to 1.0
1 to 3
4 to 6
7 to 10
>11
0.01 to 0.04
0.02 to 0.06
0.02 to 0.08
0.03 to 0.12
0.05 to 0.20
0.05 to 0.20
Adults 0.05 to 0.20
•Source: NAS, 1980
6-2
-------
Animal studies have been conducted to determine the effects of ingestion of
dietary chromium. Schroeder et al. (1964) maintained 54 male and 54 female mice
for life on drinking water containing 5 ppm of Cr(III). The mice exposed to
chromium had survival rates similar to those of control animals and mean body
weights that were 123? of control values. In this study the food was devoid of
all trace metals. In an identical study using groups of 50 male and female Long-
Evans rats, Schroeder et al. (1965) observed increased longevity in the treated
animals. In studies of the effects of the absence of chromium in the diet,
Roginski and Mertz (1969) raised rats in plastic cages on low protein diets
containing less than 100 ppb Cr(III). A second group of rats were given a
chromium supplement of 2 ppm Cr(III) in the drinking water. The chromium
supplemented rats had better weight gain than the chromium deficient animals.
This difference in weight gain was more striking if the animals were allowed free
access to an exercise wheel. When extreme care was taken to also prevent
airborne exposure to chromium of rats maintained on chromium deficient diets,
there was a high incidence of moderate hyperglycemia and glycosuria as compared
to animals on chromium supplemented diets. These studies suggest that small
amounts of dietary Cr(III) were beneficial to the health of these rats. Preston
et al. (1976) maintained female guinea pigs on low protein diets containing
0.125, 0.5, or 50 ppm chromium along with adequate levels of vitamins and some
other trace metals. After 8 to 13 weeks of maintenance on these diets, the
animals were mated and allowed to deliver pups. Although the diets had no effect
on the survival of animals not mated, there was a significant decrease in
survival of mated animals on the low chromium diet (13 of 24 animals died) as
compared to animals on the low chromium supplemented diet (4 of 24 animals died)
and the high chromium supplemented diet (3 of 24 animals died). It was suggested
that the added stress of mating and pregnancy along with chromium deficiency
6-3
-------
resulted in the high level of mortality. Other parameters examined included
weight gain and glucose loading; serum cholesterol levels were not altered by
dietary chromium. It was concluded that chromium was beneficial to survival in
guinea pigs as was previously reported for mice and rats.
6.2. GLUCOSE TOLERANCE FACTOR
Chromium deficiency in the diet results in glucose intolerance in both the
experimental animals and humans (Saner, 1980). Schwarz and Mertz (1959) have
identified a glucose tolerance factor (GIF) which was required in rats to main-
tain normal rates of glucose removal from the blood stream. This factor has been
obtained from natural sources including brewer's yeast and hydrolyzed pork
kidney. In further studies, Mertz and Schwarz (1959) demonstrated that rats
maintained on some commercial diets had normal glucose removal rates, while
maintenance on other diets resulted in poor removal of glucose from the blood.
When the second diets were supplemented with GTF, glucose removal rates returned
to normal values. Mertz and Schwarz (1959) have identified Cr(III) as the active
agent in the GTF. It was further demonstrated that administration of a variety
of Cr(III) compounds to glucose intolerant rats resulted in an increase in the
rate of removal of glucose from the blood. In this assay, the less stable
Cr(III) compounds were most effective, and some Cr(VI) compounds were totally
without activity.
The effects of chromium on glucose metabolism were suggested to result from
chromium being a co-factor for insulin. Mertz et al. (1965) measured insulin
stimulated C0? production in adipose tissue from chromium deficient rats
receiving supplements of 0.0, 0.01, 0.05, or 0.1 pg Cr/100 g body weight. The
adipose tissue of rats receiving 0.05 pg Cr/100 g produced significantly more
6-H
-------
COp than tissues obtained from other groups. Farkas and Roberson (1965) per-
formed similar studies with measurements of glucose utilization in rat lenses
taken from animals on chromium deficient and supplemented diets. The chromium
treatment alone did not affect glucose utilization; however, chromium treatment
in conjunction with insulin significantly increased the utilization of glucose.
Administration of 2 ppm of chromium in the drinking water of rats also facili-
tated the insulin transport of amino acids into the heart and the incorporation
of amino acids into protein. Providing chromium supplements to rats increased
the animals' sensitivity to many of the effects of insulin.
Saner (1980) states that certain groups of persons may be prone to chromium
deficiency; these groups include the elderly, diabetics, pregnant women,
malnourished children, offspring and siblings of diabetics, persons with early
coronary heart disease and their offspring. Jeejeebhoy et al. (1977) have
described a female patient placed on total parenteral nutrition for 3 1/2 years
in whom chromium deficiency was indicated. Blood chromium levels were reported
as 0.55 ng/mJl and hair chromium levels as 154 to 175 ng/gm. In addition to
glucose intolerance, weight loss, neuropathy, elevated fatty acid levels,
reduced respiratory quotient, and abnormal nitrogen metabolism were reported.
Daily administration of 250 pg chromium chloride for 2 weeks in the parenteral
infusate, followed by 20 pg/day maintenance, resulted in a reversal of symptoms.
Freund et al. (1979) report similar finding of chromium deficiency in a patient
receiving total parenteral nutrition. Supplementation of the infusate with 150
pg chromium/day resulted in the reversal of the adverse clinical findings of
impaired glucose tolerance, weight loss, and confusion.
Chromium has been known to affect glucose tolerance in humans. Levin et al.
(1968) performed oral glucose tolerance tests on 9 male and 6 female elderly
subjects. Of the subjects tested, 10 daily dietary supplements with 150 pg/day
6-5
-------
chromium for 3 to 1 months resulted in normal glucose tolerance in four subjects.
The remaining six subjects still had abnormal tests after receiving the chromium
supplement. Glinsmann and Mertz (1966) observed improved glucose tolerance in
three of six diabetics given 150 to 1000 yg of Cr(III) for 15 to 120 days. In a
similar study, Sherman et al. (1968) observed no change in the glucose tolerance
of ten diabetics who received chromium supplements of 100 pg/day for 16 weeks.
In malnourished infants, Gurson and Saner (1971) and Hopkins et al. (1968)
observed dramatic increases in glucose removal and utilization following treat-
ment with Cr(III). To ensure that the improved glucose tolerance was the result
of chromium therapy, Hopkins et al. (1968) treated five infants in a manner
similar to the others with the exception that chromium was not administered.
These five infants showed no improvement in glucose tolerance. Dieatary chromium
appears to have some effect on human glucose tolerance; however, the therapeutic
effect of chromium supplementation on subjects with abnormal glucose tolerance
was variable.
6.3- SUMMARY
Animal studies have demonstrated that chromium deficient rodents gain less
weight and have a shorter lifespan than animals maintained on a diet containing
adequate chromium levels. Chromium deficiency results in glucose intolerance in
rats and this intolerance can be reversed by dietary treatment with Cr(III).
More effective in reversing glucose intolerance is a chromium complex which has
been isolated from brewer's yeast and designated as glucose tolerance factor
(GTF). GTF may be formed in mammals following ingestion of inorganic chromium.
Some humans with abnormal glucose tolerance, such as the elderly, diabetics, and
malnourished infants have responded to dietary supplements of chromium.
6-6
-------
Although the exact level of chromium needed for good health is not known, it is
assumed from the lack of observed effects of chromium deficiency that the average
American intake of 0.05 to 0.2 mg/day is adequate. It is also worth noting that
there is a considerable difference between the low levels of chromium intake that
are associated with nutritional deficiency, and those discussed in the following
Chapter 7 which have been associated with toxic effects.
6-7
-------
7. CHROMIUM TOXICOLOGY
7.1. ACUTE EFFECTS OF CHROMIUM EXPOSURE IN MAN AND ANIMALS
7.1.1. Human Studies. Chromium metal is biologically inert and has not been
reported to produce toxic or other harmful effects in man. When in contact with
the skin, compounds of chromium in the trivalent state combine with proteins in
the superficial layers, but do not cause ulceration (NAS, 1974).
Cr(VI) compounds are responsible for the majority of the health problems
associated with all chromium compounds. They are irritating and corrosive, and
may be absorbed by inhalation, cutaneously, or by ingestion. Acute systemic
poisoning is rare; however, it may follow deliberate or accidental ingestion or
result from absorption through broken skin (NAS, 1974).
Much of the information on the effects of Cr(VI) is obtained from
occupational exposures, where the predominant exposures and related effects are
on the respiratory system and skin (NAS, 1974). As with most information derived
from uncontrolled settings, exact knowledge about length of exposure,
concentration of the chemical, and other variables are not known, making
quantitive dose-effect relationships difficult.
7.1.2. Animal Studies. Cr(III) compounds have a very low order of toxicity when
administered orally. Oral LD,-0 values for the rat have been reported as follows:
chromic chloride, 1.87 g/kg; chromium acetate, 11.26 g/kg; chromium nitrate,
3.25 g/kg (Smyth et al., 1969). Kobayashi et al. (1976) have determined oral
LDSO for chromium trioxide in mice and rats to be 135 to 177 mg/kg and 80 to
114 mg/kg, respectively, with death occurring between 3 to 35 hours.
7-1
-------
Symptomatology included diarrhea, cyanosis, tail necrosis, and gastric ulcer.
Surviving animals showed increases in liver and testes weight without micro-
scopic changes.
Cr(VI) is more acutely toxic than Cr(III). A primary effect of acute
exposures is kidney failure. Oral administration of high doses results in
gastric corrosion. The oral LD^ of sodium dichromate in humans has been
reported as 50 mg/kg (NIOSH, 1979).
Kidney effects are the primary result of acute exposures to chromium by
various routes. Relevant studies are summarized in the following paragraphs.
Mathur et al. (1977) injected rabbits intraperitoneally with 2 mg/kg
chromium nitrate or potassium chromate daily for 3 or 6 weeks. After 3 weeks of
exposure, kidneys from animals dosed with chromium nitrate showed marked conges-
tion, extravasation of red blood cells in the intratubular spaces and tubular
necrosis. Further treatment for 6 weeks did not produce additional changes.
The kidneys from animals given potassium dichromate showed marked conges-
tion, and the walls of the small blood vessels were thickened. Glomerular tufts
were shrunken in some places, while proliferation of endothelial cells,
obliterating the Bowman space, was seen in others. There was necrosis and
desquamation of the epithelium of the convoluted tubules. Red blood cells were
found in the intertubular spaces. Changes were similar whether animals were
exposed for 3 or 6 weeks.
Hunter and Roberts (1933) dosed monkeys subcutaneously with 1 to 5 m2, of a
2% solution of potassium dichromate over a 5-week period. One monkey died
following the final dose. The authors reported injury to the proximal and distal
convoluted tubules and the glomeruli of the kidneys.
Kirschbaum et al. (1981) injected rats subcutaneously with 20 mg/kg sodium
chromate. Epithelial cell injury in the kidney occurred 2 to U hours post-
7-2
-------
injection. They postulated that the interaction of chromium with specific
elements of the microfilamentous system, which are responsible for directing
intracellular flow of reabsorbed solutes, may be the initial effect of chromium
nephropathy. Baines (1965) also reports renal damage following exposure of rats
to acute, subcutaneous doses of potassium dichromate.
Evan and Dail (197*0 also report data which indicate effects on the proximal
convoluted tubules following intraperitoneal administration of 10 or 20 mg/kg
sodium chromate and the formation of large lysosomal vacuoles in this region.
They also report effects on mitochondrial configuration shortly following
exposure.
Berndt (1976) reports in vitro data which indicate there may be species
differences in kidney susceptibility to chromium nephropathy. In kidney slices
from rats, effects were independent of valence state. Kidney slices from rabbits
were more sensitive to transport process inhibition by Cr(VI).
Mathur et al. (1977) documented effects in other target organs following
acute exposure to chromium. Rabbits were dosed intraperitoneally with 2 mg/kg
chromium as chromium nitrate or as potassium dichromate. Doses were given daily
and the animals were sacrificed after 3 or 6 weeks. Administration of Cr(III)
for 3 weeks produced changes in the brain, including occasional neuronal
degeneration in the cerebral cortex, marked chromatolysis, and nuclear changes
in the neurones. Six weeks of exposure resulted in marked neuronal degeneration
in the cerebral cortex accompanied by neuronophagia, neuroglial proliferation,
and meningeal congestion.
Following 3 weeks of exposure to Cr(VI), congestion with perivascular
infiltration by inflammatory cells was noted. In addition, some neurones in the
cortex showed pyknotic nuclei and dissolution of Nissl's substance. Neuro-
nophagia and focal neuroglial profileration were also evident throughout the
7-3
-------
cerebral cortex. Changes following 6 weeks of exposure were similar. The
myocardium of animals exposed to Cr(III) for 3 weeks appeared to be normal.
However, following 6 weeks of exposure, the myocardium showed marked congestion
and degeneration of muscle fibers. Exposure to Cr(VI) for 3 weeks did not
produce any abnormalities. Exposure for 6 weeks produced changes similar to
those seen in animals exposed to the trivalent salt.
Tandon et al. (1978) report hepatic changes in rabbits exposed to chromium.
Exposure conditions were the same as in the previous study. Exposure to Cr(III)
for 3 weeks produced marked congestion and dilation of the central veins and
sinusoids. Discrete foci of necrosis were noted in liver parenchyma. After 6
weeks, in addition to marked congestion, extensive hemorrhage was seen in the
parenchyma. Slight nuclear pleomorphism and multinucleated cells were noted in
the lobules. Bile duct proliferation, increased cellularity and proliferation
of fibroblasts around portal tracts were noted.
Exposure to Cr(VI) for 3 weeks produced more extensive pathology than did
Cr(III) exposure. The liver capsule was thickened and there was marked conges
tion of central veins and adjacent sinusoids. Large areas of necrosis were seen
throughout the parenchyma. Changes seen following 6 weeks of exposure were
similar to those described following 6 weeks of exposure to the trivalent salt.
In summary, the kidney appears to be the main target for acute chromium
toxicity, with effects occurring at 1-2 mg Cr(VI)/kg body weight. Although
hepatic effects have also been observed, the kidney has received the most intense
study.
7.1.3- Chromium Hypersensitivity.
7.1.3.1. CHROMIUM SENSITIVITY AND CONTACT DERMATITIS ~ Chromic acid and
the chromates are powerful skin irritants, and, in lower concentrations, the
7-4
-------
chromates are sensitizers (NAS, 1974). Workmen exposed to the steam of boiling
dichromate solutions developed an acute primary irritant contact dermatitis
(Schwartz et al., 1957; White, 1934). White (193*0 described a diffuse erythema-
tous dermatosis that resulted from dichromate; some progressed to an exudative
phase.
Various chromium compounds have been implicated in giving rise to allergic
dermatitis with varying degrees of eczema. Parkhurst (1925) reported the case of
a woman employed in blueprint production using a process in which a 1$ potassium
dichromate solution was used as a fixative. A Q.5% potassium dichromate solution
was rubbed on the right thigh of the woman, and soom after the application, there
was a local sensation of itching and burning. Twelve hours later, the patient
developed a follicular erythematopapular dermatitis at the exposure site.
Itching and burning was reported when a similar application was made to the left
thigh.
Smith (193D reported a case of chromium sensitization in a man who had been
hospitalized after occupational exposure to ammonium dichromate. The patient
had ulcerations in the skin of both hands, and complained of asthma, and muscular
weakness and tenderness. He had a previous history of asthma and hay fever and
had further asthmatic attacks upon exposure to chromium. Following a patch test
o
with 1/t ammonium dichromate solution on a 1 cm area of normal skin on his
forearm, the man developed a mild erythema after 24 hours. After 3 days, the
erythematous area doubled in size, and there was the appearance of vesicles. An
intradermal injection of 0.1 m£ of a 0.5$ aqueous solution of ammonium dichro-
mate was given in the right forearm 8 days later. Within an hour, the patient
developed a generalized pruritis with soreness at the injection site. This
progressed with the development of a vesicular erythematous dermatitis covering
the entire hands and lower parts of the forearm. In addition, he developed
7-5
-------
diaphoresis and sibilant rales. These symptoms which required hospitilization
abated after his exposure to Cr(VI) ceased. Three control subjects who were
similarly injected showed no reaction.
Hall (19^*0 reported on 132 aircraft workers who developed dermatitis after
contact with a primer consisting of a suspension of zinc chromate powder and
magnesium silicate in a xylene solution of certain resins, including a phenol-
formaldehyde resin. Those workers who had dermatitis from the primer and who
were allergic to zinc chromate pigment had a mean duration of employment of
7 months (range: 1 week to 9 years). A battery of patch tests showed that 90 of
the workers (68?) were sensitive to the zinc chromate pigment only. (Apparently,
the zinc chromate pigment was a mixture of zinc chromate and calcium carbonate.)
Only a few workers had positive patch tests for other compounds encountered on
the job.
Forty-five cases of allergic contact dermatitis, observed in the Helsinki
area from 1945 to 1948, were reported by Pirila and Kilpio (1949). Positive
patch-testing with a 0.5? aqueous solution of potassium dichromate (pH 4.15) was
observed in 41 of the workers. The workers were involved in the following
occupations: 11 bookworkers, 10 cement and lime workers, 7 radio factory
workers using a photostatic procedure, 4 metal factory workers, 4 painters and
polishers, 3 fur workers, and 6 others.
Engebrigsten (1952) reported eight cases of cement eczema among 300 to 400
Norwegian workers exposed "more or less directly" to cement dust that contained
0.002 to 0.020$ water-soluble Cr(VI), which was described only as "water-soluble
chromates." Positive patch testing was reported in 7 of 8 patients exposed to
0.5$ aqueous solution of potassium dichromate. Positive reactions to cement
patch tests were observed in 4 of 8 patients. Of the 10 persons who served as
controls, none gave any positive reactions. Engebrigtsen (1952) subsequently
7-6
-------
tested the same eight patients with a cement slurry that had been washed free of
Cr(VI), and none of the people reacted positively.
Denton et al. (1954) patch tested a patient with a "strong specific hyper-
sensitivity to potassium chromate" with three solutions: (1) a 50 ppm solution
of potassium dichromate, (2) 1 ppm water-soluble Cr(VI) filtrate from American
Portland cement, and (3) 4 ppm water-soluble Cr(VI) filtrate from American
Portland cement. Each patch test resulted in an erythematous, edematous, and
papulovesicular reaction. There was no reaction to distilled water and none of
the control subjects reacted to any of the three Cr(VI) solutions.
Six out of 200 employees who worked in a diesel locomotive repair shop were
incapacitated by chromate dermatitis (Winston and Walsh, 1951). All were exposed
to an alkaline diesel locomotive radiator fluid that contained 0.08$ sodium
dichromate. Positive patch testing was reported for both sodium dichromate
(pH 4.25) and the radiator fluid (pH 10).
Walsh (1953), in a summary report on chromate hazards in industry, described
the following patch test results: 2% "chromate acid" applied for 24 hours on
superficial skin abrasions produced a crusted lesion in 3 weeks; 0.5$ sodium
dichromate, reapplied daily for 3 days, produced a crusted lesion in 3 weeks;
0.5$ potassium chromate, applied 8 hours/day for 3 days, produced lesions in
3 days; 0.5$ sodium chromate, 0.005$ sodium dichromate, and pure zinc chromate
also produced lesions in 3 days after being in contact with skin for 8 hours/day
for 3 days. Lead chromate did not produce a reaction after the same exposure
period. A 10$ solution of Cr(III) nitrate produced redness after the solution
was reapplied daily for 3 days.
Edmundson (1951) obtained only two positive reactions with patch testing in
56 men who had chrome ulcers. All men were said to have a history of chrome
dermatitis. He concluded that the presence of chrome ulcers did not necessarily
indicate sensitization.
7-7
-------
Other investigators (Morris, 1955; McCord et al., 1931; Levin et al., 1959)
have demonstrated sensitization to other chromium compounds in workers employed
as tanners and lithographers. Many of the workers gave positive results for
patch tests.
There have been numerous case reports of contact dermatitis resulting from
exposure to a variety of chromium-containing products, such as matches, auto-
mobile primer paint, and fumes from welding rods (Fregert, 1961; Engle and
Calnan, 1963; Newhouse, 1963; Fregert and Ovrum, 1963; Shelley, 1961). In most
of these reports, the subjects developed a positive reaction to patch testing for
a solution of potassium dichromate.
A study conducted in France by Jaeger and Pelloni (1950) demonstrated that
workers with cement eczema were sensitive to potassium dichromate. In their
study, the authors patch tested 32 patients with cement eczema and 168 patients
with eczema from other causes. Those with cement eczema gave positive patch
tests (94$) to an aqueous 0.5$ solution of potassium dichromate, while only 5$ of
the other eczema patients exhibited positive reactions from the dichromate.
However, Perone et al. (197*0 reported that only 2 of 95 construction
workers who regularly worked with cement gave a positive reaction to patch
testing to a solution of 0.25$ potassium dichromate or a solution containing
M50 ng/g Cr(VI) extracted from cement. The authors suggested that the cement
dermatitis present in the construction workers was associated with the initia-
tive nature of the cement rather than a hypersensitivity.
There have been two reported cases of eczema in men who regularly worked
with cement and who each had a green tattoo which contained chromium (Cairns and
Calnan, 1962; Loewenthal, 1960). Both men developed a positive reaction to patch
testing to a solution of potassium dichromate (concentration range: 0.1 to 2$).
Negative results following patch testing were observed when the solution was
7-8
-------
Cr(III) sulfate. The green pigment in one of the tatoos contained Cr(VI), but
the oxidation state of chromium in the other tattoo was not determined.
Krishna et al. (1976) reported that chrome workers exposed to 0.21 to
0.80 mg/m3 Cr(VI) displayed various dermatologic disorders, including skin
ulcers, dermatitis, and nasal perforations. Dermatologic irritation has been
reported by Clausen and Rastogi (1977), in auto workshop employees exposed to
0.01 to 3.75 pg/m3; by Tandon et al. (1977), in electroplaters, polishers, and
pigment workers; and by Spruit and Martin (1975), in offset printing employees
who came in contact with materials (usually inks) containing up to 60,000 ppra
chromium. Peltonen and Fraki (1983) found that in a population of 822 healthy
volunteers, 2% of 410 men and 1.5$ of 412 women showed a patch-positive test
reaction to 0.5$ K?Cr,,07 solution. Most of the reactions occurred among workers
in the printing industry.
Various dermatologic disorders among workers exposed to 0.001 to 0.020 mg
Cr/m in air and chromium on work surfaces have been reported by Lucas and
Kramkowski (1975). Chromium-induced dermatitis or eczema has been associated
with workers having contact with various chromium-containing lubricants and oils
(Rosensteel and Lucas, 1975; Weisenberger, 1976; Clausen and Rastogi, 1977).
Dermatitis associated with various chromium pigments and coloring agents has
been described by Somov et al. (1976), Fisher (1977), Fregert and Gruvberger
(1976), Venediktova and Gudina (1976), and Evans (1977a,b,c).
Allergic sensitization to chromium-containing compounds has been reported
by Burry and Kirk (1975) in several industrially exposed individuals, and by
Wahlberg and Wennersten (1977) and Kaaber and Deien (1977), who used potassium
dichromate in the patch tests. Husain (1977) showed sensitization in the
"general population" (not occupationally exposed), where, among 1312 patients
tested, 11.58$ had contact sensitivity to 0.5$ potassium dichromate in a patch
7-9
-------
test (15.59? of the males reacted to the chromate as opposed to 8.18$ of females
tested).
Perisic and Jovovic (1977) have identified allergic contact sensitivity to
chromates among 219 Yugoslavian housewives. The incidence of allergic contact
sensitivity was confirmed in H3 cases (19.62?); the percentage of positive cases
was 31-M? in women holding permanent jobs, 23.51$ in cleaners, and 14.58? in
unemployed housewives. It was concluded by the authors that chromates contained
in a variety of household products used in routine cleaning operations are the
cause of allergic sensitivity to chromates in women. No control or reference
data were available.
Jovovic et al. (1977) have confirmed the allergic contact sensitivity of
shoemakers exposed to chromates in occupational settings. Sixty percent of the
shoemakers with allergic sensitivity showed a positive reaction to potassium
dichromate. Allergic reactions in various glues to which the shoemakers were
exposed were also reported.
Exposure to chromium may sensitize certain individuals resulting in
asthmatic attacks upon subsequent re-exposure (NAS, 1974) (see Section 7.2.5.1).
7.1.3.2. SENSITIVITY TO CHROMIUM IN PROSTHESES — In recent years, there
has been an increase in the use of metal-to-plastic prostheses in orthopedic
surgery. Many of the metal implants consist of alloys containing nickel,
chromium, cobalt, and molybdenum.
Work by Swanson et al. (1973) and Coleman et al. (1973) has shown that when
two cobalt/chrome alloy surfaces rub together, cobalt and chrome are released
locally, pass into the blood and circulate throughout the tissue of the body, and
are finally excreted in the urine. In patients sensitive to metal, the vessels
supplying the bone in which the prosthesis is inserted may show obliterative
7-10
-------
changes, and in theory, lead to death of the bone, with weakening of the fixation
between the bone and prostheses, and consequent loosening (Anonymous, 1976).
There is considerable conflicting evidence in this area, as Deutman et al.
(1977) reported metal sensitivity in 4 of 66 patients induced by metal alloy
implants following total hip replacement arthroplasty; however, positive patch
tests occurred only with cobalt and nickel, while tests with chromium were
negative. Brown et al. (1977) gave patch tests for chromium sensitivity to 20
patients with sterile loose hip replacements. Of these 20 patients, none were
positive for chromium sensitivity, and of 17 patients that were reoperated on,
none showed histologic evidence of delayed hypersensitivity around the implant.
In addition, contact dermatitis and allergic sensitivity to chromium-
containing dental prostheses have been reported by Levantine (197*0 and
Ovrutskii (1976).
7.1.3.3. ANIMAL STUDIES ON CHROMIUM SENSITIVITY ~ Numerous investigators
have demonstrated that sensitization of laboratory animals can be produced by
exposure to various chromium compounds, including both Cr(VI) and (III).
In studies of delayed dermal hypersensitivity to potassium chromate or
chromium chloride, guinea pigs treated with 2.4 x 10" M chromic sulfate
developed signifiant cross sections (Gross et al., 1968). The ability of Cr(III)
compounds to evoke an allergic reaction in chromium-sensitive guinea pigs
decreased gradually in the following order: chromic chloride > chromic nitrate >
chromic sulfate > chromic acetate > chromic oxalate (Scheiner and Katz, 1973).
According to the authors, this difference depended on the free Cr(III) concentra-
tion in the solution which, in turn, depended on the degree to which the Cr(III)
formed coordination complexes with the original ligands present; strong ligands
prevented the formation of complete antigens. Sensitization of guinea pigs with
7-11
-------
chromium (III) sulfate (using Triton X-100) was 86$ successful, whereas sensiti-
zation with an aqueous solution of potassium chromate was successful in only 56?
of the animals tested (Schwartz-Speck and Grundsmann, 1972).
Jansen and Berrens (1968) reported that guinea pigs could be sensitized to
both Cr(III) and Cr(VI) by the subcutaneous injection of aqueous solutions of the
appropriate chromium salts in a 1:1 emulsion with Freund's complete adjuvant.
Hicks et al. (1979) produced hypersensitization in guinea pigs with both
Cr(VI) and (III) salts. The contact sensitization potentials of the Cr(III)
complexes in guinea pigs were proportional to the release of chromium from the
complex (Schneeberger and Forck, 1971*). Siegenthaler et al. (1983) found that
guinea pigs sensitized with either trivalent chromium chloride or potassium
dichromate reacted in vitro and in vivo to challenges with both salts. They
concluded that chromium-reactive lymphocytes are directed against common or
closely related determinants.
7-12
-------
7.2. EVALUATION OF THE CARCINOGENICITY OF CHROMIUM
The purpose of this section is to provide an evaluation of the likelihood
that chromium is a human carcinogen and, on the assumption that it is a human
carcinogen, to provide a basis for estimating its public health impact, including
a potency evaluation in relation to other carcinogens. The evaluation of
carcinogenicity depends heavily on animal bioassays and epidemiologic evidence.
However, other factors, including mutagenicity, metabolism (particularly in
relation to interaction with DNA), and pharmacokinetic behavior, have an important
bearing on both the qualitative and the quantitative assessment of carcinogenicity.
The available information on these subjects is reviewed in other sections of
this document. This section presents an evaluation of the animal bioassays,
the human epidemiologic evidence, the quantitative aspects of assessment, and
finally, a summary and conclusions dealing with all of the relevant aspects of
the carcinogenicity of chromium.
7.2.1. Animal Studies. A number of animal studies have been performed to
determine whether or not chromium compounds are carcinogenic. In these studies
metallic chromium and salts of both the hexavalent (Cr VI) and trivalent (Cr III)
states were administered by various routes. The discussions that follow are
grouped according to these various routes of administration.
7.2.1.1. INHALATION STUDIES — Baetjer et al. (1959) exposed three strains
of mice (strain A, Swiss, and C57BL) to chromium-containing dust. These strains
have,'respectively, high, medium, and low spontaneous lung tumor incidences.
The dust was similar to that found in the chromium chemical manufacturing
industry, containing 13.7% hexavalent chromium oxide (Cr03) and 6.9% trivalent
chromium oxide (0^03), along with other metal oxides. In addition to the
chromium compounds in the dust, potassium dichromate (K^C^Oy) was added at a
7-13
-------
level of 1.0%. The animals were exposed to the dust-laden atmosphere containing
between 0.5 and 1 mg of total chromium 4 hours/day, 5 days/week, for an average
of 39.7 weeks (range of 16 to 58 weeks). Table 7-1 describes the specifics for
each exposure group. At death or termination of exposure, the lungs were
examined by means of a low-power microscope, and abnormal tissues were submitted
for histologic confirmation of tumors.
TABLE 7-1. INHALATION EXPOSURE OF MICE TO CHROMIUM-CONTAINING DUST
(Baetjer et al. 1959)
Strain
Swiss
Swiss
Swiss
Strain A
Strain A
Strain A
Strain A
C57BL
C57BL
Sex
F
M
F
F
F
F
M
M
F
Number of
animals at
start
11
10
127
34
45
110
52
50
61
Number of
animals at
end
7
6
51
31
45
38
36
13
14
Duration of
exposure
(weeks)
39
39
58
16
24
38
46
42
41
Estimate
of total
Cr inhaled
(mg of Cr)
520
520
692
141
210
359
406
667
589
The incidence of lung tumors was not different in exposed mice of any
strain as compared with approximately equal numbers of the appropriate strains
of unexposed mice of the same age. There was also no difference in those
strains having high spontaneous tumor incidence with regard to the average number
7-14
-------
of tumors per mouse or the percent of mice with multiple tumors. The lung
tumors present in both control and treated animals were adenomas, which appeared
to be histologically similar; however, in exposed animals, the adenomas developed
slightly earlier in the strain A mice. Three additional small groups of mice
(two groups of 10 Swiss female mice and one group of nine female strain A mice)
were exposed to high concentrations of chromium dust (7.8 to 13 rag Cr/m-*) in
a nose-only chamber 0.5 hours/day, 5 days/week, for 43, 52, and 20 weeks,
respectively. Again, no increase in the incidence of lung tumors was observed.
In a lifetime chronic study, Nettesheim et al. (1971) exposed 136 C57BL/6
mice of each sex 5 hours/day, 5 days/week, to an atmosphere containing calcium
chromate (CaCrO,) dust at a level of 13 mg/m , with 95% of the particles less
than 0.6 microns in size. During the study, seven or eight mice were removed for
interim sacrifice at 6, 12, and 18 months; the sex distribution and exact number
of animals removed at each period were not stated. At autopsy, sections were
taken from all major organs, and with the exception of the lungs, it was
indicated that no increase in tumor incidence was observed. In the lungs,
tumors developed in six males and eight females, as compared with three males
and two females in the control group. The tumors were described as alveologenic
adenomas and adenocarcinomas, although the numbers of the different types of
tumors were not given. The authors claim in the discussion that calcium chromate
exposure resulted in a significant increase in lung tumors; however, the
performance of statistical analysis was not described in the results. In a
review of this study, the International Agency for Research on Cancer (IARC
1980) maintained that no excess in treatment-related tumors was observed.
Because of difficulties in determining the numbers and sexes of the animals removed
during the interim kills, it is impossible to perform independent statistical
analysis of these data.
7-15
-------
Two other experimental groups in the aforementioned study were exposed to
calcium chromate dust following prior treatment with 100 R whole-body X-radiation
or infection by influenza virus. A slightly greater incidence of lung*tumors
was observed after combined exposure to influenza virus, as compared with
exposure to calcium chromate alone. As a result of exposure to two potential
cancer-producing agents, these last two groups of animals cannot be used to
provide any supporting evidence for the carcinogenicity of calcium chromate.
With the limited experimental detail presented in this study, it is not possible
to determine if the small increase in lung tumors represents a significant
treatment-related increase in tumor incidence.
In the study described previously, Baetjer et al. (1959) also exposed 110
(57 males and 53 females) mixed-strain rats (from Wistar and McCollum stock) to
the same chromium dust to which the mice were exposed; 100 rats (48 males and
52 females) of similar age distribution were kept as controls. The level of
chromium in the air was 1 to 1.5 mg/m^, with exposure again for 4 hours/day, 5
days/week, for >70 weeks. During the study, nearly half of the experimental
rats died, three developed lymphosarcomas involving the lungs, while two
additional suspected lymphpsarcomas involving the lungs were also identified.
No rats developed bronchogenic carcinomas. The authors considered these findings
suggestive of a chromium-induced tumorigenic response; however, since lymphosarcomas
are common in these rats, the experiment was repeated. In the second study
(Steffee and Baetjer 1965), Wistar rats were exposed to chromium-containing
dust under the same regime as described previously. Following autopsy, three
alveogenic adenomas were detected in the treated rats and two in the controls,
and four lymphosarcomas were present in both groups. From this second study,
it was concluded that J.ymphosarcomas in rats are not associated with exposure
to chromiura-cpntaining dust.
7-16
-------
Also, Steffee and Baetjer (1965) exposed eight rabbits and 50 guinea pigs
to mixed chromium dust containing 1.5 to 2.0 mg of total chromium/m3 for 4 to
5 hours/day, 4 days/week, for 50 months and for life, respectively. The mixed
•»
chromium dust exposure consisted of 2 days/week exposure to roast dust, as
described by Baetjer et al. (1959), supplemented with 1.0% dry potassium
dichromate and the mist of a 5.0% solution of potassium dichromate, followed
by a 1 day/week exposure to the mist of a 17.5% solution of sodium chromate
and a 1 day/week exposure to residue dust (roast dust from which sodium chromate
was extracted) supplemented with 1.0% dry potassium dichromate. .There were no
lung tumors in the rabbits, and the incidence of lung tumors and other body
tumors was similar in the exposed and control guinea pigs. Under these experi-
mental conditions, inhalation of mixed chromium dust did not increase the
incidence of lung tumors.
Laskin (1972) conducted a study in which rats (number and strain not
specified) were exposed by inhalation to air containing 2 mg/m3 hexavalent
calcium chromate. The total number of exposures was 589 over a period of 891
days. The author reported one squamous cell carcinoma of the lung and larnyx
and one malignant peritoneal tumor. Because of incomplete reporting of the
experiment, this study is considered to be inadequate to assess the carcino-
genicity of calcium chromate by inhalation.
7.2.1.2. INTRATRACHEAL INSTILLATION STUDIES — In further attempts to
illustrate a lung tumorigenic response from chromium compounds, mixed chromium
dust and chromium salts were instilled into the tracheas of experimental animals.
Baetjer et al. (1959) suspended a chromium dust, similar in composition to
that used in the inhalation studies, in olive oil, and zinc chromate and barium
chromate in saline prior to intratracheal instillation into strain A, Swiss,
and C57BL mice and mixed-breed rats (Wistar and McCollum stocks). The mice
each received five to six instillations of 0.01 to 0.05 mg of chromium at 4- to
7-17
-------
6-week intervals, while the rats received 15 instillations at the same dose at
2-week intervals. The total duration of the studies was between 32 and 52 weeks.
The mice treated with chromium had a tumor incidence similar to age-matched
controls, and the rats in both the treated and control groups had no benign or
malignant tumors. In continuing work, Steffee and Baetjer (1965) treated 62
strain A mice with 0.01 to 0.03 mg chromium by intratracheal instillation of
zinc chromate (ZnCrC^). The instillations were performed at 2-week intervals
for a total of six injections, and the animals were observed until death. The
incidence of lung adenomas was 31 of 62 in the treated animals and 7 of 18 in
the untreated controls. In the controls treated with zinc chromate, 3 of 12
animals developed lung adenomas. The incidence of lung tumors in treated
animals was statistically different from controls. In the same, study, the
instillation of chromium dust, zinc chromate, and lead, ch.ro.mate (PbCrC>4) into
the tracheas of guinea pigs (13 to 21 animals/group) and rabbits (7 to 10
animals/group) produced no increase in lung adenomas. Hueper and Payne (1962)
also reported similar negative results after instillation of stro,n.tium chromate
(SrCrO^) or calcium chromate suspended in gelatin; however, the experimental
detail in the report was insufficient for adequate evaluation. There is no
convincing evidence that intratracheal instillation of chromium compounds
results in the development of lung cancer.
Steinhoff et al. (1983 unpublished draft report) investigated the carcino-
genicity of soluble sodium dichromate and calcium chromate in r^ts via intra-
trachael administration. The study consisted of 10 treatment groups, one negative
control group, and two positive control groups. Each test group contained 40 male
and 40 female Sprague-Dawley rats (10 weeks old at the beginning of the test).
The design of the dose levels selected were such as to assess the impacts, of the
chemicals delivered in single high doses as contrasted to the same dose distri-
buted over a 5-day period. Table 7-2 describes the dose levels used in this study.
7-18
-------
TABLE 7-2. DOSAGE REGIMEN FOR INTRATRACHEAL INSTILLATION
OF SODIUM DICHROMATE AND CALCIUM CHROMATE
(adapted from Steinhoff et al. 1983)
Intratracheal instillation
Test group3 1 time per week 5 times per week
Untreated — —
Physiological NaCl 5 mL/kg 1 mL/kg
Na2Cr207 * 21^0 0.05 mg/kg 0.01 mg/kg
Na2Cr207 * 21^0 0.25 mg/kg 0.05 mg/kg
Na2Cr207 * 21^0 1.25 mg/kg 0.25 mg/kg
1.25 mg/kg 0.25 mg/kg
aEach group consisted of 40 male and 40 female Sprague-Dawley rats.
The study was conducted for 2 years and 8 months. The total dose administered
for the highest dose group was 160 mg/kg. The dose levels of 5 x 0.25 mg/kg/week
of sodium dichromate as well as calcium chromate were considered to be maximum
tolerated doses. The maximum tolerated dose was demonstrated by the decline in
the body weight gain and the increases in the absolute and relative lung weights;
however, the decline in body weights was more pronounced in males than in females.
Rats administered sodium dichromate or calcium chromate one or five times
per week had no significant reduction in survival periods as compared to controls,
except in the case of females treated with calcium chromate 5 x 0.25 mg/kg
week (P <^0.05 for saline controls and P <^0.01 for untreated controls).
The distribution of lung tumors (adenomas, adenocarcinomas, and squamous
cell carcinomas) is presented in Table 7-3. An increased incidence of lung tumors
as compared to controls was observed in only one treated group in which sodium
dichromate was administered at 1 x 1.25 mg/kg/week. No lung tumors were observed
in any of the groups treated with 5 x 0.01, 5 x 0.05, or 5 x 0.25 mg/kg/week,
7-19
-------
TABLE 7-3. LUNG TUMOR INCIDENCE IN SPRAGUE-DAWLEY RATS FOLLOWING INTRATRACHEAL
INSTILLATION OF SODIUM DICHROMATE OR CALCIUM CHROMATE
(Steinhoff et al. 1983)
Lung tumor incidence
Adeno- Squamous cell
Adenomas carcinomas carcinomas
Treatment group
Untreated
Physiological NaCl 5x/wk
Physiological NaCl Ix/wk
Na2Cr207 * 21^0 5x0.01 mg/kg/wk
Na2Cr207 * 21^0 5x0.05 mg/kg/wk
Na2Cr207 * 21^0 5x0.25 mg/kg/wk
Na2Cr20? * 21^0 1x0.05 mg/kg/wk
Na2Cr20? ' 2H20 1x0.25 mg/kg/wk
Na2Cr20? * 2H20 1x1.25 mg/kg/wk
CaCrO^ 1x1.25 mg/kg/wk
CaCr04 5x0.25 mg/kg/wk
M
0/40
0/40
0/40
0/40
0/40
0/40
0/40
1/40
6/40a
9/40
5/40
F
0/40
0/40
0/40
0/40
0/40
0/40
0/40
0/40
6/40
2/40
0/40
M
0/40
0/40
0/40
0/40
0/40
0/40
0/40
0/40
2/40
1/40
0/40
F
0/40
0/40
0/40
0/40
0/40
0/40
0/40
0/40
0/40
0/40
0/40
M
0/40
0/40
0/40
•0/40
0/40
0/40
0/40
0/40
3/40
l/40b
0/40
F
0/40
0/40
0/40
0/40
0/40
0/40
0/40
0/40
3/40a
1/40
1/40
alncluded two tumors of doubtful origin. These were, in the first case
(female rat), a squamous cell carcinoma, and in the second case (male rat),
an adenocarcinoma in the paratracheal lymph nodes, probably metastases of
tumors of unknown origin.
^Included a squamous cell carcinoma of the lung (male rat). It is not certain
whether this was a primary lung tumor or a metastasis of a histologically
proven squamous cell carcinoma of the jaw.
with the exception of one adenoma in a male rat treated with sodium dichromate
(1 x 1.25 mg/kg/week). In rats administered calcium chromate, statistically
significant increases in lung tumors were found in groups treated with 1 x 1.25
rag/kg/week, as well as in a group treated with 5 x 0.25 mg/kg/week distributed
over a -period of 5 days. Table 7-4 gives a retabulation of the results presented
in Table 7-3, in which adenomas and adenocarcinomas were combined. The combined
lung tumor incidence was statistically significant in male and female rats
treated with 1 x 1.25 mg/kg/week of sodium dichromate. For calcium chromate,
the combined adenomas and carcinomas were statistically significant only in male
7-20
-------
TABLE 7-4. COMBINED LUNG TUMOR INCIDENCE IN SPRAGUE-DAWLEY RATS FOLLOWING
INTRATRACHAEL INSTILLATION OF SODIUM DICHROMATE AND CALCIUM DICHROMATE
(Steinhoff et al. 1983)
Lung tumor incidence
Treatment group
Untreated
Physiological NaCl 5x/wk
Physiological NaCl Ix/wk
Na2Cr207 ' 2H20 1x0.25 mg/kg/wk
Na2Cr20? ' 2H20 1x1.25 mg/kg/wk
CaCrO^ 1x1.25 mg/kg/wk
CaCr04 5x0.25 mg/kg/wk
Adenomas and
Adenocarcinomas
Male
0/40
0/40
0/40
1/40
8/40a
P=0,003b
10/40
P=0.0005b
5/40
P=0.027b
Female
0/40
0/40
0/40
0/40
6/40
P=0.013b
2/40
0/40
Squamous cell
carcinomas
Male
0/40
0/40
0/40
0/40
3/40
1/40C
0/40
Female
0/40
0/40
0/40
0/40
3/40
1/40
1/40
alncluded two tumors of doubtful origin. These were, in the first case (female
rat), a squamous cell carcinoma, and in the second case (male rat), an
adenocarcinoma in the paratracheal lymph nodes, probably metastases of tumors
of unknown origin.
bP-values calculated using the Fisher Exact Test.
clncluded squamous cell carcinoma of the lung (male rat). It is uncertain
whether this was a primary lung tumor.
rats treated with single weekly doses of 1.25 mg/kg/week, as well as with the
same dose distributed over a period of 5 days (5 x 0.25 mg/kg). The adequacy
of this study is validated by the carcinogenic effect of the positive controls
dimethylcarbamylchloride and benzo[a]pyrene.
In conclusion, sodium dichromate and calcium chromate administered
intratracheally in solution form in Sprague-Dawley rats resulted in positive
carcinogenic effects.
7-21
-------
7.2.1.3. INTRABRONCHIAL IMPLANTATION STUDIES — Laskin et al. (1970)
investigated the carcinogenic effects of chromium compounds using the intra-
bronchial pellet technique. The compounds used in this investigation were:
chromic chromate, chromic oxide, chromic trioxide, calcium chromate, and process
residue. Pellets were prepared from molten mixtures of materials dispersed in
equal quantities of cholesterol carrier. These studies included material of
differing solubilities and valences. A total of 500 rats were under observation
for periods of up to 136 weeks. Lung cancers that closely duplicate human
pathology were found in these studies (Table 7-5).
TABLE 7-5. CARCINOMAS PRODUCED WITH CHROMIUM COMPOUNDS IN RATS
(Laskin et al. 1970)
Material
Process residue
Calcium chromate
Chromic chromate
Chromic oxide
Chromic trioxide
Cholesterol control
Number of Squamous cell
animals carcinoma
100 1
100 6
100
98
100
24
Adeno- Heptocellular
carcinoma carcinoma
_ i
2 1
1
-
2
-
With the calcium chromate, eight cancers were found in an exposed group of
100 animals. Six of these were squamous cell carcinomas and the other two were
adenocarcinomas.
The National Institute for Occupational Safety and Health (NIOSH 1975)
criteria document on hexavalent chromium (Cr VI) described a written communica-
7-22
-------
tion from L.S. Levy in 1975 about an animal study done at the Chester Beatty
Research Institute, London. Random-bred Parton Wistar rats of both sexes
received a pellet in the left inferior bronchiolus via trachectomy under
anesthesia. The rats were kept for 2 years. One hundred rats were set up for
each of the chromium-containing-material test groups. The pellets that were
implanted contained 2 mg of test material suspended 50/50 (w/w) in cholesterol.
Negative control groups received either blank metal pellets or pellets and vehicle.
Positive control groups received 3-methylcholanthrene. The lungs of all rats
either dying during the study or killed at its termination were examined both
macroscopically and microscopically. Apart from those in the lung, tumors were
similar both in type and number in all groups. The bronchial tumors found and
microscopically confirmed are given in Table 7-6, along with the average induction
periods. Additional lung tumors not of bronchial origin and not considered by
the authors to be causally related to implantations are also listed in Table
7-6. The majority of bronchial tumors were large keratinizing squamous cell
carcinomas. Intrathoracic invasions, particularly to the right lung in the
hilar region, were common, and metastases to local lymph nodes and to kidneys
were seen.
Squamous cell carcinomas were found in 8/100 (P < 0.05) rats receiving
calcium chromate, 3/100 rats receiving zinc chromate (zinc potassium chromate),
3/100 rats receiving chromic chromate dispersed in silica, and 1/100 rats
receiving ground chromic acid. It may be that the chromic acid implantation
produced a carcinoma only because it was converted to a less-soluble hexavalent
chromium material by reaction with cholesterol. Because of its extremely great
oxidizing ability, some of the chromic acid may have been chemically reduced by
cholesterol, forming chromic chromate. Calcium chromate produced carcinomas in
5/100 (P < 0.05) rats when mixed with primene, and carcinomas in 7/100 (P <
0.05) rats when mixed with diphenylguanidine. Primene 81-R benzoate and
7-23
-------
TABLE 7-6. LIVING TUMORS FOUND AND MICROSCOPICALLY CONFIRMED
[Levy and Venitt (1975), NIOSH (1975)]
Experi-
mental
group
no.
Cora-
pound
no. Test material
No. of
rats
in
group
Bronchial
carcinoma
of left
lung
Induction
period
in days
(range)
Lung tumors
not associated
with treatment
Ground chro-
mite ore
Bolton high
lime residue
Residue after
alumina pre-
cipitation
Residue from
slurry tank-
free of
soluble Cr
Residue from
vanadium
filter
Residue from
slurry
disposal tank
Sodium dichro-
mate dihydrate
100
it
it
101
100
Pulmonary aden-
oma of left lung
Anaplastic car-
cinoma of upper
left lung
Adenoma of
right lung
Fibrosarcoma of
upper left lung
8
9
10
11
12
8
9
10
11
12
Sodium chromate "
Chromic acid "
(ground)
Chromic oxide "
Calcium chromate "
Chromic chloride "
hexahydrate
it
1
0
8
0
560
—
604(473-474) P < 0.05a
Lymphoma of
— right lung
*P-value is calculated using the
Fisher Exact Test (one-tailed).
c potassium chromate.
(continued on the following page)
7-24
-------
TABLE 7-6. (continued)
Experi-
mental
group
no.
13
14
21
22
23
24
25
26
Com-
pound
no.
13
14
15A
15B
16A
16B
17A
17B
No. of
rats
in
Test material group
Zinc chr ornate "
type IIb
Chrome tan "
Diphenyl- "
guanidine (DPG)
DPG + calcium "
Primene 81 -R 100
benzoate
Primene + cal- "
cium chr ornate
Chromic chromate "
Chromic chromate "
Bronchial
carcinoma
of left
lung
3
0
it
7
0
5
0
3
Induction
period
in days
(range)
708(657-734)
—
—
656(502-732)
—
620(440-732)
—
698(666-730)
Lung tumors
not^ associated
with treatment
P < 0.05a
P < 0.05a
dispersed in
silica
15
16
28
20
17
18
19
27
15
16
28
20
17
-18
19
27
Pellet + chol- 150
esterol
Blank pellet "
Pellet + chol- 100
esterol +
Kieselguhr
100% 3 -MCA 48
100% 3 -MCA "
50% 3 -MCA "
25% 3 -MCA
50% 3 -MCA 50
0
ii
34
36
18
13
36
—
•"• *™
493(217-730)
498(270-701)
474(284-696)
517(297-698)
498(269-732)
Adenoma of
right lung
Adenocarcinom,
of right lung
aP-value is calculated using the
Fisher Exact Test (one-tailed)
"Zinc potassium chromate.
7-25
-------
•Iguanidine failed to produce tumors when administered by themselves. No
i.al carcinomas were found in negative control groups or in rats receiving
sodium dichromate dihydrate or sodium chromate.
Levy and Martin (1983) conducted an extensive investigation of 21 chromium-
containing test materials in Wistar rats, using an intrabronchial implantation
technique. In this procedure, a stainless steel wire mesh pallet with anchoring
hooks was loaded with approximately 2 mg of test material suspended in 2 mg of
cholesterol as an inert carrier, and was surgically implanted into the lower
left bronchus of 8-week-old rats. The study comprised 21 test groups, one
vehicle-control group, and another positive control group. Details of the
study design are given in Table 7-7. Weight gain and survival rate between the
test and control groups showed a compound related effect. The rats were allowed
to live for 2 years, after which the study was terminated. As shown in Table
7-7, an examination of the lung tissues indicated that bronchial carcinomas
were found to be statistically significant in comparison with controls only in
the groups given strontium chromate, zinc chromate, and calcium chronmte. The
authors of this study concluded that the carcinogenicity of chrom,iu,m compounds
depends on the extent of their solubility. Only sparingly soluble chromium
compounds were observed to be carcinogenic.
7.2.1.4. INTRAPLEURAL INJECTION STUDIES — Production of pulmonary
tumors by the intrapleural injection of chromium compounds has also been
attempted. Hueper and Payne (Hueper 1955, Hueper 1958, Payne 1960a, Hueper
1961, Hueper and Payne 1962) described a series of studies in rats Created, by
intrapleural injection of a number of hexavalent or trivalent chromium com-
pounds. Hueper (1955) injected powdered metallic chromium into the pleural
cavity of rats, guinea pigs, and mice under the dose schedule described in
Table 7-8. No significant increase in tumor incidence, either at the injection
7-26
-------
TABLE 7-7. INCIDENCE OF LUNG TUMORS IN RATS FOLLOWING INTRABRONCHIAL IMPLANTATION
OF VARIOUS CHROMIUM COMPOUNDS
(adapted from Levy and Martin 1983)
Group
no.
1
2
3
4
5
6
^j 7
1
N)
*"J 8
9
10
11
12
13
14
15
Test material
Lead chromate
Primrose chrome
yellow
Strontium
chromate
Barium chromate
Molybdate
chromate orange
Zinc chromate
(low sol. )
Zinc
tetroxychromate
Cholesterol
(negative control)
Light chrome
yellow
LD chrome
yellow
Calcium chromate
(positive control)
Chromic acid
Medium chrome
yellow
Zinc chromate
(Norge)
Sodium
dichromate
No.
of
rats3
100
100.
100
101
100
100
100
100
100
100
100
too
100
100
100
No. of lungs
examined
98
100
99
101
100
100
100
100
100
100
100
100
100
100
99
No. of bronchial
carcinomas Probability Significance
1 0.2 - 0.15 NSt>
1 0.2-0.15 NS
43 <0. 00005 sc
0 - -
0 - -
5 0.015-0.010 S
1 0.2 - 0.15 NS
0 - -
0 - -
1 0.2 - 0.15 NS
25 <0. 00005 S
2 0.08 - 0.07 NS
1 0.2-0.15 NS
3 0.05-0.04 S
1 0.15-0.10 NS
Valence of
chromium
6
6
6
6
6
6
6
6
6
6
6
6
6
6
aFor groups 1-22, 52 females and 48 males were used. For group 23, 24 females and 24 males were used.
bNS - Not statistically significant.
CS = Statistically significant at the 5Z level.
(continued on the following page)
-------
TABLE 7-7. (continued)
i
NJ
00
Group
no.
16
17
18
19
20
21
22
23
Test material
High lime
residue
Vanadium
solids
High silica
chrome ore
Kiln frit
(2% limestone)
Recycled residue
(2% limestone)
Silica encaps.
medium chrome
yellow
Strontium
chromate
20-methyl-
cholanthrene
(positive control)
No.
of
rats
100
100
101
100
100
100
100
48
No. of lungs No. of bronchial Valence of
examined carcinomas Probability Significance chromium
99 1 0.15 - 0.10 NS 6+3
100 1 0.2 - 0.15 NS 6
99 0 - 3
100 2 0.08 - 0.07 NS 6+3
100 0 - - 6+3
100 0 6
99 62 <0. 00005 S 6
48 22 <0. 00005 S 6
aFor groups 1-22, 52 females and 48 males were used.
t>NS * not statistically significant.
CS = statistically significant at the 5% level.
For group 23, 24 females and 24 males were used,
-------
TABLE 7-8. EXPOSURE SCHEDULE FOR BIOASSAY OF CHROMIUM COMPOUNDS BY INTRAPLEURAL INJECTION
(Hueper 1955)
NS
VO
Species Strain
mice C57BL
mice Strain A
rats Bethesda Black
rats Bethesda Black
rats Bethesda Black
rats Bethesda Black
rats Bethesda Black
rats Osborne-Mendel
rats Osborne-Mendel
guinea pigs NR
Number of Compound
animals Compound (mg)
50
55
25
35
42
39
14
25
25
26
metallic 0.001
chromium powder
mixed 1 or 2
chromium dust
chromite roast 25
chromite roast 25
minus Na2Cr04
chromium acetate 25
K2Cr207 2
CaCr04 12.5
metallic 16.8
chromium powder
chromite ore 36.7
metallic 67.2
chromium powder
Number of
injections
6 injections at
2-week intervals
4 injections at
4- to 6-week
intervals
single implant
single implant
in 50 mg of fat
8 implantations
over 13 months
single implant
single implant
6 injections at
monthly, intervals
6 injections at
monthly intervals
6 injections at
monthly intervals
Reference
Hueper
Baet jer
Hueper
Payne 1
Hueper
Hueper
Hueper
Hueper
Hueper
Hueper
1955
et al. 1959
1958
960a
and Payne 1962
and Payne 1962
and Payne 1962
1955
1955
1955
NR = Not reported.
-------
site or in other organs, was observed. Payne (1960a) implanted chromite roast,
from which the soluble sodium chromate was extracted, into the pleural cavity
of 35 rats. Each rat received 25 mg of this material plus 50 mg of sheep fat,
which corresponded to 2 mg of total chromium, of which 0.4 was hexavalent
chromium. None of the 35 control animals developed tumors, while three of the
treated animals developed implantation site tumors. In an earlier study (Hueper
1958) using chromite roast not leached of sodium chromate, none of the
25 treated male Bethesda rats developed implantation site tumors during 24
months; however, the early deaths of nine of the treated animals appreciably
decreased the number of animals at risk. Hueper and Payne (1962) noted that no
implantation site tumors were observed in 42 rats during a 24-month period
following eight implantations of 25 mg of trivalent chromium acetate in gelatin
over a 13-month period. Using a single implant of 2 mg of potassium dichromate
into the pleural cavity, 1 of 39 rats developed a tumor at the implantation
site. In this study, only calcium chromate elicited a high tumor incidence
following implantation. Of the 14 rats treated with 12.5 mg of calcium chromate,
eight developed tumors at the site of implantation. These studies suggest that
intrapleural implantation of some hexavalent chromium compounds might be carcino-
genic, with calcium chromate producing the most dramatic response. Hueper
(1961) also reported that many hexavalent chromium compounds produce tumors
upon intrapleural implantation, while trivalent compounds were less effective;
however, no experimental detail, including dose, was provided. A summary of
the tumor incidences reported is presented in Table 7-9.
Baetjer et. al. (1959) also used intrapleural injection to assess the car-
cinogenicity of mixed chromium dust, containing both trivalent and hexavalent
chromium, in 30 male and 25 female strain A mice. The mice received four doses
of dust suspended in olive oil, with each dose containing 0.07 mg of chromium
7-30
-------
(Table 7-9). No increase in tumor incidence or number of lung tumors per mouse
was observed during the period extending 52 weeks after the first treatment.
TABLE 7-9. COMPOUNDS REPORTED TO HAVE BEEN TESTED FOR CARCINOGENICITY
BY INTRAPLEURAL IMPLANTATION3
(Hueper 1961)
Compound
Calcium chromate
Sintered calcium chromate
Strontium chromate
Lead chromate
Barium chromate
Sodium di chromate
Zinc yellow
Chromic chromate
Chromite roast residue
Chromium acetate
Sheep fat controls
Valence
+6
+6
+6
+6
+6
+6
+6
+6, +3
+6, +3
+3
NAC
Number of rats
with tumor s^
20
17
17
3
1
Oa
22
26
5
1
0
Percent of
tumor incidence
57
49
74
9
3
0
63
74
14
3
0
aAll animals died by the 13th to 15th month of the study,
bThere were 35 rats per group at the start.
CNA = Not applicable.
Davis (1972) injected trivalent chromite [FeO(CrAl^03] into the pleural
cavity of 25 BALB/c mice and observed only small granulomas in the lungs. The
animals were treated with a single injection of 5 mg of finely ground ore (<1
urn) containing 1 mg of chromium. The mice were killed at intervals of 2 weeks
to 18 months, and the lungs were examined with the aid of an electron microscope.
7-31
-------
A carcinogenic response was not observed in mice following intrapleural injec-
tion; however, the number of chromium compounds studied was limited.
The potential of trivalent chromium to produce lung tumors has been studied
in the sensitive strain A mouse. Shimkin and Leiter (1940) injected 5 mg of
chromite ore (39 to 60% 0^03) into the tail vein of 37 animals. Periodic kills
of 10, 10, and 17 animals were performed at intervals of 2, 4, 5, and 6 months.
respectively, and the lungs were examined for tumors. The mice treated with
chromite ore had neither a greater incidence of lung tumors nor a greater num-
ber of tumors per tumor-bearing lung than the 72 control mice. More recently,
Stoner et al. (1976) and Shimkin et al. (1978) reported similar negative results
for trivalent chromium sulfate following intraperitoneal administration to strain A
mice. In this study, groups of 20 mice received 100%, 50%, or 20% of the maximum
tolerated dose (2,400 tng/kg) in 24 injections given three times/week. The ani-
mals were examined for tumors at the end of 30 weeks. In all of these studies,
positive controls were used to demonstrate the sensitivity of the strain A mouse
to the development of chemically induced multiple lung tumors.
Although increased risk of lung cancer has been associated with the chromium
industry, as discussed below, it has proven difficult to demonstrate a carcinogenic
response in the lungs of experimental animals. Trivalent chromium (Cr III) com-
pounds have not produced lung tumors after inhalation, intratracheal implantation,
or intrapleural implantation, while hexavalent chromium (Cr VI) was not carcino-
genic by inhalation or intratracheal instillation. Some hexavalent chromium
compounds did produce tumors following intrabronchial or intrapleural implanta-
tion; however, the small number of animals (14) used by Hueper and Payne (1962)
in the study of calcium chromate, and the lack of detail in the report of
Hueper (1961), where a number of hexavalent chromium compounds were reported to
be carcinogenic, make it difficult to evaluate the carcinogenicity of these
7-32
-------
compounds to rodent respiratory tissue. For these reasons, studies of respira-
tory cancer in animals do not provide substantial confirmation of lung cancer
associated with workers in the chromium industry. However, the limited data
does suggest that of the two valences, hexavalent chromium is more likely to be
the etiologic agent in chromium-induced cancer.
7.2.1.5. INJECTION STUDIES FOR SITES OTHER THAN LUNG — Attempts have
been made to demonstrate chromium-induced carcinogenesis in other than respi-
ratory tissue. In an early study, Hueper (1955) injected either powdered chro-
mium or chromite ore into the marrow cavity of the femur of rats, rabbits, and
dogs. The experimental conditions are described in Table 7-10. Of the animals
treated, only one rat developed a tumor at the site of injection, and other
tumors observed in the treated and control groups were not considered to be
treatment-related. Similar negative results were obtained by Hueper (1955)
following intraperitoneal injection of chromium powder in rats and mice and
intravenous administration of chromium powder in mice, rats, and rabbits. The
experimental conditions used in these studies are also presented in Table 7-10.
Although some tumors were present in treated rats, these tumors were similar to
those in the controls, except in the cases of two treated rats that developed
unique insulomas of the pancreas after intraperitoneal administration of powdered
chromium, and two treated rats that developed lung adenomas. While no tumors
were observed in the controls, the causal relationship between chromium exposure
and the observed tumors is highly questionable.
Injection site tumors developed in animals after subcutaneous administration
of chromium compounds in some, but not all, studies. Payne (1960a) treated groups
of 52 C57BL mice (26 males and 26 females) with size-fractionated particles of
chromium residue dust or chromic phosphate. The animals received a single sub-
cutaneous injection of 10 mg, after which they were observed for life. The com—
7-33
-------
TABLE 7-10. EXPERIMENTAL CONDITIONS USED TO STUDY THE EFFECT OF INTRAFEMORAL, INTRAPERITONEAL, AND INTRAVENOUS
ADMINISTRATION OF CHROMIUM
(Hueper 1955)
Route
intrafemoral
intrafemoral
intrafemoral
intrafemoral
-j
(jj intrafemoral
intrafemoral
intraperitoneal
intraperitoneal
intravenous
intravenous
Species
rats
rats
rabbits
dogs
rats
rabbits
mice
rats
mice
rats
Number of
animals
Strain (M, F)
Osborne-Mendel 25 M
Wistar 25 M
Dutch 8 F
mixed breed 5 F
Osborne-Mendel 15 M, 10 F
Dutch 4 F
C57BL 53 M
Wistar 25 M
C57BL 25 M
Wistar 25 M
Chromium
compound
powdered chromium
powdered chromium
powdered chromium
powdered chromium
chromlte ore
44% Cr203
chromite ore
44% Cr203
powdered chromium
powdered chromium
powdered chromium
powdered chromium
Dose
100 mg
100 mg
140 mg
170 to 399
mg followed
in 24 months
by 340 to
to 798 mg
36 mg
147 mg
1 mg/week
for 4 weeks
5 mg/week
for 6 weeks
0.25 mg/week
for 6 weeks
9 mg/week
for 6 weeks
Duration of
observation
24 months
24 months
3 to 58 months
60 months
24 months
20 to 50
months
21 months
NR
18 months
NR
intravenous
rabbits
NR
8 F
powdered chromium
2.5 mg/kg/week 36 months
for 6 weeks,
treatment
repeated 4
months later
NR - not reported.
-------
position of the dust fraction as related to trivalent and hexavalent chromium is
presented in Table 7-11. A low incidence of injection site tumors (3 of 52) was
observed in animals treated with the unfractionated residue dust, while no tumors
were present in the controls or in animals treated with smaller particles, even
though these smaller particles had a higher proportion of hexavalent chromium. In
a study of identical design, Payne (1960b) treated mice with sintered chromium
oxide, sintered calcium chromate, and calcium chromate. Only one injection site
tumor was observed, and this was in an animal treated with calcium chromate.
Roe and Carter (1969) reported that 20 weekly injections of calcium chromate at
a dose of 5 mg for the first 2 weeks and 0.5 mg for the remaining 18 weeks,
resulted in a 75% (18 of 24) incidence of injection site tumors. Although the
title of Roe and Carter's (1969) article and the legend to the tabulated results
described the route of administration as subcutaneous injection, the experimental
section and the conclusions described the treatment as an intramuscular injection
into the flank. As a result of this uncertainty, it is unclear whether any
chromium compound has been demonstrated to produce injection site tumors following
subcutaneous administration.
7.2.1.6. IMPLANTATION STUDIES — Intramuscular implantation has also been
used with varying success to demonstrate that chromium compounds are carcinogenic.
Hueper (1958) and Payne (1960a) implanted chromite roast or chromium residue
dust mixed with sheep fat into the thighs of Bethesda Black rats. The respective
incidence of injection site tumors was 3 of 31 for animals given 25 mg of
chromite roast and 1 of 35 for animals receiving the same doe of dust. In both
studies, no tumors were present in the sheep fat-treated control animals. Payne
(1960a) also implanted 10 mg of chromium dust into the thighs of 52 C57BL mice
and observed no injection site tumors. Hueper and Payne (1959) and Payne
(1960b) used similar techniques in the investigation of pure chromium compound.
7-35
-------
TABLE 7-11. LEVELS OF HEXAVALENT CHROMIUM IN
FRACTIONATED RESIDUE DUST
(Payne 1960a)
Material
Weight of chromium as Cr/dose
Hexavalent (mg) Total (mg)
Vehicle
(tricaprylin)
Dust residue extracted
with H20
Dust residue
5 to 10
Dust residue
<2
Chromic phosphate
0.037
0.17
0.45
0.003
0.50
0.69
0.68
2.64
In a small study, Payne (1960b) observed two injection site tumors in six
Bethesda Black rats after implantation of a gelatin capsule containing 12.5 mg of
calcium chrornate. In the study by Hueper and Payne (1959), 25 mg of a chromium
compound was mixed with sheep fat prior to implantation into groups of 35
Bethesda Black rats. The implantation site tumor incidence was 8 of 35 for
calcium chromate, 8 of 35 for sintered calcium chromate, 15 of 35 for chromium
oxide, and 0 for 35 for barium chromate. On implantation of 10 mg of sintered
calcium chromate into the thighs of 52 C57BL mice; nine implantation site tumors
developed; however, only one tumor developed following the implantation of calcium
chroma.te. Hueper (1961) reported on the development of implantation site tumors
following treatment of rats with a number of chromium compounds (Table 7-12);
however, details of this study, including dose given, were not reported, and
thus it is difficult to relate this study to the other reports of implantation
7-36
-------
TABLE 7-12, COMPOUNDS REPORTED TO HAVE BEEN TESTED FOR CARCINOGENICITY
BY INTRAMUSCULAR IMPLANTATION
(Hueper 1961)
Compound
Valence
Number of tumors3
aThere were 35 rats/group at the start.
NA = Not applicable.
Percent of
tumor incidence
Calcium chr ornate
Sintered calcium
chromate
Strontium chromate
Lead chromate
Barium chromate
Sodium dichromate
Zinc yellow
Chromic chromate
Chromite roast
residue
Chromium acetate
Sheep fat control
+6
+6
+6
+6
+6
+6
+6
+6, +3
+6, +3
+3
NA
9
12
15
1
0
0
16
24
1
1
0
25
34
43
3
0
0
46
69
3
3
0
site tumors. The intramuscular implantation technique has provided relatively
consistent findings that some hexavalent chromium (Cr VI) compounds are tumorigenic
in laboratory animals.
Neither trivalent (Cr III) nor metallic chromium compounds have produced
implantation site tumors following intramuscular implantation. Hueper and
Payne (1962) implanted 25 tng of chromic acetate into the thighs of 35 Bethesda
Black rats. After a 24-month observation period, only one animal developed an
injection site tumor. In a study using powdered chromium, Sunderman et al.
7-37
-------
(1974) observed no tumors in 24 male Fischer rats after a 112-week observation
period. Atomic absorption spectroscopy indicated that each rat received 2 mg
of dust. Similar results were obtained by Hueper (1955) following repeated
injection of powdered chromium into the thighs of 25 C57BL mice. Each animal
received two injections of 0.1 mg at 2-week intervals, and this was repeated 3
weeks later for a total dose of 0.4 mg. No tumors developed in the 15 mice
that survived 3 to 13 months. Although trivalent chromium was not tumorigenic
following intramuscular implantation, only one compound was tested under a
limited experimental protocol, and it is only speculative that zero-valent and
trivalent chromium would continue to give negative results with further testing.
Maltoni (1974, 1976) gave single subcutaneous injections of 30 mg hexavalent
lead chromate or hexavalent lead chromate oxide in water to groups of 40 Sprague-
Dawley rats. Lead chromate produced 24/40 and 27/40 sarcomas respectively at the
site of injection within 117 to 150 weeks. No local sarcomas were observed in
60 vehicle-treated control rats, and in 80 untreated control rats receiving com-
parable subcutaneous injection of unspecified iron pigments, only one local
sarcoma was observed.
Furst et al. (1976) studied the carcinogenicity of hexavalent lead chromate
and calcium chromate in rats. Groups of 25 male and 25 female Fischer-344 rats
were given monthly intramuscular injections of 8 mg hexavalent lead chromate
suspended in trioctanoin for 9 months or 4 mg hexavalent calcium chromate
suspended in trioctanoin for 12 months.
Calcium chromate produced three fibrosarcomas and two rhabdomyosarcomas
at the injection site in 5/45 rats, whereas hexavalent lead chromate produced
14 fibrosarcomas and 17 rhabdomyosarcomas at the site of injection in 31/47 rats.
In addition, 3/24 lead chromate treated male rats developed renal carcinomas.
None of the above tumors were found in controls injected with the vehicle.
7-38
-------
Furst et al. (1976) also investigated the carcinogenicity of hexavalent
lead chromate in mice. A total of 25 female NIH Swiss mice were given four
monthly intramuscular injections of 3 mg hexavalent lead chromate in trioctanoin.
Two lymphomas were observed within 16 months and three lung adenocarcinomas within
24 months among 17 mice necropsied. Similar incidences were observed in vehicle-
injected and untreated female NIH Swiss mice.
7.2.1.7. ORAL STUDIES — Trivalent chromium (Cr III) has been tested
for carcinogenicity by the oral route in mice and rats. Schroeder et al.
(1964) exposed a group of 108 (equal numbers of male and female) Swiss mice to
drinking water containing 5 ppm of chromium as chromium acetate. The lifetime
exposure to this level of chromium had no effect on the longevity of females,
and only a slight decrease was noted in the longevity of males. There was no
increase in tumor incidence in the treated animals as compared with controls.
In a similar study, Schroeder et al. (1965) exposed 46 male and 50 female Long
Evans rats to drinking waLet couLdiaiuy 3 ppm of chromium as chromium acetate.
Again, lifetime exposure to this level of chromium had only a slight effect on
longevity, with no increase in tumors in treated as compared with control
animals. It should be noted that only one dose level of chromium was used in
this study, and from the lack of overt signs of toxicity, it may be concluded
that higher dose levels could be tolerated. Higher dose levels would increase
the likelihood of detecting a carcinogenic response from a weak carcinogen.
Lane and Mass (1977) observed carcinogenic activities of chromium carbaryl as
well as synergistic activities with benzo[a]pyrene in rats following tracheal
grafting techniques.
Ivankovic and Preussman (1975) incorporated trivalent chromium oxide into
the diets of 60 male and female BD rats. The trivalent chromium oxide was
baked into bread at levels of 1, 2, or 5%, and fed to the rats 5 days/week for
7-39
-------
2 years. The only effect of treatment was a dose-dependent decrease in liver
and spleen weight. Both the longevity and tumor incidence in the treated
animals were similar to that of control animals. Again, the lack of major
toxic effects of treatment may indicate that this chromium compound could have
been tested at higher levels in the diet. The authors commented that the
negative observations may have resulted from the poor absorption of chromium
from the gastrointestinal tract.
7.2.1.8. SUMMARY OF ANIMAL STUDIES — A summary of the animal carcino-
genicity studies of chromium is presented in Table 7-13. To date, chromium
compounds have not induced significantly increased incidences of tumors in
laboratory animals following exposure by the inhalation and ingestion routes.
Neither trivalent (Cr III) nor hexavalent (Cr VI) chromium compounds have induced
significantly increased incidences of lung tumors by inhalation. Similar results
have been obtained following the ingestion of trivalent chromium compounds;
however, studies have not been reported in detail. There is some positive
evidence that chromium, particularly some hexavalent chromium compounds, is
carcinogenic following subcutaneous injection or intrabronchial, intrapleural,
intramuscular, or intratracheal implantation; however, implantation site tumors
have only consistently been demonstrated using intramuscular implantation. Of
all the chromium salts, calcium chromate is the only one that has been consistently
found to be carcinogenic in rats by several routes. Calcium chromate, strontium
chromate, zinc chromate, sodium dichromate, lead chromate, lead chromate oxide,
and sintered chromium trioxide have produced local sarcomas or lung tumors in
rats at the site of application. Although the studies available indicate that
metallic chromium powder and trivalent chromium compounds are not carcinogenic,
these compounds have been studied less extensively than hexavalent chromium
compounds. The relevance of studies using intramuscular implantation to human
7-40
-------
TABLE 7-13. CARCINOGENICITY OF CHROMIUM COMPOUNDS IN EXPERIMENTAL ANIMALS
Route of
administration
Inhalation
inhalation
inhalation
inhalation
^j
1
.p-
h— '
Inhalation
inhalation
inhalation
Compound
chromium-containing
dust
chromium-containing
dust
chromium-containing
dust
CaCrO^ dust
chromium-containing
dust
chromium— containing
dust
chromium-containing
mist and dust
Species/ Dose
strain as chromium
mice/Strain A 0.5 to 1 mg/m3
mice/Swiss 0.5 to 1 mg/m3
mlce/C57BL 0.5 to 1 mg/m3
mice/C57BL 4.33 mg/m3
rats/mixed breed 1 to 1.5 mg/m3
Wlstar and McCollum
rats/Wlstar 1 to 1.5 mg/nt3
rabbits 1.5 to 2 mg/ra3
guinear pigs
Duration
of exposure
4 h/d, 5 d/wk
for 16 to 54 wk
4 h/d, 5 d/wk
for 39 to 58 wk
4 h/d, 5 d/wk
for 41 to 42 wk
5 h/d, 5 d/wk
for life
4 h/d, 5 d/wk
for >70 wk
5 h/d, 5 d/wk
for life
4 to 5 h/d, 4 d/wk
the life of
guinea pig or 50
months for rabbits
Findings
No increase in the
incidence of lung
tumors or number of
tumors / lung
No increase in the
incidence of lung
tumors or number of
tumors/ lung
No lung tumors
observed
6 of 136 males and 8
of 136 females
developed lung tumors
as compared with 3 of
136 females and 2 of 136
male controls; the
significance is not clear
Increased incidence of
lymphosa rcomas
involving the lungs, 3
of 100 in experimental
and 0 of 85 in control
group
No change in lung
tumor incidence
No increase in the
incidence of lung
tumors
Reference
Baetier et al.
1959
Baetier et al.
1959
Baetier et al.
1959
Netteshe iro
et al. 1971
Baet jer er al.
1959
Steffee and
Baetier 1965
Steffee and
Baetier 1965
-------
TABLE 7-13. (continued)
Route of
administration Compound
intratracheal chromium dust or
BaCr04 or ZnCr04
intratracheal ZnCr04
intratracheal chromium
instillation dust
intratracheal ZnCr04 or
instillation PbCrO^
intratracheal ZnCr04 or
instillation PbCr04
intratracheal Na2Cr20?
application
I
i
N)
intratracheal CaCr04
application
intrapleural Cr powdered metal
injection
intrapleural mixed chromium dust
injection
intrapleural FeO(CrAl)203
injection
intrapleural chromite ore
injection
intrapleural Cr powdered metal
injection
Species/
strain
mice/Strain A
Swiss, C57BL
mice/Strain A
rats/mixed breed
Wistar and McCollum
rabbit
guinea pigs
rats/
Sprague-Dawley
rats/
Sprague-Dawley
mice/C57BL
mice/Strain A
mice/Balb/c
rats /Os borne-Mendel
rats/Os borne-Mendel
guinea pigs
Dose
as chromium
0.01 to 0.5 mg/
injection
0.01 to 0.03 mg/
injection
0.02 mg/injection
2.3 to 2.8 mg/
injection
0.7 to 0.86 mg/
injection
1 x per week
0.05 mg/kg/wk
0.25 mg/kg/wk
1.25 mg/kg/wk
5 x per week
0.01 mg/kg/wk
0.05 mg/kg/wk
0.25 mg/kg/wk
1 x per week
1.25 mg/kg/wk
5 x per week
0.25 mg/kg/wk
0.001 mg/
injection
0.07 mg/
injection
1 mg
36.7 mg (of ore)
16.8 mg
67.2 mg
Duration
of exposure
5 to 6 injections
at 4 to 6 wk
intervals
6 injections at
2 wk intervals
15 injections at
2 wk intervals
3 to 5 injections
at 3 mo intervals
6 injections at
6 wk intervals
30 months
30 months
30 months
30 months
6 injections at
2 wk intervals
4 injections at
4 to 6 wk intervals
single injection
6 injections at
1 mo intervals
6 injections
at 1 mo intervals
Findings
No increase in the
incidence of lung
tumors or the number
of tumors /lung
No statistical increase
in lung tumors, 31
No lung tumors
observed
No lung tumors
observed
No increase in lung
tumor incidence
Significant
increased incidence
of lung tumors
at the highest
dose
No lung tumors
observed
Significant
increased Incidence
of lung tumors
Increased
incidence
of lung tumors
No significant increase
in injection site
tumors
No increase in lung
tumor incidence or
number of lung tumors/
mouse
Only small
granulomas observed
No significant increase
in injection site tumors
No significant increase
in injection site tumors
Reference
Baetjer et al.
1969
Steffee and
Baetler 1965
Baetjer et al.
1959
Steffee and
Baetjer 1965
Steffee and
Baetjer 1965
Steinhoff et al.
1983
Steinhoff et al.
1983
Steinhoff et al.
1983
Steinhoff et al.
1983
Hueper 1955
Baetjer et al.
1959
Davis 1972
Hueper 1955
Hueper 1955
(continued on the following page)
-------
TABLE 7-13. (continued)
Route of
administration
intrapleural
implant
intrapleural
implant
intrapleural
implant
intrapleural
implant
intrapleural
implant
intrabronchial
intrabronchial
intrafemoral
intrafemoral
Species/ Dose
Compound strain as chromium
chromite roast rats/Bethesda Black 25 mg (of roast)
minus NaCr04
K2Cr2t>7 rats/Bethesda Black 0.35 mg
CaCr04 rats/Bethesda Black 4.2 mg
Cr(C2H302)3 rats/Bethesda Black 5.2 mg/implant
chromite roast rats/Bethesda Black 25 rag (of roast)
variety of rats/Parton 2 mi;
chromium compounds Wistar
21 compounds rats/Parton 2 mg
Wistar
metallic chromium rat/Osborne-Mendel 100 mg
rats/Wistar 100 mg
rabbits/Dutch 140 mg
chromite ore rats/Osborne-Mendel 15 -mg
rats/Dutch 64 mg
Duration
of exposure
single implant
in sheep fat
single implant
in sheep fat
single implant
in sheep fat
8 implants over
13 mo
single implant
single implant
single implant
single injection
single injection
Findings
3 of 35 animals
developed implant
site tumors; none
in controls
1 of 39 animals
developed implant
site tumors; none
in controls
8 of 14 animals
developed implant
s ite tumors , none
in controls
No implantation site
tumors in 24 animals
No implant site
tumors
See Table 7-6
See Table 7-7
No injection site
tumors developed
except for a single
tumor in one rat
No injection site tumors
Reference
Payne 1960a
Hueper
and Payne 1962
Hueper
and Payne 1962
Hueper
and Payne 1962
Hueper 1958 .
Levy and
Venitt 1975
(as reported
in NIOSH
1975)
Levy and
Martin 1983
Hueper 1955
Hueper 1955
-------
Table 7-13. (continued)
Route of Species/
administration Compound strain
Intraferaoral metallic chromium dogs/mixed breed
Intraperltoneal 0^(804)3 mice/Strain A
intraperitoneal metallic chromium mice/C57BL
-J
1
*; intraperitoneal metallic chromium rats/Wistar
intravenous chromite ore mice/Strain A
intravenous metallic chromium mice/C57BL
rats/Wistar
intravenous metallic chromium rabbits/NR
Dose
as chromium
170 to 399 mg
followed by
340 to 798 mg
1.3 to 6.6 mg/
injection
1 mg/lnjection
5 rag/inject ton
1.95 to 3 mg
0.25 mg/injectlons
9.0 mg/injections
2.5 mg/kg
Duration
of exposure
second treatment
given 24 mo
after first
24 injections given
3/wk
4 injections at
1 wk intervals
6 injections at
1 wk intervals
single injection
6 injections at
1 wk Intervals
6 Injections at
1 wk Intervals
repeated in 4 mo
Findings
No injection site
tumors
No Increase in lung
tumor incidence or
number of lung
tumors /mouse
No tumors observed
Pulmonary adenomas in
2 of 25 animals; none
in controls
No increase in lung
tumor incidence or
number of lung tumors/
mouse
No tumors in mice
while tumors In rats
were identical to
controls
No adverse effects
after 36 months
Reference
Hueper 1955
Stoner et
al. 1976;
Shimkin et al.
1978
Hueper 1955
Hueper 1955
Shimkin and
Lelter 1940
Hueper 1955
Hueper 1955
(continued on the following page)
-------
Table 7-13. (continued)
Route of
administration
subcutaneous
subcutaneous
intramuscular
•^j
1
Ui intramuscular
intramuscular
intramuscular
intramuscular
intramuscular
intramuscular
Compound
chromium residue dust
chromium residue dust.
5 to 10 p
chromium residue dust,
<2 p
chromic phosphate
sintered Cr03
sintered CaCr03
CaCr04
lead chr ornate
(VI) oxide
lead chromate
(VI) oxide
sintered
chromium
(VI) trioxide
metallic chromium
CaCr04
sintered CaCr04
chromium
residue dust
metal lie chromium
Cr(C2H302)3
Species/
strain
mice/CSTBL
.
rats/
S prague-Daw ley
rats/
Fischer 344
rats/
Bethesda Black
mice/C57BL
mice/C57BL
mice/C57BL
rats/Fischer
rats/Bethesda Black
Dose
as chromium
0.5 mg
0.69 mg
0.68 mg
2.64 mg
0.52 mg
0.37 mg
0.33 mg
30 mg
8 mg
25 mg
0.1 mg
3.3 mg
3.3 mg
10 mg (of dust)
2 mg
5.2 mg
Duration
of exposure
single injection
single injection
single injection
single implant
2 injections at
2 wk interval?
repeated 3 wks
later
single implant
single implant
single implant
single implant
Findings
3 of 52 animals
receiving chromium
residue dust and 1 of
52 receiving CaCrO^
developed injection
site tumors
injection site sarcomas
In 27/40 treated and 0/60
vehicle control animals
injection site sarcomas
in 31/47 treated and 0/22
vehicle control animals;
3 renal carcinomas
Implantation site sarcomas
in 15/35 treated and 0/35
control animals
No tumors in 25
animals
9 of 52 mice treated
with sintered CaCr04
had Implantation site
tumors , while 1 of 52
treated with CaCr04
developed tumors
No Injection site
tumors developed
No tumors in 24
animals
1 Implantation site
tumor in 35
Reference
Payne 1960a,b
Maltoni 1974,
1976
Furst et al. 1976
Hueper and Payne
1959
Hueper 1955
Payne 1960b
Payne 1960a
Sunderman et
al. 1974
Hueper
and Payne 1962
treated animals
-------
risk following inhalation or oral exposure to chromium compounds is not clear;
however, these animal studies may indicate that some hexavalent chromium com-
pounds are likely to be the etiologic agent in human chromium-related cancer*
7.2.2. Epidemiologic Studies.
7.2.2.1. CHROMATE PRODUCTION WORKERS — The early association of respi-
ratory cancer with employment in chromium compound-related industries has been
reviewed by Baetjer (1950a). The first case report appeared in 1890, with a
total of 122 reports of respiratory cancer in chromium compound-related indus-
tries collected between this date and 1950. These cases were predominantly in
German and American industries, with one case reported in Scotland and one case
reported in Switzerland. The workers in the chromate industry were exposed to
chromite ore (trivalent chromium) and sodium monochromate, sodium bichromate,
and chromic acid (hexavalent chromium), along with vapors and gases associated
with the chromate manufacturing process. According to Baetjer, the early German
investigators suggested LtiaU liexavaleut chromium was the etiologic agent in
respiratory cancer, since respiratory cancer was not associated with the mining
of trivalent chromite ore. These early observations do not provide information
on the relative incidence of respiratory cancer in the chromate workers as com-
pared with the general population, nor were the studies sufficiently large or
controlled to support the conclusion that chromium exposure was related to
respiratory cancer. However, these early reports were the impetus for initiating
a number of epidemiologic studies of the chromate manufacturing industry in the
United States, Great Britain, and Japan. Although a relatively large number of
epidemiologic studies have been conducted of this industry, the individual
studies were often analyzing cancer incidence from the same cohort of workers.
In order to clarify the interrelationship of the cohorts in these studies, the
7-^6
-------
locations of the plants from which each study derived its exposed population are
presented in Table 7-14. It should be noted that studies of the same plant by
different investigators often resulted in the vital statistics of individual
workers being used in more than one study. The result was that in many cases,
these epidemiologic studies verified the observations of previous studies rather
than adding evidence from additionally exposed population groups.
In response to the early reports associating lung cancer with the chromate
industry, the industry initiated a retrospective epidemiologic study of the seven
chromate plants in the United States (Machle and Gregorius 1948). A total of
1,445 workers were employed in the seven plants, with each plant employing between
50 and 5UO workers. The company group life insurance records were used to deter-
mine causes of death. Adequate records were available for six of the plants, and
the cohort of men actively working in the chromate industry consisted of 11,019
man-years of experience. In this cohort, 156 deaths were observed, 32 of which
were from lung cancer. An additional 10 lung cancer deaths were reported of the
37 deaths from the plant with work records unsatisfactory for the purpose of the
epidemiologic analysis. The period of study varied with each plant from 4 to 17
years, depending on the availability of mortality data. Of the total deaths
observed in the chromate industry, 21.8% (42 of 193) were from lung cancer, com-
pared to an expected 1.3% (10 of 733) as calculated from a comparable industrial
group not exposed to chromium (industrial life insurance policyholders for the
year 1947, Metropolitan Life Insurance Co.). This difference is statistically
significant at P < 0.01. When examined individually, five of the seven plants
were reported to have respiratory cancer proportionate mortality ratios from 13
to 31 times that expected. The crude respiratory cancer mortality rate per 1,000
males was also found to be significantly (P < 0.01) increased over the crude
lung cancer mortality rate of the group of life insurance holders. This was
-------
TABLE 7-14. LOCATION OF CHROMATE MANUFACTURING PLANTS WHICH PARTICIPATED IN EPIDEMIOLOGIC STUDIES AND PLANTS FROM
WHICH VITAL STATISTICS WERE OBTAINED FOR EACH STUDY
Machle
Location and
of
plant
Gregorlus
1948
Brlnton
et al.
1952
(also pub.
as part of
PHS 1953)
Mancuso
and
Hueper
1951
Hayes
Mancuso Baetjer et al.
1975 1950s 1979
Hill
and'
Ferguson
1979
Taylor Enterllne
1966 1974
Bldstrup
1951
Bldstrup
and
Case 1956
Alderson
et al.
1981
Ohsakl
et al.
1978
Watanabe
and
Fukuchl
1975
Satoh
et al.
1981
Korallus
et al.
1982
Blttersohl
1971
Glens Falls, +
NY
Jersey City, NJ
Plant fl +
Plant »2 +
Baltimore, +
MD
Kearny, NJ +
Newark, NJ -f
Palnes-
vllle, OH +
Bolton,
England
Rutherglen,
England
Eaglescllff,
England
+ - Participating chromate manufacturing plants.
- - Non-participating chromate manufacturing plants.
aThe plant In the Watanabe and Fukuchl (1975) and the Ohsakl et al. (1978) studies nay be one and the sane; It Is
Impossible to tell from the literature. Both plants were reported to be located on Hokkaido Island, Japan.
bPlant II is the larger of the two plants. Machle and Gregorlus (1948) reported that It had 350 employees as
compared to 150 employees In Plant 12.
-------
TABLE 7-14. (continued)
Brlnton
et al.
Machle 1952 Mancuso • Hill Watanabe
Location and (also pub. and Hayes and Bldstrup Alderson Ohsaki and Satoh Korallus
of Gregorlus as part of Hueper Mancuso Baetjer et al. Ferguson Taylor Enterllne Bidstrup and et al. et al. Fukuchi et al. et al.
plant 1948 PHS 1953) 1951 1975 I950a 1979 1979 1966 1974 1951 Case 1956 1981 1978 1975 1981 1982
Hokkaido
Islands ,
Japan - - _____ ~_ __ _ +a+a__
Tokyo ,
Japan - - _____ __ __ _ __ + _
L Leverkusen,
^ W. Germany - - _____ - _ __ _ _ _ _ +
Uerdlngen,
W , Ge rma ny- - _____ _ - __ _ ___ +
Leuna Chemical
Combine,
E » Ge rmany - - ___-- -- __ _ ____
Blttersohl
1971
-
-
-
~
+
a Participating chromate manufacturing plants,
- J Non-participating chromate manufacturing plants*
aThe pla.it in the Watanabe and Fukucht (1975) and the Ohsaki et al. (1978) studies may be one and the same;
Impossible to tell from the literature. Both plants were reported to be located on Hokkaido Island, Japan.
bPlant #1 Is the larger of the two plants. Machle and Gregorius (1948) reported that It had 350 employees as
compared to 150 employees in Plant #2.
it is
-------
true for both the age group 50 and under and the age group 50 and over. A
slightly increased crude mortality rate from digestive system cancer was also
reported (1.1-8 per 1,000 versus 0.59 per 1,000, P < 0.01).
Estimates of exposure were not made in this study, since analytical data
were not available for a large portion of the period studied, and work records
did not report the shifting of personnel to different positions in the plant
during the course of employment. A limitation of this study is that no age
adjustment was done in the comparison of the mortality rates of the control
group with the mortality rates of the exposed group. In this regard, however,
it should be noted that there was a dramatic difference in lung cancer mortality
between the chromate workers and the control group for both the 50 and under
age category and the over 50 age category. Thus, it is unlikely that a
preponderance of older persons among the chromate workers was responsible for
the large difference in lung cancer mortality between the chromate workers and
the control group. Another limitation is that the authors used a lung cancer
mortality rate for oil refinery workers in the 1933-1938 period for comparison
with the chromate workers in the 1930-1947 period. Lung cancer increased
dramatically in U.S. males between 1930 and 1947 (about five times). Thus, it
is possible that the lung cancer mortality rate for the control group may have
a lower rate than that which would have been found for the period 1930-1947.
However, it is unlikely that the dramatic difference in lung cancer between the
controls and the chromate workers could be explained by the selection of a
control group from a slightly different time period than that of the exposed
group.
Brinton et al. (1952) used the disability records between January 1, 1946
and December 31, 1950 of the seven United States chromate plants to determine
occupational diseases associated with the chromate industry. These data were
7-50
-------
subsequently published as part of the Public Health Service (1953) report,
"Health of Workers in a Chromate-Producing Industry." The cohort was limited
to men who belonged to the company disability plan, which prior to 1949 was
voluntary. After 1949, participation in the plan was mandatory in all but two
plants, although participation in these plants was near 100% by the end of the
study. Only deaths that occurred within 1 year of beginning disability status
were included in this study, and the determination of the cause of death was
made solely on what was listed as "cause of death" on the death certificate.
The health experience of a "large group of industrial workers (predominantly
white)" was used for comparison. The incidence of most disabilities was com-
parable between chromate workers and the reference population, with the exception
of cancer. Cancer at all sites had an incidence of 7.1 per 1,000 in chromate
workers and 0.7 per 1,000 in the control population (P < 0.01); however, it
should be pointed out that the actual number of individuals who developed cancer
was small (26 individuals) in the group of chromate workers. Brinton et al.
(1952) further studied the cancer experience in tne chromate industry for the
period 1940-1950, using death records obtained from the group disability insurance
plans. Information was available for two plants for fhf> entire 1940-1950 time
period: one plant had records for 1943 to 1950, three plants for 1946 to 1950,
and one plant for 1949 to 1950. During this 11-year period, 44 deaths from
cancer at all sites and 32 deaths from respiratory cancer were observed. This
corresponds to a 4.5-fold increase in cancer at all sites and a nearly 29-fold
increase in respiratory cancer in the chromate workers as compared to males in
the United States as a whole. There was an indication in the data that the
increased risk of respiratory cancer was greater in nonwhite as compared to
white chromate workers; however, the number of individuals was small when
the cohort of chromate workers was subjected to this further subdivision. The
7-51
-------
authors commented on the limitations the study placed on the make-up of the
cohort as a result of the use of death records obtained solely from the company
disability plan, and suggested that the observed increased risk of respiratory
cancer represents a minimal value.
Baetjer (1950a, b) studied a cohort consisting of lung cancer patients in
two Baltimore hospitals to determine if employees of the local chromate plant
were overrepresented in this group. The records from Johns Hopkins Hospital
were examined for the period from 1925 to 1946, and from Baltimore City Hospital
from 1930 to 1948; 198 and 92 confirmed cases of lung cancer in males were
obtained, respectively, from each institution. Control groups consisted of
age-matched male patients with a hospital stay of >10 days chosen at random
from the hospital records. An additional control was used for Johns Hopkins;
this consisted of patients with cholelithiasis, since patients with this disease,
as well as lung cancer patients, may preferentially select a large medical
facility. At Johns Hopkins, 7 of 198 lung cancer patients had worked in the
chromate industry; none of the 226 randomly sampled controls and none of the
177 cholelithiasis controls had worked in the chromate industry. At the
Baltimore City hospital, 4 of 92 lung cancer patients and 0 of 499 control
patients reported exposure to chromium compounds. The percentage of chromate
workers in the lung cancer group was statistically signficantly (P < 0.01)
greater than expected. If one were to calculate the unadjusted odds ratios of
having lung cancer and being exposed to chromium, the odds ratios would be
abouf32 for the Johns Hopkins cases and about 23 for the Baltimore City Hospital
cases. Both are statistically significant (P < 0.05).
Mancuso and Hueper (1951) used vital statistics for all employees who
worked for J> 1 year in the Painesille, Ohio chromate plant during 1931 to 1949
to investigate chromate production-associated lung cancer. Of the 2,931 deaths
7-52
-------
of males in the county where the plant was located, 34 (1.2%) were from respiratory
cancer, while of the 33 deaths among the chromate workers, 6 (18.2%) were from
respiratory cancer. This difference is significant at P < 0.01. Mancuso and
Heuper (1951) indicated that 96% of the workers were exposed predominantly to
insoluble chromate. Chemical analysis of the organs of two deceased workers,
one who had died from lung cancer and one who had died from bladder cancer,
revealed that the lungs appeared to be the major storage depot for chromium,
with approximately 390 and 250 ug of chromium per 10 g of tissue detected in
each individual respectively. The high level of chromium in the first subject,
the one who had died from lung cancer, was still detected despite the fact that
the individual had not been exposed to chromium for a period of 3.4 years.
Chromium levels measured in the lungs of nonexposed individuals were nearly
zero. Mancuso and Hueper (1951) suggested that the presence of insoluble chromium
in the lung may have been an etiologic factor in the observed higher incidence of
cancer; however, the present data are too limited to support this conclusion.
Mancuso (1975) followed the vital status until 1974 of 332 chromate plant
workers from his earlier study (Mancuso and Hueper 1951) who had been employed from
1931 to 1937. Over 50% of the 332 employees in this cohort had died by 1974.
Mancuso divided this group of workers into those employed from 1931 to 1932, 1933
to 1934, and 1935 to 1937. He found that 63.6%, 62.5%, and 58.3%, respectively,
of the cancer deaths in these groups were lung cancer deaths. The latency
period for the lung cancer deaths was found to cluster around 27-36 years. The
author found that lung cancer deaths were dose-related to insoluble (trivalent),
soluble (hexavalent), and total chromium exposure, and concluded that lung
cancer mortality was associated with both trivalent and hexavalent chromium.
Workers in this study were exposed to both trivalent and hexavalent chromium,
and as exposure to one increased so did exposure to the other. Therefore, an
7-53
-------
observed dose-response to trivalent chromium may merely be a reflection of a
dose-response to hexavalent chromium, or vice versa. Thus, it is questionable
whether the author's conclusion with regard to a dose-response to both hexavalent
and trivalent chromium is correct. Furthermore, the lung cancer death rates,
which purport to show this dose-response, are based on very small numbers, and
thus the finding of dose-response is probably questionable. For six deaths due
to lung cancer, chromium levels well above control values were found in the
lungs of the deceased at autopsy 15-195 months after the last exposure to
chromium, suggesting that the lung retains chromium for a considerable period
of time.
Taylor (1966) studied a group of 1,212 chromate workers for 24 years using
Old-Age and Survivors Disability Insurance records. All of these workers were
employed in three United States chromate plants (Baltimore, Maryland; Painesville,
Ohio; and the larger of two plants in Jersey City, New Jersey) for >3 months
during the period 1937 to 1940. The vital statistics of the group were
obtained through 1960, and where deaths occurred, the cause was determined
from death certificates. Causes of death in this cohort were compared with age-
and cause-specific mortality rates for civilian males in the United States. The
most dramatic increase in the standardized mortality ratio (SMR) for chromate
workers occurred in respiratory cancer, with excesses of 8.5 times (71 observed
and 8.3 expected) observed by the termination of the study in 1960. Also, the
length of experience in the chromate industry, used as the only indication of
the extent of exposure, was compared with the incidence of lung cancer in this
cohort. After 1937, duration of employment was determined from insurance
records, while prior to this date, duration of employment was determined by
extrapolation from the age-specific employment experience of the cohort. Using
7-54
-------
this method to determine length of exposure, Taylor found that respiratory
•
cancer mortality showed a dose-response by length of time exposed to chromate.
Bittersohl (1971) studied cancer incidence among more than 30,000 employees
of the Leuna Chemical Combine in the German Democratic Republic (East Germany).
The chemical plant comprised several departments involving different chemical
exposures. Ever since the plant had opened (in 1921 or earlier), its workers
had experienced increasing cancer incidence. During the period 1958 to 1971,
588 malignancies were found among the male workers and 170 malignancies were
found among the female workers. In most of the departments, no cases of cancer
were found. However, in those departments with exposures to asbestos and tar,
coal gases, ammonia, or isobutyl oil, or exclusive exposure to asbestos, chromate,
or coal gases, the cancer incidence per 10,000 was much higher relatively than
the average for the plant. In chromate production, the cancer rate was about
ninefold higher than that of departments where the above-described exposures
did not exist. This rate exceeded the rates in departments with exposure to
either asbestos or coal gases. It was less than the cancer rate for departments
where there was exposure to asbestos and tar, coal gases, ammonia, or isobutyl
oil.
This study appears to be nothing more than an interdepartmental comparison
of cancer incidence rates within a chemical works. The author has not defined
a cohort for his study, and thus the results could be misleading with regard to
the cancer experience of chromate production workers if there was a high turnover
rate by department. In connection with the data that are presented, the author
has provided little detail. Organ-specific cancer incidence is not given, and
the author does not state whether he is comparing crude rates or rates that are
sex- and age-adjusted. Thus, although the cancer incidence rate for the chromate
7-55
-------
production workers is strikingly high in comparison with nonexposed workers,
the study design and lack of detail render the results somewhat inconclusive.
Enterline (1974) recalculated the expected deaths for the study by Taylor
(1966), but provided no explanation as to why this was done. Enterline calculated
a respiratory cancer SMR of 942 for the 20-year observation period from 1940 to
1960. Taylor had calculated a respiratory cancer SMR of 850 for the period
from 1937 to 1960. Enterline (1974) showed that the relative risk of respiratory
cancer was greater in workers in the age group <45 years (14.1 times) as compared
with the older age group of 55 to 64 (6.8 times). Enterline also showed that
the risk of respiratory cancer was highest shortly after the cohort was identified,
suggesting "a short latent period which is probably the result of exposure to a
very potent carcinogen."
Hayes et al. (1979) conducted a cohort mortality study of workers at a
Baltimore, Maryland, chromium chemical production plant. The plant underwent
extensive changes in the mill and roast operation and bichromate operations in
1950 and 1951, when a new facility was built to house these operations, and in
the chromic acid and special products operations in 1960, when a new facility
was built to house these operations. The new facilities were constructed for
the purpose of lowering employee exposure to chromium. In this study, vital
status was determined on newly hired workers between 1945 and 1974 who had at
least 90 days of employment. The cohort consisted of 2,101 employees, 1,803
laborers, and 298 managers, of which vital statistics were obtained for 88% of
the group as of 1977. A comparison of lung cancer in this cohort was made
using .SMRs from the cause-specific mortality rates of Baltimore males, and was
tested for significance by means of the Poisson probability distribution.
Workers were divided into two exposure groups; the high or questionable exposure
group consisted of employees who worked in the old facilities and workers of
7-56
-------
unknown exposure, and the low exposure group consisted of workers employed only
in the new facilities. In this study, no information was available on the
actual levels of exposure or the extent of difference in exposure between the
new and old plants. Analysis of lung cancer mortality by specific jobs was
done after matching lung cancer cases by race, age, time of initial employment,
and duration of employment to employees that died from noncancer causes.
The SMR for lung cancer in the entire cohort of hourly workers was 202
(95% confidence limits of 155 to 263), which was statistically significant (P <
0.01). The SMRs for both short-term (90 days to 2 years) and long-term (>^ 3
years) workers of the high and low exposure groups are presented in Table 7-15.
There was an apparent dose-response relationship, as associated with length of
employment, for the group initially hired between 1950 and 1959. The lung
cancer SMR was statistically signficant (95% confidence limits did not include
1.00) for the workers with "high and questionable exposure" employed >3 years,
and who had initially been employed between 1950 and 1959. In analyzing lung
cancer by job description, only workers in the special products department or
employees who worked in both the special products and bichromate departments,
the so called "wet end" of the production process (the production process in
the mill and roast department is referred to as the "dry end"), showed a sig-
nificantly (P < 0.05) elevated relative risk of lung cancer of 2.6 and 3.3,
respectively.
In a study of the same plant investigated by Hayes et al. (1979), Hill and
Ferguson (1979) used a novel statistical analysis in an attempt to demonstrate
any trends in the risk of lung cancer associated with the modernization of the
plant. The statistical method used was "probability window analysis," which
provided a method of comparing the number of cases of lung cancer in equivalent
time periods. In comparing the number of lung cancers prior to 1951 (the time
7-57
-------
TABLE 7-15. OBSERVED NUMBER OF DEATHS, STANDARDIZED MORTALITY RATIOS (SMRs), AND 95% CONFIDENCE LIMITS (95% CL)
FOR DEATHS DUE TO CANCER OF THE TRACHEA, BRONCHUS, AND LUNG
AND TfaE NUMBER OF REPORTED DEATHS FOR WHICH NO CERTIFICATE COULD BE OBTAINED,
BY YEAR OF INITIAL EMPLOYMENT, EXPOSURE CATEGORY, AND TOTAL DURATION EMPLOYED,
FOR WORKERS INITIALLY HIRED AS HOURLY EMPLOYEES
(Hayes et al. 1979)
Duration of
employment"
Cause of
death
Exposure category3
Observed
no. of deaths
Low exposure
SMR (95% CL)C
Questionable and high exposure
SMR (95% CL)C
Observed
no. of deaths
INITIALLY HIRED 1945 to 1949
I
Ui
oo
Short
Long
Trachea, bronchus NA NA
and lung
Cause not NA NA
determined^
Trachea, bronchus NA NA
and lung
Cause not NA NA
determined^
20
25
13
0
1.8
(1.1
NA
3.0
(1.6
NA
to 2.7)
to 5.2)
aBased upon whether work exposure was exclusively in a new facility. See text.
bShort: 90 days to 2 years; long: ^3 years.
cCalculated using an assumption that the observed number of deaths is distributed as a Poisson random variable,
P = 0.025, in each tail.
^Those reported deceased for whom no death certificate could be obtained. If these are distributed by cause of death in
a similar way as the known deaths (cancer of the trachea, bronchus, and lung, 15%), the reported SMRs would be increased
slightly.
NA = Not applicable.
(continued on the following page)
-------
TABLE 7-15. (continued)
Duration of
employment'5
Short
-j Long
Ul
VO
Exposure category3
Low exposure Questionable
Cause of Observed Observed
death no. of deaths SMR (95% CL)C no. of deaths
INITIALLY HIRED 1950 to 1959
Trachea, bronchus 2 0.7 (0.1 to 2.6) 12
and lung
Cause not 3 NA 7
determined^
Trachea, bronchus 3 4 .0 (0 .8 to 1 1 . 7) 9
and lung
Cause not 0 NA 0
determined^
and high exposure
SMR (95% CL)C
1.8
(0.9 to 3
NA
3.4
(1.6 to 6
NA
.1)
.5)
All
Trachea, bronchus
and, lung
Cause not
determined"*
INITIALLY HIRED 1960 to 1974
NA
NA
NA
NA
aBased upon whether work exposure was exclusively in a new facility. See text.
bShort: 90 days to 2 years; long: >^ 3 years.
cCalculated using an assumption that the observed number of deaths is distributed as a Poisson random variable,
P = 0.025, in each tail.
^Those reported deceased for whom no death certificate could be obtained. If these are distributed by cause of death in
a similar way as the known deaths (cancer of the trachea, bronchus, and lung, 15%), the reported SMRs would be increased
slightly.
NA = Not applicable.
-------
of start-up of the new bichromate plant) with those after 1951, Hill and
Ferguson (1979) report a statistically significant (P < 0.01) decline from 23
to 7, and the authors suggested that this reduced risk of lung cancer resulted
from the engineering improvements that took place in 1951. Comparing the time
periods 1932-41, 1942-51, 1952-61, and 1962-71, the number of bronchogenic
carcinoma deaths falling within these time periods ("windows") are 9, 7, 1, and
0, respectively. The authors found a significant (P < 0.01) difference when
comparing the four groups under the null hypothesis that all classes have equal
probabilities of cancer mortality. The GAG would agree with the review of this
study by the IARC (1980), in which it was stated that no conclusion on improved
safety in this chromate plant could be made, since the analysis only compared
lung cancer cases and not rates of lung cancer. Also, this analysis does not
allow for the long latency period generally associated with lung cancer in the
chromate industry.
Chromate plants in both England and Japan have also been studied to
determine if there is an association between employment in this industry and
lung cancer. In an early survey, Bidstrup (1951) performed lung X-rays of 724
workers employed in the three chromate plants in England in 1949. This survey
detected only one case of lung cancer, while the expected number would have
been 0.4, as indicated from the mass radiography units of the Ministry of
Health. Of the 724 workers examined, 237 were employed for >15 years, and
although the numbers were too small for definite conclusions, Bidstrup (1951)
suggested that it was unlikely that a 25-fold incrased risk of lung cancer, as
reported in studies of United States plants, was associated with employment in
the British plants. Bidstrup's study was a cross-sectional study of only
currently employed workers, however. Since workers with lung cancer or with
symptoms of lung cancer are likely to have dropped out of the working population,
7-60
-------
it is probable that such results underestimate the difference in the incidence
or the risk of lung cancer in this group of workers.
Bidstrup and Case (1956) performed a follow-up study between 1949 and 1955
of the 723 workers who participated in the radiographic survey. During this
time, 217 workers were lost to the study as a result of change in employment.
Of the remaining men, all but 59 were alive in 1955 when the study was terminated.
The age- and cause-specific deaths of these chromate workers were compared with
the number expected based on vital statistics for the male population of
England and Wales. There was no difference in either death rates for neoplasms
other than lung cancer or deaths from other causes; however, for lung cancer,
there were 12 observed deaths with only 3.3 expected. This increase of 360%
was statistically significant (P = 0.005). There was also a trend for greater
risk in the age group <45 years (7 observed and 1.3 expected; P < 0.01, as
calculated by the CAG); however, the numbers were too small to demonstrate if
this trend was statistically significant. The effects of place of residence,
social class, and smoking habits on lung cancer incidence were considered too
small to account for this 360% increase in risk. The authors noted that this
study was of short duration and that continued follow-up of this cohort would
probably reveal an even greater increased risk of lung cancer. Of interest in
this study is the authors' report that 217 workers were lost to follow-up
because of change in employment. Change of employment may well indicate a
change in health status, and thus may suggest that the difference in lung cancer
mortality may again be underestimated.
Al.derson et al. (1981) conducted a cohort mortality study of workers at
three chromate plants in Great Britain. This was follow-up of the earlier
studies by Bidstrup (1951) and Bidstrup and Case (1956). Subjects were eligible
for this study if they had had an X-ray examination at work, had worked for a
7-61
-------
minimum of one year's continuous service, and had been employed from 1948 to
1977. Two of the plants, those at Bolton and Rutherglen, were closed in .1966
and 1967, respectively. Following these closings, production became concentrated
at the Eaglescliffe plant. The national mortality rates for England and Wales
were used to calculate expected mortality for the plants at Bolton and Rutherglen.
For all plants together, the observed lung cancer mortality in relation to that
expected was statistically significant (observed/expected = 2.419, P < 0.001).
For the individual plants, the observed/expected ratio was not significant
(observed/expected = 1.00, with 5 observed and 4.98 expected; P < 0.44) at the
Bolton plant while it was significant (P < 0.001) at the Eaglescliffe plant
(observed/ expected =» 2.156, with 36 observed and 16.20 expected) and the
Rutherglen plant (observed/expected = 2.854, with 75 observed and 26.18 expected).
It should be noted, however, that the cohort at the Bolton plant was relatively
small (202 workers)—not large enough to be reasonably able to detect a difference
between observed and expected lung cancer deaths. It should also be noted that
the observed/expected nasal cancer mortality ratio was statistically significant
(P < 0.05) at the Rutherglen plant. The authors were interested in determining
if plant modifications affected the relative risk of disease in men in the
earlier study by Bidstrup and Case (1956). The authors found that the observed/
expected ratio of lung cancer deaths decreased from 3.0 (P < 0.01) for those
who worked before the plant modifications, to 2.0 (P < 0.005) for those who
worked before and after the modifications, to 1.9 (P < 0.290) for those who
worked after the modifications were completed.
Because of the confounding factors of age distribution of the workforce,
duration of employment, duration of follow-up, and environmental influences,
the authors did a multivariate analysis of the data. For each individual, the
analysis compared the risks of developing lung cancer based on the following
7-62
-------
factors: duration of employment, duration of follow-up, calendar period of
employment, factory, age at entry, and estimated degree of chromate exposure.
The authors found the greatest contribution to the risk of lung cancer mortality
to be that of duration of employment followed by duration of follow-up, calendar
period of employment, entry age, and estimated degree of chromate exposure. In
this study the authors reported that an earlier, unpublished report of the
study cohort presented data concerning the smoking habits of 70% of these workers,
who had completed a questionnaire. The results of this survey indicated that
the percentage of heavy smokers was lower for the study cohort than that reported
for England and Wales (Todd 1962). The authors concluded that the questionnaire
responses did not provide evidence that the respondents were at greater risk of
lung cancer due to to smoking than was the population as a whole. Such a
conclusion must be considered questionable, however, considering that 30% of
the cohort did not respond. Other studies have found that the percentage of
smokers among nonrespondents to questionnaires is higher than among respondents
(Doll and Hill 1964, Criqui et al. 1979).
Watanabe and Fukuchi (1975) followed a group of 136 chromate production
workers who worked or had worked in a chromate production factory on Hokkaido
Island, Japan. The criteria for inclusion in the cohort was that the worker
must have been exposed for more than 9 years to chromium compounds. The follow-up
period was 14 years, from 1960 to 1973. Eight cases of lung cancer and two
possible cases of lung cancer were identified during the follow-up period. The
authors did not state how the ten cases were identified. Seven of these ten
cases had died during the observation period, including the two questionable
cases. The dates and causes of the deaths were ascertained by the workers'
death certificates and by hospital records. Based on national vital statistics
data, the expected number of lung cancer deaths for this population would be
7-63
-------
0.33. Thus the observed number of lung cancer cases is 21.2 times greater (if
the 2 questionable cases are counted with the observed lung cancer deaths) or
15.2 times greater (if only the five certain cases of lung cancer are counted).
Both observed numbers are significantly (P < 0.05) greater than expected,
however.
Ohsaki et al. (1978) studied the incidence of lung cancer in a cohort of
67 active chromate workers and 487 retired chromate workers at a factory on
Hokkaido Island in Japan. It is not known at this time whether this is the
same factory as was studied by Watanabe and Fukuchi (1975). The average exposure
was 22.8 years (10 to 36 years), with 133 workers exposed for >10 years. In
this population, the authors diagnosed 10 patients with lung cancer, and deter-
mined from death records that an additional four cases had occurred. The
incidence of lung cancer in these chromate workers was 658 per 100,000, as
compared to 13.3 per 100,000 for the Japanese population (P < 0.01 as calculated
by the GAG). (It is presumed that the latter rate applies to the male population,
although the authors did not state this explicitly.) No correction was made in
this study for smoking habits, even though all but two of the individuals with
lung cancer were heavy smokers.
Sano and Mitohara (1978) reported that of 36 deceased chromate workers in
Tokyo, 19 had died of cancer of the respiratory organs. Although this report
merely related case histories, the authors maintained that it supported the risk
of cancer associated with the chromate industry. Of particular interest in
this report were the total metal analyses of two workers exposed to chromium
who died of respiratory cancer. Along with excessive levels of chromium, there
were elevated levels of nickel, cobalt, beryllium, vanadium, and manganese in
the lungs, indicating exposure to other possible carcinogens.
7-64
-------
Satoh et al. (1981) conducted a mortality and morbidity study of 896 workers
engaged in the manufacture of chromium compounds for one or more years during
the period 1918 to 1975 at a plant in Tokyo, Japan. It was reported that during
the period 1934 to 1975, 84% of the chromium compounds manufactured were
hexavalent and 16% were trivalent. The plant was closed in 1975; workers were
followed until 1978 or until death. Data on the causes and dates of death were
collected from death certificates or other "reliable written testimony." In
addition to the 896 workers who were followed, 165 chromium workers were not
included in the study due to a lack of "necessary information." All of the
latter group were retired workers whose permanent residence and current status
were unknown and whose vital status as of the end of 1978 could not be determined.
Of these 165, approximately 80% had left chromium work prior to 1949, and for
65% the date of birth was unknown. The average number of years as a chromium
worker for the 165 not included in the study was about 7 years, less than the
10-year average for the 896 workers who were included in the study. The authors
analyzed the mortality data by four different time periods, 1918-49, 1950-59,
1960-69, 1970-78, and the overall time period 1918-78. They reported the
observed and expected number of deaths and SMRs for various diseases, and found
that there was an excess risk of lung cancer for each of the time periods
(mortality ratio = 9.5 for the 1918-78 time period; P < 0.005). No excess risk
of death from any other disease was found for any of the time periods or for
the overall time period. Satoh et al. also analyzed the data by length of
working experience (1-10 years, 11-20 years, and 21+ years) for the time periods
1950-59, 1960-69, and 1970-78. For lung cancer deaths, but not for deaths from
"all other cancers," the ratio of observed to expected deaths increased by
length of working experience.
7-65
-------
Satoh et al. studied morbidity by examining health insurance records for
the period from 1974 through 1977 to determine if any sickness occurred in the 81
chromium-exposed workers as compared with the 82 nonexposed workers in the plant
during that period. The authors reported that although the numbers were small,
there was no apparent difference between the exposed and the nonexposed groups
in the number of cases for any major disease category. It should be noted that
respiratory cancer and perforation of the septum are compensable and thus do
not appear in the health insurance records. Three years after the end of the
chromium exposure, all 94 workers who had been exposed for 1 to 28 years (average
14.9 years) were given a complete series of liver and kidney function tests.
All values were reported to fall within the normal range characteristic of
Japanese males.
Korallus et al. (1982) reported the lung cancer SMRs for two West German
chromate-producing plants during the period from 1948 through 1979. The popula-
tion of North Rhineland-Westphalia was used as the comparison group in calculating
the SMRs. The respiratory cancer SMRs for both plants, Leverkusen and Uerdingen,
1.92 and 2.24 respectively, were both statistically significant (P < 0.05) for
the period of the study. The authors reported that when the study period was
divided into six five-year intervals (1948-52, 1953-57, 1958-62, 1963-67,
1968-72, 1973-77) and one two-year interval (1978-79) that the respiratory
cancer SMRs for these intervals generally decreased in both plants over the
period of the study. This decrease was rather inconsistent in the Uerdingen
plant, however. Additionally, three subcohorts were defined: individuals with
beginning of exposure prior to January 1, 1948 (Group I), individuals exposed
before and after the "change in manufacture" (the change in manufacture began
in 1948 and was completed in 1957 at Leverkusen and in 1963 at Uerdingen)
(Group II), and individuals with at least half their exposure after the
7-66
-------
completion of the "change in manufacture" (Group III). The "change in manufacture"
refers to the initiation of "no-lime" processing of the chrome ore—a change
which is believed to have resulted in a reduction of the carcinogen risk. For
both plants there was a clear drop in the lung cancer SMR from Group I to Group
III. For Uerdingen, the lung cancer SMRs were 2.76, 2.60, and 0.96 for Groups
I, II, and III. For Leverkusen, the corresponding lung cancer SMRs were 2.85,
1.97, and 0.54. In neither the Leverkusen plant nor the Uerdingen plant,
however, were the SMRs for Groups I, II, or III significantly different (P <
0.05) from each other. Thus, the decrease in the respiratory cancer SMR from
Groups I to III might be due to chance. A reduction in nasal perforations,
symptomatic of chromium exposure, was also observed for Groups I to III. This
latter finding is statistically significant for both plants at P < 0.05 (chi
square test for linear trend in proportions), and would support a conclusion of
a decrease in chromium exposure from Group I to Group III.
7.2.2.2. CHROME PIGMENT WORKERS — Two morcalicy studies in which workers
were exposed only to hexavalent chromium have been conducted in the chrome pigment
industry. Langard and Norseth (1975) reported on three pigment plants in Norway
that were in operation between 1948 and 1972. One of the plants, however, was
brought on line only in 1972, the year the study ended. Between these dates,
133 workers were identified as being employed at the three plants. Of the 133,
24 had been employed >3 years, and of this cohort, six cases of cancer (three
of lung cancer and one of gastrointestinal cancer) were identified through the
Cancer Registry of Norway. All three of the lung cancer cases had been employed
5 years or longer. Two other cases, one with prostate cancer and one with
nasal cancer, were identified among the 133 workers. These cases did not
qualify for membership in the cohort employed for >3 years, however. Data from
7-67
-------
the Cancer Registry indicated an expected number of lung cancer cases among
those employed of 0.079; thus, the observed number of cases was 38 times that
expected. Exposure levels as determined by personal monitoring were reported
for the plants for the year 1972, the year in which the study ended, with
chromium levels in the two older plants ranging between 0.04 and 1.35 mg/m-',
and levels in the new plant between 0.01 and 0.08 mg/m^. The distribution of
the number of employees per year was not presented, although it was indicated
tht only seven or eight employees worked in the plants between 1948 and 1950,
with this number increasing slowly to a level of 30 workers by 1972. Although
an increased risk of lung cancer was indicated, two of the individuals with
lung cancer were moderate to heavy smokers. Nevertheless, a relative risk of
lung cancer of 38 would not be explained by differences in smoking between the
study cohort and the Norwegian population.
Davies (1978, 1979) studied three chromate pigment plants in England, of
which plants A and B produced both zinc and lead chromate, while plant C produced
only lead chromate. The cohort of exposed workers consisted of employees with XI
year of service, who were first hired between 1933 and 1967 for plant A, 1948
through 1967 for plant B, and 1946 through 1961 for plant C (these years were
governed by the availability of complete employment records), and for whom
vital statistics were available as of 1977. Using these guidelines, 396, 136,
and 114 subjects were obtained from plants A, B, and C, respectively. These
groups were further subdivided into high and medium exposure and low exposure
group's. The observed mortality from lung cancer in the different plants by
exposure category was compared to the expected mortality as calculated from
national lung cancer mortality rates for all males in England and Wales. These
data are presented in Table 7-16. The exposure categories of high and medium
were combined, the author stated, because they were "similar." No details
7-68
-------
TABLE 7-16. LUNG CANCER IN WORKERS IN THE CHROMATE PIGMENT INDUSTRY
(Davies 1979)
Plant and year of
initial employment
Plant
71 1932
S 1955
Plant
1948
Plant
1946
A
- 1954
- 1967a
B
- 1967
C
- 1967
High and medium
exposure
Number of Observed Expected
men lung cancer lung cancer
175 18b
62 0
116 7C
95 1
8.17
1.14
1.43
2.46
Low exposure
Number of Observed
men lung cancer
77 2
14 0
20 0
19 1
Expected
lung cancer
2
0.16
0.1
0.37
aPlant modification in Plant A in 1955 considerably reduced employee exposure to chromates. Thus, results
were divided into two time periods, 1932-54 and 1955-67, for the purpose of analysis.
bP < 0.01.
CP < 0.001.
-------
were given in this regard, but it is presumed the author meant that the
categories were similar with regard to the ratios of observed to expected lung
cancer deaths. Adjustments to the expected number of lung cancer death values
were made in the following manner: (1) the expected number of lung cancer
cases was adjusted upward because the proportion of unskilled and semiskilled
workers in the study population was higher than in the general population, and
these persons are known to smoke more than the national average; (2) the expected
number was adjusted downward for plant B and upward for plant C to reflect the
respective differences in local lung cancer mortality in comparison to the
national lung cancer mortality. The author did not state how these adjustments
were calculated. An elevated risk of lung cancer was present only in the com-
bined high and medium exposure group in plants A and B, while plant C, which
manufacture^ .only lead chromate, showed no elevated risk. Also, workers in
plant A who had been hired after production modification in 1955 showed no
increased risk of lung cancer, even though there was a minimum followup period
of 15 years. The authors suggested that these data indicate that zinc chromate
was associated with the etiology of lung cancer, while lead chromate was not,
and although the data were limited by the small sample size, the authors claim
that engineering controls had effectively lowered the risk of lung cancer in
plant A.
Frentzel-Beyme (1983) reported that the observed number of lung cancer
deaths exceeded those expected among workers in five chromate pigment plants in
the Netherlands and West Germany. In only one factory, however, was this excess
statistically significant. The authors did not find a lung cancer mortality
dose-response by intensity of exposure or duration of exposure. The numbers of
deaths in each exposure category were rather small, however.
7-70
-------
7.2.2.3. CHROME PLATING WORKERS — Royle (1975) studied mortality in
the chromium plating industry of England in a retrospective study between 1969
and 1972, and also reported on the first 2 years of a prospective study that
began in 1972. Workers in this industry are exposed to hexavalent chromium in
the form of chromic acid mist and some sodium dichromate dust. In the retro-
spective study, 1,238 chrome plating workers employed for >3 months were traced
along with 1,284 manual laborers used as controls. The control subjects were
matched to the exposed workers for sex and the last date they were known to be
alive, while the subgroup of workers who were currently employed in the chrome
plating industry were also matched for smoking habits. Similar success was
achieved in tracing both groups, and from the response to questionnaires, there
was little difference in the smoking habits of the groups. The death rate was
higher in the chrome plating group, with 109 deaths observed as compared to 85
in the control group. In examing cause-specific deaths, there was a significant
(P < 0.05) difference in the death rate for cancer at all sites, 3.15% in chrome
platers as compared to 1.63% in controls; while deaths from malignancy of the
lung and from malignancy of the gastrointestinal tract were each increased,
although not significantly. Other causes of death were similar in the two
groups. In a small group (220 subjects) with high exposure, there was greater
mortality associated with employment of 1 year or more as compared with exposure
for less than 1 year. The results of the first 2 years of the prospective
study indicate an increased proportionate cancer mortality with 12 cancer deaths
(the tables indicate 12, although the text reports 9) of 33 deaths reported in
the chrome-exposed group, compared with 3 cancer deaths of 19 deaths in the
control group.
In addition to the mortality study, Royle also conducted a morbidity study
of the chrome plating workers. The controls were the same as those in the
7-71
-------
mortality study. The author stated that detailed results of the morbidity
study would be published in a separate article. Among platers, it was reported
that there was a significant increase in the prevalence of a large majority of
different respiratory symptoms. The possible causes of this increase were
examined in some detail. In the course of the investigation, it was found
that a significantly larger proportion of controls (8.3%, 93 controls) than
platers (3.6%, 36 platers) had been engaged in asbestos processing. Thus, the
risk of lung cancer due to chromium exposure in platers as compared to that of
the controls may have been underestimated. The results of this study were
inconclusive because of the relatively short (3-year) follow-up period in this
retrospective study, and because results from the prospective study are only
preliminary.
Okubo and Tsuchiya (1979) conducted a cohort study of 889 Tokyo chrome
platers, with an unspecified number of controls selected from the same factories
as the chromium platers. The follow-up period was April 1, 1970 to September
30, 1976. Vital statistics were ascertained using the records of the Tokyo
Health Insurance Society of the Plating Industry. The type of work and chromium
exposure history of the members of this society were investigated by means of a
questionnaire sent to the manager of each factory. The recovery rate of the
questionnaire was 70.5%. Survival information on retired subjects was obtained
from the offices of the Japanese family registration system. Among the 889
male chromium platers, 19 deaths were observed, or about 50% of those expected.
The expected number of deaths was calculated using the annual male mortality
rates,, by age, for Tokyo. In contrast, the authors reported a slightly higher
percentage of deaths in the control group.
The authors noted that two possible types of error may have occurred in
this study: 1) deaths from the chromium group may have been incorrectly assigned
7-72
-------
to the control group, and 2) the 30% who were nonresponders may have represented
a significant number of factories in which chromium-related deaths occurred.
In addition to the problems noted by the authors, the follow-up period for
this study, six years, was probably not long enough to be able to detect
differences in lung cancer mortality resulting from chromium exposure. Also,
the size of the cohort, 889, was relatively small for the detection of significant
differences in lung cancer mortality.
Silverstein et al. (1981) found a statistically significant increase (P <
0.001) in the lung cancer proportionate mortality ratios for both male and
female white employees in a die-casting and electroplating plant. In this
plant, workers were exposed to chromium during electroplating, but nickel and
copper were also used in electroplating. The other operations of the plant were
zinc alloy die-casting and buffing, polishing, and cleaning of zinc and steel
parts. Because of the employees' exposure to other potential carcinogens, no
conclusion can be made from this study regarding the association of chromium
electroplating and lung cancer mortality.
7.2.2.4. FERROCHROMIUM WORKERS — Pokrovskaya and Shabynina (1973)
studied cancer mortality among workers at a chromium ferroalloy plant in the
Soviet Union for the period 1955-1969. Deaths among the workers were identifed
through the archives of the city registrar's office. Age-specific mortality
rates for the plant employees were compared with the corresponding age-specific
mortality rates for persons in the city where the plant was located. Persons
who were exposed to chromium in other occupational settings were excluded from
the comparison group.
Workers in the plant were reported to be exposed to low-solubility chromium
compounds. The valence of the chromium and the characteristics of the chromium
7-73
-------
exposure (dust or aerosol) varied throughout the different sections of the plant.
The greatest concentration of chromium in the air was noted during the smelting
of refined ferrochromium. In addition to chromium, the authors reported that
large amounts of "tarry substances" entered the work area, together with furnace
gases, during the smelting process. These substances included benzo[a]pyrene,
which is formed during the caking of the electrode mass. The highest concentrations
of tarry substances were observed at the work stations of electrode workers; a
significantly lower concentration was observed at the work stations of smelters
and batchers.
The mortality ratios for cancer of all sites and for lung, stomach, and
esophageal cancer for the ferroalloy production workers are reported in Table
7-17.
TABLE 7-17. MORTALITY RATIOS RESULTING FROM MALIGNANT TUMORS AMONG
WORKERS IN CHROMIUM FERROALLOY PRODUCTION
(adapted from Pokrovskaya and Shabynina 1973)
Relative risks of cancer mortality
All sites Lungs Stomach
Esophagus
Age groups
30-39
40-49
50-59
60-69
Males
2.6
0.5
3.3a
2.0
Females Males Females Males Females Males Females
2.8 4.4 - 3.8 -
1.3 - - 0.8 4.0
7.9 6.6a - 3.2 - 2.0a
- - - - 11.3a
aP=0.001.
As can be seen from the table, the mortality ratio for all malignant tumors was
higher for the ferroally workers in almost all age groups of both sexes. The
7-74
-------
mortality ratio (MR) for lung cancer was significantly higher (MR = 6.6, P - 0.001)
among males in the 50-59 year age group and was 4.4 among males in the 30-39
year age group. (The statistical significance of this latter mortality ratio
was not reported.) The mortality ratio for esophageal cancer was significantly
higher (P = 0.001) among males in the 50-59 and 60-69 year age groups (RR = 2.0
and 11.3, respectively).
The average age of workers in ferrochromium production who died of cancer
was reported to be significantly lower than that of the city residents who died
of cancer—49.4 and 62.9 years, respectively. A higher level of cancer mortality
was also reported among workers subjected to the greatest dust load (charge
loaders, metal breakers, and smelters). The authors reported a high mortality
rate from cancer among workers in the charge preparing and finishing sections,
where high chromium-containing dust pollution was observed but no exposure to
benzo[a]pyrene was found. This study appears to be merely a comparison of
cancer mortality between the ferroalloy plant workers and the local population
for the time period 1955-1969. Very little detail was provided by the authors.
No cohort was defined. The number of person-years and the number of cancer cases
among the ferroalloy workers were not reported; nor were the number of cases and
and the number of individuals in the comparison population. Because of the
sketchiness of the reporting and the lack of an adequately defined cohort, the
results of this study are open to question. In addition, if a high turnover
rate existed among employees of the plant, the results of the study could be
misleading.
Langard et al. (1980) studied ferrochronu'iini workers at a ferrochromium
plant in Norway who were predominantly exposed to trivalent chromium. In an
industrial hygiene study of the plant in 1975, ambient chromium levels of
between 0.01 and 1.34 mg/m^ were detected in the ferrochromium department, and,
7-75
-------
of this chromium, it was determined that 11% to 33% was hexavalent. The study
consisted of 976 employees who had worked for >1 year, were alive after 1953,
and had been initially employed prior to 1960. The study was divided into 10
subgroups by job description, with only 325 subjects specifically associated
with ferrochromium production and considered exposed to chromium. Comparison
of cancer incidence for all sites and for different sites were made between the
Norwegian male population and the ferrochromium worker population, using the
Norwegian Cancer Registry. Foisson distribution was used to test for statistical
significance. A comparison of total mortality was also made. Of the 10 job-
related subgroups, only the ferrochromium workers had a significant difference
between the observed and expected number of "lung" cancer cases (7 observed
versus 3.1 expected, P < 0.05). (Note: The statistical significance was calculated
by Langard et al. to be P = 0.08. The GAG tested the difference using the Poisson
distribution, and found the significance to be P < 0.05). There is a problem
with the authors' use of the term "lung cancer." Although the authors primarily
refer to "lung cancer" in the text, they also use the terms "cancer of the
respiratory tract" and describe the "lung cancer" cases in one of the tables as
being of International Classification of Disease (ICD) codes 162 and 163. ICD
codes 162 and 163 include cancer of the respiratory tract other than lung
cancer. This ambiguity raises some question as to the authors' comparisons of
observed and expected cases. If the expected number of cases is calculated for
ICD codes 162 and 163, and the observed number of cases are only lung cancer
cases', then the relative risk of lung cancer has been underestimated. If the
opposite is true, then the risk has been overestimated. If the observed number
of "lung cancer" cases mistakenly includes cases of tnesothelioma, then the
stated "lung cancer" risk due to chromium may instead be a reflection, at least
7-76
-------
partially, of asbestos exposure. This ambiguity must therefore be considered
when interpreting the results.
Total mortality and cancer incidence at all sites were similar to the
general population. The relative risk of "lung cancer" among the ferrochromium
workers as compared to the Norwegian male population may have been underestimated,
however, since the county in which the plant was located, Hordaland, had an age-
adjusted "lung cancer" incidence that was 58% of Norway's "lung cancer" incidence
rate. If this 58% is multiplied by the expected number of "lung cancer" cases
calculated from national rates, the newly calculated expected number differs
significantly from the observed number of cases (P < 0.01). If non-chromium-
exposed workers in this plant had been used as a control population, the risk
of "lung cancer" in chromium-exposed workers would have increased to 8.5, which
is significant at P < 0.01. The ferrochromium workers were possibly exposed to
two other carcinogens, asbestos and polycyclic aromatic hydrocarbons; however,
this was unlikely to have been an important factor, the authors indicated, since
a group of 243 ferrosilicon workers in the same plants were believed to have been
exposed to these two known carcinogens to the same degree, and no increased
risk of "lung cancer" (0 observed and 2.78 expected) was observed among these
workers. However, the sample size for the ferrosilicon workers (number = 243)
would have to be considered too small to be able to detect any significant
excess of "lung cancer."
In studies of the ferrochromium industry in Sweden, Axelsson et al. (1980)
concluded that there was no association between employment in the ferrochromium
industry and risk of respiratory cancer. The study cohort consisted of 1,876
men who had been employed for one or more years between 1930 and 1975, and who
were alive as of 1951. Observed cases were included for comparison if they
occurred 15 years after first exposure. The exposed population was compared
7-77
-------
with males in the county in which the plant was located, and comparisons were
tested for significance by two-sided P-values using the Poisson distribution.
The workers were subdivided into four groups consisting of arc-furnace workers;
workers in transport, metal grinding, and sampling; maintenance workers; and
office workers. It was estimated that these groups had been respectively
exposed to 2.5, 0.5 to 2.5, 2.5, and 0 mg/rn^ of trivalent chromium and elemental
chromium, and 0.25, 0.01 to 0.05, 0.05, and 0 mg/m^ of hexavalent chromium,
respectively. Medical examination of employees during the last 3 years of the
study detected three cases of perforated septum of the nose, suggesting that
some exposure to hexavalent chromium had occurred in at least a portion of the
work force. No statistically significant difference in cancer mortality between
that observed in the workers and that expected based on national mortality data
was found. The only significant increase in cancer incidence was an increase
in respiratory cancer (4 observed and 1 expected, P = 0.038) in the 315 maintenance
workers. Of the four cases of respiratory cancer, two were diagnosed as meso-
theliomas, and the authors suggested that these cases may have resulted from
exposure to asbestos. It was also noted that the control population was mainly
rural dwellers, while the exposed group consisted of urban dwellers, and that
rural residents in Sweden generally smoked less. From the results of this
study, the authors concluded that no association existed between exposure to
predominantly trivalent chromium and elemental chromium and the development of
cancer. Because of the confounding effects of smoking and exposure to asbestos,
no definite conclusions can be drawn from this study.
7-78
-------
7.2.3. Quantitative Estimation. This quantitative section deal t,
v
risk for chromium in air and the potency of chromium relative to H
gens that the GAG has evaluated. The unit risk estimate for an
is defined as the incremental lifetime cancer risk over the back,
in a hypothetical population in which all individuals are exposed continuously
to a concentration of 1 ug/m3 of the agent in the air that they breathe. This
calculation is done to estimate in quantitative terms the impact of the agent
as a carcinogen. Unit risk estimates are used for two purposes: 1) to compare
the carcinogenic potency of several agents with each other, and 2) to give a
crude indication of the population risk which might be associated with exposure
to these agents, if the actual exposures are known.
The data used for quantitative estimation can be of two types: 1)
lifetime animal studies, and 2) human studies where excess cancer risk has been
associated with exposure to the agent. It is assumed, unless evidence exists
to the contrary, that if a carcinogenic response occures at the dose levels
used in a study, then responses will also occur at all lower doses, with an
incidence determined by an extrapolation model.
There is no solid scientific basis for any mathematical extrapolation
model that relates carcinogen exposure to cancer risks at the extremely low
concentrations that must be dealt with in evaluating environmental hazards.
For practical reasons such low levels of risk cannot be measured directly either
by animal experiments or by epidemiologic studies. We must, therefore, depend
on our current understanding of the mechanisms of carcinogenesis for guidance
as to which risk model to use. At the present time the dominant view of the
carcinogenic process involves the concept that most cancer-causing agents
also cause irreversible damage to DNA. This position is reflected by the fact
that a very large proportion of agents that cause cancer are also mutagenic.
7-79
-------
A is reason to expect that the quantal type of biological response, which
is characteristic of mutagenesis, is associated with a linear non-threshold
dose-response relationship. Indeed, there is substantial evidence from
mutagenesis studies with both ionizing radiation and a wide variety of chemicals
that this type of dose-response model is the appropriate one to use. This is
particularly true at the lower end of the dose-response curve; at higher doses,
there can be an upward curvature, probably reflecting the effects of multistage
processes on the mutagenic response. The low-dose linearity and nonthreshold
dose-response relationship is also consistent with the relatively few epidemiologic
studies of cancer responses to specific agents that contain enough information
to make the evaluation possible (e.g., radiation-induced leukemia, breast and
thyroid cancer, skin cancer induced by arsenic in drinking water, liver cancer
induced by aflatoxins in the diet). There is also some evidence from animal
experiments that is consistent with the linear nonthreshold model (e.g., the
initiation stage of the two-stage carcinogenesis model in rat liver and mouse skin).
Because its scientific basis, although limited, is the best of any of the
current mathematical extrapolation models, the linear nonthreshold model has
been adopted as the primary basis for risk extrapolation to low levels of the
dose-response relationship.
' The quantitative aspect of carcinogen risk assessment is included here
because it may be of use in the regulatory decision-making process, e.g.,
setting regulatory priorities, evaluating the adequacy of technology-based
controls, etc. However, it should be recognized that the estimation of cancer
risks to humans at low levels of exposure is uncertain. At best, the low-dose
linearity extrapolation model used here provides a rough but plausible estimate
of the upper limit of risk; i.e., it is not likely that the true risk would be
much more than the estimated risk, but it could very well be considerably lower.
7-80
-------
The risk estimates presented below should not be regarded as accurate representa-
tions of the true cancer risks even when the exposures are accurately defined.
The estimates presented may, however, be factored into regulatory decisions to
the extent that the concept of upper risk limits is found to be useful. ^___
There are many epidemiologic studies demonstrating that hcxavalent chromium
(Cr VI) is a potential human carcinogen, but few of these studies provide adequate
exposure data for use in risk estimation. One study by Mancuso (1975) provides
what the GAG feels is limited but adequate information for this purpose, however.
Mancuso1s data is used as the main data base for estimating the carcinogenic
potency of hexavalent chromium. For comparison, data from three foreign studies
on ferrochromium plants are also used in the potency calculations. From the
quantitative risk assessment viewpoint, these studies are less adequate than
the Mancuso study. For the Norwegian study (Langard et al. 1980), the exposure
measurements were taken in 1975, while some workers could have been exposed to
chromium as early as 1928, when the ambient dust levels were much higher than
in recent years. For the Swedish study (Axelcson et al. 1980), the chromium-
exposed workers did not show a significant increase of lung cancer, and thus
only the statistical upper bound of the response can be used in potency estimation.
For the above reasons, it is expected that the use of data from the Norwegian
and Swedish studies would result in an overestimation of the true carcinogenic
potency of hexavalent chromium. While the Russian study (Pokrovskaya and
Shabynina 1973), does not have the deficiencies of the other two foreign studies,
the cohort in this study is not well defined, and thus the validity of the data
reported is open to question. In an effort to provide alternative potency
estimates, the data from these less adequate studies have also been used by the
CAG to calculate the potency of hexavalent chromium. The potency calculated on
the basis of Mancuso1s data is consistently smaller than (but within 5 folds of
magnitude of) the lower-limit estimates calculated from the three foreign studies.
7-81
-------
Animal data from intratracheal studies have not been used to estimate the
carcinogenic potency of chromium by inhalation because there is no pharmacokinetic
information relating the distribution of chromium to lung tissues by inhalation
and by intratracheal administration. This information is needed because the
dose distribution is clearly different between these two exposure patterns.
Furthermore, the physiological mechanism of dose distribution by intratracheal
administration may depend (in a non-linear fashion) on the dose levels used in
the experiment, as evidenced by the observation that a single administration of
sodium dichromate induced a carcinogenic response in Sprague-Dawley rats but
failed to induce a response when the same weekly dose was given over 5 days.
7.2.3.1. ESTIMATION OF THE CARCINOGENIC PPTENCY OF HEXAVALENT CHROMIUM
(Cr VI) BASED ON THE MANCUSO (1975) DATA -- The Mancuso study was based on a
cohort of 332 white male workers who were employed in a chromate plant between
1931 (when the plant began to operate) and 1937, and who were followed to 1974.
In his study, Mancuso reported lung cancer death rates by levels of exposure to
soluble, insoluble, and total chromium concentrations. Because only lung
cancer mortality for total chromium exposure was reported by age group, the GAG
has used only the dose-response data for total chromium to estimate the carcino-
genic potency of hexavalent chromium. Although the use of dose-response data
for total chromium will result in an underestimation of the potency of hexavalent
chromium, we believe that the effect of this underestimation is approximately
compensated for by other factors that may overestimate the risk. These issues
will be discussed further in the pages that follow.
Exposure information in the Mancuso study was derived from an industrial
hygiene study of the plant conducted in 1949. In this study, time-weighted
averages of exposure to insoluble, soluble, and total chromium per cubic meter
7-82
-------
were calculated for each occupation and for each worker in every department.
Using these data and company personnel records, Mancuso was able to calculate
an estimate of exposure to soluble, insoluble, and total chromium by duration
of exposure (in mg/nH-years) for each member of the 1931-37 cohort. In 1949,
after the industrial hygiene study had been conducted, the company initiated a
comprehensive program designed to reduce employees' exposures and improve
manufacturing efficiency. Until that time, however, the company had not under-
taken any programs for the purpose of reducing employee exposure. It should be
noted, however, that Bourne and Yee (1950), who conducted the industrial hygiene
survey in 1949, reported that "in order to meet price and quality competition,
improvements in equipment and processes have been made periodically during the
past 18 years, and it is the universal experience of industrial hygiene personnel
that greater process efficiency is almost invariably associated with a more
healthful working environment. Therefore, there seems little doubt that
atmospheric contamination in the past was greater than in 1949." Nonetheless,
no concerted effort was made to reduce employee exposure until late in 1949,
and because this particular plant was a relatively modern one at the time of
the industrial hygiene survey, it is unlikely that improvements in efficiency
over the period 1931 to 1949 would have reduced employee exposure to a great
extent. Thus, the GAG considers Mancuso"s data to be a reasonable approximation
of what workers in the study cohort were exposed to during their entire working
history. The GAG recognizes the possibility that the exposure may be slightly
underestimated because of the likelihood that a greater proportion of the
"total exposure" was contributed prior to 1949 than post-1949.
The effects of underestimating the exposure concentration, as well as the
effects of other uncertainties on the estimation of potency, will be addressed in
the Discussion section.
7-83
-------
7.2.3.1.1. Data Available for Potency Calculations — Table 7-18,
which is taken directly from Mancuso (1975), presents age-specific lung cancer
deaths, corresponding person-years, and range of exposures to total chromium.
To estimate the lifetime cancer risk due to exposure to chromium, it is
assumed that an exposure, D (rag/m^ -years) , as presented in Table 7-18, is
equivalent to the continuous exposure d (ug/nr) calculated by
x _i x 240 x 103 Ug/m3
_P
fLe 24 365
where Le is the midrange in each age category, f is the fraction of time in age
exposed, and 8/24 and 240/365 are the fractions of a day and year, respectively,
o
that a worker spent at the plant. For instance, if D = 8 mg/m -years, Lg = 60,
and f = 0.65, then d = 44.96 ug/m3. The assumption of f « 0.65 implies that
the cohort exposure to chromium began approximately at age 20. The assumption
is that the particular exposure pattern (unknown to us) leading to the cancer
mortality rates as observed is equivalent to the continuous constant exposure
starting from the age when exposure began. This assumption may or may not be
realistic. However, it would be more unrealistic to make a different assumption
concerning the exposure pattern when all that is given is an exposure which
itself was calculated by taking the weighted average of the duration of exposure
for each respective job the worker had.
Since the person-year in each category presented in Table 7-18 is very
small, the exposure categories are combined as shown in Table 7-19 to increase
statistical stability. The last column of Table 7-19 is given for the purpose
of identifying which exposure categories in Table 7-18 are combined. The
midrange of age and exposure concentration is used in Table 7-19. Data in this
table are used to estimate the lifetime cancer risk due to chromium exposure.
7-84
-------
TABLE 7-18. AGE-SPECIFIC LUNG CANCER DEATHS AND GRADIENT EXPOSURES TO TOTAL CHROMIUM
(Mancuso 1975)
Age
<1.00
Exposure to total chromium (mg/m^-year)
1.0-1.99 2.0-3.99 4.0-5.99 6.0-6.99
7.0-7.99
8+a
45-54 Deaths
i
00
Ln
Person-years
886
459
583
348
159
140
262
55-64 Deaths
Person-years 707
356
462
250
113
98
203
65-74 Deaths
Person-years 235
166
182
80
42
41
81
aData in the last column are not used in the CAG1s risk assessment because the range of exposure in this class is
not known, and it does not appear reasonable to assume that all three age groups had an identical exposure
distribution in this class.
-------
TABLE 7-19. COMBINED AGE-SPECIFIC LUNG CANCER DEATH RATES AND TOTAL CHROMIUM EXPOSURE (in ug/m3)
I
oo
Age
50
50
50
60
60
60
70
70
Concentration
(ug/m3)a
5.66
25.27
46.83
4.68
20.79
39.08
4.41
21.29
Deaths
3
6
6
4
5
5
2
4
Person-years
1345
931
299
1063
712
211
401
345
Background
rateb
6.05
6.05
6.05
1.44
1.44
1.44
1.57
1.57
x 10"4
x 10"4
x 10"4
x 10~3
x 10~3
x 10~3
x 10~3
x 10~3
Exposure range as presented
in Table 7-18
<_ 1
2.0
6.0
<_ 1
2.0
6.0
_< 1
2.0
.99
- 5.99
- 7.99
.99
- 5.99
- 7.99
.99
- 7.99
formula described in the section "Data Available for Potency Calculations." The concentrations presented
in this table are the averages of several exposure categories weighted by corresponding person-years.
^Background rate is estimated from 1964 U.S. Vital Statistics. The year 1964 is selected because it is
estimated that a large proportion of lung cancer deaths occurred during that year.
-------
7.2.3.1.2. Choice of Dose-Response Model — It has been widely recognized
(e.g., Doll 1971) that the age-specific incidence curve tends to be linear on
doubly logarithmic graphs, or equivalently, the age-specific incidence follows the
mathematical form
where b and k are parameters that may be related to other factors such as dose,
and T may be one of the following three cases:
1. T is age when cancer is observed,
2. T is the time from the first exposure to observed cancer, or
3. T is the time from exposure to cancer minus the minimum time for a
cancer to be clinically recognized.
This model has been shown to arise from the somatic mutation hypothesis of
carcinogenesis (Armitage and Doll 1954, Whittemore 1978, Whittemore and Keller
1978). It has also been shown to arise from the epigenetic hypothesis when the
reversible cellular change is programmed to occur randomly (Watson 1977).
These authors and many others have used this model to interpret and/or estimate
potency from human data.
Since the data that could be used for risk estimation are limited, a simple
model that fits the data should be used. Therefore, the observed age-specific
incidence is assumed to follow the model
I(t,d) = B(t) + h(t,d)
where B(t) is the background rate at age t and h(t,d) = Q(d) tk-1 with
r\
Q(d) = q^d + q2d , a function of dose d.
Once the parameters qi , q2 , and k are estimated, the lifetime cancer risk
associated with an exposure d by age t, taking into account the competing risk,
can be calculated by
7-87
-------
t S
P(t,d) = f h(s,d)exp r-[ f h(y,d)dy + A(s)] Ids
oo
where exp[-A(s)] is the probability of surviving to age s and h(t,d) =
I(t,d) - B(t), the age-specific incidence after adjusting the background rate.
7.2.3.1.3. Estimation of the Risk Model — To estimate the parameters in
h(t,d) we assume, as is usually done, that the number of lung cancer deaths, X,
at age t, follows the Poisson distribution with the expected value
E(X) = N x (B + Q(d) tk-1)
where N is the person-year associated with X, B is the background rate at age
f\
t, and Q(d) = q^d + q2d .
Using the BMDP computer program P3R and the theory relating the maximum
likelihood and non-linear least square estimation by Jennrich and Moore (1975),
the parameters qj, q2, and k are estimated by the method of maximum likelihood
—7 —9
as q1 = 1.11 x 10 , q2 = 1.84 x 10 , and k » 2.915; the corresponding standard
deviations are respectively 7.8 x 10~7, 1.2x10"^, and 1.7.
Thus, the age-specific cancer death incidence at age t due to chromium
exposure d ug/m^ is given by
h(t,d) = Q(d) tl-915
where
Q(d) = 1.11 x 10~7d + 1.84 x 10~9d2
The model fits the data well, as can be seen from the goodness of fit
statistic
X2 = j. (0-E)2/E =1.60
7-88
-------
which has, asymptotically, a chi-square distribution with 5 degrees of freedom
under the model specified. The observed and predicted values used in calculating
X2 are (3, 2.5), (6, 7.2), (6, 5.1), (4, 3.1), (5, 6.7), (5, 4.1), (2, 1.4)
and (4, 4.3).
Taking into account the competing risk, the lifetime probability of lung
cancer death due to exposure to chromium d ug/m3 is given by
L
P(L,d) = / h(t,d)exp |-[(Q(d)/2.915) t2«915 + A(t)] }dt
o
where L is the maximum human lifetime and is mathematically equivalent to
infinity, since the probability of surviving beyond L is 0.
At low doses, approximately,
P(L,d) = d x P(L,1)
where P(L,1) is the lifetime cancer risk due to exposure to 1 ug/m3 of
chromium. The unit risk, P(L,1), has been adopted by the GAG as an indicator
of the carcinogenic potency of a chemical compound.
7.2.3.1.4. Calculation of the Risk at 1 ug/m3 — To calculate the unit
risk, P(L,1), it is necessary to know exp[-A(t)], the probability of surviving
to age t. Since this probability can only be estimated, it is assumed that the
survival probability is constant over a 5-year interval, as provided in the U.S.
Vital Statistics.
Using this approximation and by integrating the formula P(L,1), we have
P(L,1). = £[exp(-3.87 x 10~8 t^2*915) - exp(-3.87 x 10~8ti2*915) ] x P£
- 1.16 x ID"2
where (t^_j, t^) is a 5-year interval and P£ is the probability of survival up
7-89
-------
to the age t£_i. P£ is assumed to be a constant over the interval and is
estimated from the 1975 U.S. Vital Statistics.
7.2.3.1.5. Alternative (Crude) Approach for Calculating the Carcinogenic
Potency of Chromium from the Mancuso (1975) Data — As a crude approximation, the
carcinogenic potency of chromium can be calculated by B = (R-l) x P0/d, where
PQ = 0.036 is the estimated lung cancer mortality rate for the U.S. population,
R is the relative risk of the lung cancer deaths in the cohort, and d is the
"standardized" lifetime dose concentration to which the workers were assumed to
be exposed. This approach is used by the GAG to calculate carcinogenic potency
when the only data available are the relative risk estimate and an average
exposure concentration.
For the Mancuso (1975) data, the relative risk R and the "standardized"
dose d are estimated respectively to be R = 7.2 and d = 15.5 ug/m3. They are
calculated by combining the relative risks and dose concentrations in each of
the age-exposure categories, weighted by the relative magnitude of person-years,
as shown in Table 7-19.
Therefore, the carcinogenic potency of hexavalent chromium (Cr VI) is
estimated to be
B = (7.2-1) x 0.036/15.5 = 1.4 x 10-2/ug/m3.
This crude estimate is only slightly higher than the previous estimate,
1.2 x 10-2/ug/m3.
7.2.3.1.6. Discussion — The following discussion is intended to provide
some insight about the uncertainties of estimating the carcinogenic potency of
hexavalent chromium on the basis of the Mancuso (1975) data.
1. As noted previously, the risk of hexavalent chromium is estimated on
the basis of the total chromium obtained from all the soluble and insoluble
7-90
-------
chromium to which workers were exposed. Since only some compounds of hexavale..
chromium are^known to be carcinogenic, the potency presented above is likely to
be underestimated. This underestimation seems unlikely to be more than sevenfold
(or smaller) on the basis of Bourne and Yee (1950). Bourne and Yee reported
that the ratios of trivalent chromium (Cr III) and hexavalent chromium (Cr VI)
concentrations in the airborne dust in nine major departments in the plant
ranged from 1 to 3 except for two departments where the ratios were 6 for the
lime and ash operation and 52 for the ore preparation. Excluding the ore
operation, the maximum ratio of trivalent chromium and hexavalent chromium is
6, and thus the underestimation of the risk would not be more than sevenfold.
2. As indicated previously, there is a possibility that the use of 1949
hygiene data may result in a slight underestimation of what workers were actually
exposed to. However, because the plant was relatively modern at that time, the
underestimation is unlikely to be very considerable. If an underestimation of
2 times were assumed, then the unit risk would be reduced from 1.2 x 10~^/ug/m^
to 6.0 x 10~3/ug/m3.
3. The risk presented in the present report may be somewhat overestimated in
the sense that it is implicitly assumed that the smoking habits of chromate workers
were similar to the general white male population, while it is generally accepted
that the proportion of smokers is higher for industrial workers (thus the higher
background incidence rates) than for the general population. As a sample
calculation, it was found that if the background rate of lung cancer mortality for
the cohort in Table 7-19 is increased by 40%, then the corresponding unit risk
would be reduced by about 25%, or from 1.2 x 10"^ to 8.7 x 10~3.
The background age-specific rate of lung cancer at ages 50, 60, and 70
could be 40% more than those presented in Table 7-19, should it be assumed that
80% of chromate workers are ever-smokers (individuals who smoke at least 100
7-91
-------
cigarettes during their lifetimes) and only 50% of the general white male
population are ever-smokers. It could also result from other assumptions about
the smoking habits of the chrornate workers and the general population. For
instance, the background rate could be increased by approximately 40%, based on
the age-specific cancer rates provided by Doll (1971), if the proportion of
smokers in each smoking level is distributed as follows:
NUMBER OF CIGARETTES SMOKED PER DAY
0
Chromate workers 0 .3
General population 0.5
1-14
0.2
0.2
15-24
0.3
0.2
25 or more
0.2
0.1
Other reasonable assumptions about the smoking habits of the cohort workers would
not reduce the risk estimate by more than 50%. Therefore, it is unlikely the
<*"N
f risk is overestimated by more than four times, considering both the smoking
(^habits and the underestimates of exposure by a factor of two.
7.2.3.2. POTENCY ESTIMATION BASED ON LANGARD ET AL. (1980) — The
study population consisted of 976 employees of a ferrochromium plant in Norway
who worked for at least one year, were alive after 1953, and were initially
employed prior to 1953. All members of the study population were considered
"under observation" from the beginning of 1953 until 1977 inclusively, or in the
case bf later initial employment, from one year after employment in the plant.
A subgroup of 325 workers were identified as specifically associated with
ferrochromium production and considered exposed to chromium. The relative risk
of lung cancer in chromium-exposed workers was estimated to be 8.5 when non-
chromium-exposed workers in the same plant were used as a control population.
7-92
-------
The authors considered the use of this control group more appropriate than the
other reference populations because both chromium-exposed and non-chromium-
exposed workers were similar with respect to the factors that could influence
lung cancer response. The chromium concentration to which the workers were
exposed is not available. In a complete industrial hygiene survey carried out
during 1975, the ambient chromium levels in the plant were found to be between
0.01 mg/m3 and 1.34 mg/m3, and of this chromium, it was determined that 11% to
33% was hexavalent. Clearly, the measurements taken in 1975 underestimate the
actual chromium ambient levels to which most of the workers were exposed.
These concentrations are used in our potency calculations, with the understanding
that the potency so estimated can only be considered an upper-bound estimate.
Assuming that the hexavalent chromium content in the sample was 19% (i.e.,
the geometric mean of 11% and 33%), the hexavalent chromium ambient concentrations
ranged from 1.9 ug/m3 to 254.6 ug/m3. These concentrations are "standardized"
to the lifetime exposure, 0.1 ug/m3 and 14.0 ug/m3, as calculated by the formula
d x (8/24) x (240/365) x (£/L)
where d is the ambient level (either 1.9 or 254.6 ug/m3); values (8/24) and
(240/365) reflect that a worker worked 8 hours per day, 240 days for a year; and
the factor (fc/L) represents the ratio of time in years an "average" worker was
exposed and the average age of the cohort when the study terminated. In the
absence of detailed information, the ratio (£/L) is assumed to be 1/4, as
usually estimated in other cohort studies.
The carcinogenic potency of hexavalent chromium (Cr VI) is calculated to
range from
B = (8.5-1) x 0.036/14.0 - 1.9 x 10~2/ug/m3
7-93
-------
to
B = (8.5-1) x 0.036/0.1 = 2.7/ug/m3.
The geometric mean of these two limits is 0.23/ug/m-^. It is noted that even the
lower range of the estimate (1.9 x 10~^/ug/m^) is greater than the potency estimate
calculated from the Mancuso (1975) data (1.2 x 10""2/ug/m3).
7.2.3.3. POTENCY ESTIMATION BASED ON AXELSSON ET AL. (1980) — The study
cohort consisted of 1,876 men who were employed for one or more years between
1930 and 1975 and who were alive as of 1951. The chromium-exposed population was
compared with males in the county in which the plant was located. The workers
were classified into four groups consisting of (1) arc-furnace workers; (2) workers
in transport, metal grinding, and sampling; (3) maintenance workers; and (4)
office workers. The study did not show a significant increase of respiratory
cancer in the exposed population, except in the case of the maintenance workers.
Four respiratory cancers were observed in the group of 315 maintenance workers
versus one expected. Of the four cases of respiratory cancer, two were diagnosed
as mesotheliomas which the authors suggested may have resulted from exposure to
asbestos. Excluding the two cases of mesotheliomas, the relative risk for lung
cancer is 2, which is not statistically significant. To utilize the data from a
negative study, the 95% upper limit of the relative risk, 3.7, is used to
calculate an upper-bound estimate of potency.
The ambient levels of hexavalent chromium to which the maintenance workers
were exposed was estimated to be 50 ug/m-*. The authors cautioned against the
use of this exposure estimate. However, the authors did not indicate how the
estimate was obtained and did not provide hints as to whether this estimate is
likely to be higher or lower than the true ambient levels.
7-94
-------
Using the relative risk of 3.7 and the ambient level of 50 ug/m3, the
carcinogenic potency of hexavalent chromium is estimated to be
B = (3.7 - 1) x 0.036/2.7 - 3.5 x 10~2/ug/m3
where the lifetime dose 2.7 ug/m3 is calculated by
50 x (8/24) x (240/365) x (1/4) = 2.7 ug/m3.
7.2.3.4. POTENCY ESTIMATION BASED ON POKROVSKAYA ET AL. (1973) —
Although this study showed a significant increase of lung cancer mortality over
the control group, the validity of the data is questionable because the study
cohort is not clearly defined. The report indicates that the cancer mortalities
over the period 1955-1969 in workers from a ferroalloy plant in the Soviet Union
were compared with the population of similar ages in the city where the plant
was located, but it fails to indicate the criteria by which workers were
included in the cohort. The lung cancer mortality ratios were reported to be
4.4 for the age group 30-39 and 6.6 for the age group 50-59 among male workers.
Concentrations of hexavalent chromium wf>re reported to exceed the marginally
allowable value (0.01 mg/m3) by 2 to 7 times on the average. The length of
employment was from 7 to 20 years, with an average of 15 years.
Based on the information that the average ambient concentrations of hexava-
lent chromium exceeded the marginally allowable value 0.01 mg/m3 by 2 to 7
times, workers' exposure to hexavalent chromium ranged from 0.02 mg/m3 to 0.07
mg/m3. The lifetime doses corresponding to 0.02 mg/m3 and 0.07 mg/m3 are,
respectively, as follows:
dl = 0.02 x 103 x (8/24) x (240/365) x (1/4) =1.1 ug/m3
7-95
-------
and
d2 = 0.07 x 103 x (8/24) x (240/365) x (1/4) = 3.8 ug/m3
If 6.6 is taken to be an estimate of the average relative risk for the cohort,
then the carcinogenic potency for hexavalent chromium (Cr VI) is calculated to
range from
B = (6.6-1) x 0.036/3.8 = 5.2 x 10~2/ug/m3
to
B = (6.6-1) x 0.036/1.1 = 0.18/ug/m3
The geometric mean of the two limits is 9.7 x 10~^/ug/m3. It is about 8 times
larger than 1.2 x 10~^/ug/m3, the potency calculated on the basis of the Mancuso
(1975) data.
7.2.3.5. COMPARISON OF UNIT RISK ESTIMATES ON THE BASIS OF DIFFERENT HUMAN
STUDIES — The carcinogenic potency, B, calculated on the basis of data from the
three foreign studies cited above, can be used to calculate unit risks by means
of the formula P(d) = l-exp(-B x d) with d = 1 ug/m^. When B x d is small,
P(d) can be approximated by B x d. The results of these calculations, along
with the unit risk estimate of Mancuso (1975), are presented in Table 7-20.
Lower, best, and upper limits are provided. These limits reflect various
uncertainties associated with the calculation of potency estimates. For the
Mancuso study, the lower limit represents the uncertainties associated with
possible underestimation of the chromium exposure levels and the smoking habits
of the cohort workers. It is assumed that the risk calculated without correcting
these factors is overestimated by four times. The upper limit for the Mancuso
study is calculated by assuming that the ratio of trivalent chromium (Cr III) and
hexavalent chromium (Cr VI) is 6, and thus the risk is underestimated by seven
7-96
-------
times. For Langard et al. (1980) and Pokrovskaya et al. (1980), the "best"
estimate is the geometric mean of the lower and upper limits, which correspond,
respectively, to the maximum and minimum hexavalent chromium ambient levels
reported. For Axelsson et al., the statistical upper limit (95% confidence
limit) is used as the best and upper-limit estimate because the study was negative.
TABLE 7-20. COMPARISON OF UNIT RISKS (LIFETIME RISK DUE TO
1 ug/m3 OF HEXAVALENT CHROMIUM IN AIR)
Data base
Mancuso (1975)
Langard et al. (1980)
Axelsson et al.
(1980)
Pokrovskaya et al.
(1973)
Lower limit
3.0
1.9
Not
5.2
x 10"3
x 10-2
available
x 10-2
Best
1.2
1.3
3.5
9.2
estimate
x 10-2
x 10-1
x 10~2
x 10-2
Upper
8.4 x
9.3 x
3.5 x
1.6 x
limit
10-2
10-1
10-2
io~i
7.2.3.6. RELATIVE CARCINOGENIC POTENCY OF CHROMIUM ~ Figure 7.1 is a
histogram representing the frequency distribution of the potency indices of 53
suspect carcinogens evaluated by the CAG. Table 7-21 presents the potency
index for these 53 suspect carcinogens. The potency index for a compound is a
rounded-off number expressed in terms of (mMol/kg/day)"l. Where human data are
available for an agent, they have been used to calculate the index. Where no
human"data are available, animal oral studies have been used in preference to
animal.inhalation studies.
Based on the occupational study by Mancuso (1975) and the assumption that
the daily air intake of a 70-kg man is 20 m3, the potency index for chromium is
calculated as 4 x 10+3, which lies in the first quartile of the distribution.
7-97
-------
0-
-1
4th 3rd 2nd 1st
QUARTILE QUARTILE QUARTILE QUARTILE
+ 1 +2 -t-3
1 x 10 4 x 10 2 x 10
2345
LOG OF POTENCY INDEX
7 8
Figure 7-1. Histogram representing the frequency distribution of the
potency indices of 53 suspect carcinogens evaluated by the
Carcinogen Assessment Group.
7-98
-------
TABLE 7-21. RELATIVE CARCINOGENIC POTENCIES AMONG 53 CHEMICALS EVALUATED BY
THE CARCINOGEN ASSESSMENT GROUP AS SUSPECT HUMAN CARCINOGENS1.2.3
Slope
Compounds (mg/kg/day)"1
Acrylonitrile
Aflatoxin Bi
Aldrin
Allyl chloride
Arsenic
B[a]P
Benzene
Benzidene
Beryllium
Cadmium
Carbon tetrachloride
Chlordane
Chlorinated ethanes
1 , 2-dichloroethane
hexachloroethane
1,1,2, 2-tetrachloroethane
1 , 1 ,2-trichloroe thane
Chloroform
Chromium
DDT
Dichlorobenzidine
1 , 1-dichloroethylene
Dieldrin
0.24(W)
2924
11.4
1.19x10-2
15(H)
11.5
5.2xlO-2(W)
234(W)
1.40(W)
7.8(W)
1.30X10-1
1.61
6.9xlO-2
1.42x10-2
0.20
5.73x10-2
7x10-2
41(W)
8.42
1.69
1.47x10-1(1)
30.4
Molecular
weight
53.1
312.3
369.4
76.5
149.8
252.3
78
184.2
9
112.4
153.8
409.8
98.9
236.7
ib/.y
133.4
119.4
100
354.5
253.1
97
380.9
Potency
index
1x1 0+1
9x1 0+5
4x1 0+3
9x10-1
2x1 0+3
3xlO+3
4x10°
4x1 0+4
1x10+1
9x10+2
2x10+1
7x10+2
7x10°
3x10°
3x10+1
8x10°
8x100
4xlO+3
3xlO+3
4x10+2
1x10+!
lxlO+4
Order of
magnitude
dogl"
index)
+1
+6
+4
0
+3
+3
+1
+5
+1
+3
+1
+3
+1
0
+1
+1
+1
+4
+3
+3
+1
+4
(continued on the following page)
7-99
-------
TABLE 7-21. (continued)
Compounds
2,4-Dinitrotoluene
Diphenylhydrazine
Epichlorohydrin
Bis(2-chloroethyl)ether
Bis (chloromethyl)ether
Ethylene dibromide (EDB)
Ethylene oxide
Heptachlor
Hexachlorobenzene
Hexachlorobutadiene
Hexachlorocyclohexane
technical grade
alpha isomer
beta isomer
gamma isomer
Hexachlorodibenzodioxin
Methylene chloride
Nickel
Nitrosamines
Dimethylnitrosamine
Diethylnitrosamine
Dibutylnitrosamine
N-nitrosopyrrolidine
N-nitroso-N-ethylurea
N-nitroso-N-methylurea
N-nitroso-diphenylamine
Slope Molecular
(mg/kg/day)_i weight
0.31
0.77
9.9xlO-3
1.14
9300(1)
8.51
1.26(1)
3.37
1.67
7.75xlO-2
4.75
11.12
1.84
1.33
6.2xlO+3
6.3xlO~4
1.15(W)
25.9(not by q*)
43.5(not by q*)
5.43
2.13
32.9
302.6
4.92x10-3
182
180
92.5
143
115
187.9
44.1
373.3
284.4
261
290.9
290.9
290.9
290.9
391
84.9
58.7
74.1
102.1
158.2
100.2
117.1
103.1
198
Potency
index
6x1 0+1
lxlO+2
9X10"1
2x1 0+2
1x1 0+6
2x10+3
6xlO+1
1x10+3
5x1 0+2
2x1 0+1
1x10+3
3x10+3
5x1 0+2
4x10+2
2x1 0+6
5xlO"2
7xlO+1
2x10+3
4x10+3
9xlO+2
2xlO+2
4x10+3
3xlO+4
1x10°
Order of
magnitude
(logic
index)
+2
+2
0
+2
+6
+3
+2
+3
+3
+1
+3
+3
+3
+3
+6
-1
+2
+3
+4
+3
+2
+4
+4
0
(continued on the following page)
7-100
-------
TABLE 7-21. (continued)
Order of
Slope
Compounds (mg/kg/day)_i
PCBs
Phenols
2 , 4 , 6-t richlorophenol
Tetrachlorodibenzo-p-dioxin
Tetrachloroethylene
Toxaphene
Trichloroethylene
Vinyl chloride
4.34
1.9 9x1 0~2
1.5 6x1 0+5
3.5x10-2
1.13
1.9xlO-2
1.75x10-2(1)
Molecular
weight
324
197.4
322
165.8
414
131.4
62.5
magnitude
Potency (logjo
index index)
1x1 0+3
4x10°
5x1 0+7
6x10°
5x10+2
2.5x10°
1x10°
+3
+1
+8
+1
+3
0
0
Remarks:
1. Animal slopes are 95% upper-limit slopes based on the linearized multistage
model. They are calculated based on animal oral studies, except for those
indicated by I (animal inhalation), W (human occupational exposure), and H
(human drinking water exposure). Human slopes are point estimates based on
the linear non-threshold model.
2. The potency index is a rounded-off slope in (mMol/kg/day)"^ and is calculated
by multiplying the slopes in (mg/kg/day)"1 by the molecular weight of the
compound.
3. Not all of the carcinogenic potencies presented in this table represent the
same degree of certainty. All are subject to change as new evidence becomes
available.
7-101
-------
This provides an estimate of the relative potency of chromium in comparison
with other suspect carcinogens evaluated by the GAG.
7.2.4. Summary. Epidemiologic studies of chromate production facilities in
the United States, Great Britain, West Germany, and Japan have all found an
association between occupational exposure to chromium and lung cancer. Workers
in the chromate production industry are exposed to both trivalent chromium (Cr
III) and hexavalent chromium (Cr VI) compounds. Most of the epidemiologic
studies did not attempt to determine which chromium compounds were the etiologic
agents.
The strength of the association of exposure in the chromate production
industry with lung cancer is evident by the high lung cancer mortality ratios
found in various studies, the consistency of results by different investigators
in different countries, the dose-response found in several studies, and the
specificity of the tumor site (i.e., the lung). The respiratory cancer mortality
ratio for chromate production workers was louud to be as high as 29 by Brinton
et al. (1952) and as low as 2.0 by Korallus et al. (1982). Comparison of
mortality ratios across studies would not be valid because of differences in
exposure intensity and duration, length of observation period, and percentages
of the cohort lost to follow-up. Certainly, however, the magnitude of the
mortality ratios found in several studies (studies of three independent cohorts
of chromate production workers found lung cancer mortality ratios of at least
9.5 or greater) lends strong support to the association of exposure in the
chromate production industry with lung cancer.
Three studies of the chrome pigment industry, one in Norway, one in Great
Britain, and the third in the Netherlands and Germany, have also found an
association between occupational exposure to chromium and lung cancer. The
predominant chromium exposure in the chrome pigment industry is to hexavalent
7-102
-------
chromium (Cr VI). One study of the chromium plating industry in England
(Royle 1975) reported that workers exposed primarily to hexavalent chromium
(chromic acid mist and some dichromate dust) had significantly (P < 0.05)
higher mortality from cancer at all sites than the control group; respiratory
cancer mortality was not found to be elevated, however, and the follow-up
period was only for 3 years. Thus, the results are inconclusive. The results
of a Japanese study of chrome platers (Okubo and Tsuchiya 1979) were negative.
A proportionate mortality study by Silverstein et al. (1981) of a die-casting
and electroplating plant where chromium was used for electroplating found an
excess of lung cancer mortality, but because of the presence of other carcinogens
in the plant, no conclusions regarding the association of chromium and lung
cancer can be drawn from this particular study. The results of studies of the
ferrochromium industry are inconclusive as to lung cancer risk.
In most of the studies, smoking data were inadequate for any detailed
analyses of smoking as a confounding variable with respect to lung cancer.
However, the relative risks of lung cancer found by several of the studies of
chromate production workers were higher than the risks that would be expected
on the basis of differences in smoking habits between the study group and the
controls. Also, the dose-response in terms of length of time worked in the
industry seen in the Hayes et al. (1979) and the Taylor (1966) studies, and the
dose-response found by Mancuso (1975), in terms of mg/m^-years of total chromium,
are not likely to be explained by differences in smoking habits among the dose
groups.
Using the International Agency for Research on Cancer (IARC) classification
scheme for the assessment of human evidence of carcinogenicity, the CAG considers
that the epidemiologic evidence of the carcinogenicity of chromium is sufficient
to establish a causal relationship between chromium and cancer in humans.
7-103
-------
Chromium compounds have not induced significantly increased incidences of
tumors in laboratory animals following exposure by the inhalation and ingestion
routes. Neither trivalent (Cr III) nor hexavalent (Cr VI) chromium compounds
have induced significantly increased incidences of lung tumors by inhalation.
Similar results have been obtained following the ingestion of trivalent chromium
compounds; however, these studies have not been reported in detail. There is
some positive evidence that hexavalent chromium compounds are carcinogenic
following subcutaneous injection or intrabronchial, intrapleural, intramuscular,
or intratracheal implantation; however, implantation site tumors have only
consistently been demonstrated using intramuscular implantation. Of all the
chromium salts, calcium chromate is the only one that has been consistently
shown to be carcinogenic in rats by several routes. Other chromium compounds,
strontium chromate, zinc chromate, sodium dichromate, lead chromate, lead
chromate oxide, and sintered chromium trioxide, have produced local sarcomas or
lung tumors in rats at the site of application. Although the studies available
indicate that metallic chromium powder and trivalent chromium compounds are
not carcinogenic, these compounds have been studied less extensively than
hexavalent chromium compounds. The relevance of studies using intramuscular
implantation to human risk following inhalation or oral exposure to chromium
compounds is not clear; however, these animal studies suggest that some hexa-
valent chromium compounds are likely to be the etiologic agent in human
chromium-related cancer.
' Although a number of epidemiologic studies have found an association between
exposure to chromium compounds and lung cancer, the best data that can be used
for estimating cancer risks due to exposure to chromium compounds are found in
the study by Mancuso (1975). Mancuso (1975) reported age-specific lung cancer
mortality data for chromate production workers in terms of total chromium
7-104
-------
exposure, which included exposure to both trivalent (Cr III) and hexavalent
(Cr VI) chromium compounds. As noted previously, available data suggest that
only hexavalent chromium compounds are carcinogenic.
Thus, a cancer risk estimate based on total chromium expooure will under-
estimate the risk due to hexavalent chromium alone. In the Bourne and Yee
(1950) study, the ratio of trivalent (Cr III) to hexavalent (Cr VI) chromium
in the airborne dust in the plant's nine major departments ranged from 1 to 3,
except in the case of two departments, where the ratios were 6 for the lime
and ash operation and as high as 52 for the ore preparation operation. There-
fore, the ratio of trivalent and hexavalent chromium in the plant did not
exceed 52, seems unlikely to exceed 6, and may be smaller. Thus, the under-
estimation of the risk from hexavalent chromium when the Mancuso (1975) data
on exposures to total chromium are used is unlikely to be more than sevenfold,
if the ratio is assumed to be 6. There are two other factors, however, that
may result in an overestimation of the risk: (1) there is a possibility that
the use of 1949 hygiene data may result in some degree of underestimation of
worker's exposures, and (2) the risk presented in this report may be somewhat
overestimated because of the implicit assumption that the smoking habits of
chromate workers were similar to those of the general white male population.
It is difficult to determine how much the risk has been overestimated in this
regard. However, it seems reasonable to assume that the risk is not over-
estimated by more than 4 times on the assumption that 80% of the chromium
workets and 50% of the control population smoked cigarettes, and that the
exposure may be underestimated by a factor of 2.
Thus, these factors must be considered when interpreting the CAG's approx-
imation, calculated from the Mancuso (1975) data, of the lifetime cancer risk
due to a constant exposure to air containing 1 ug/m-* of hexavalent chromium
7-105
-------
(Cr VI). The unit risk, is calculated as 1.2 x 1CT2 on the basis of total
chromium in the chromate plant. As discussed previously, this figure could
either underestimate or overestimate the risk of hexavalent chromium. The use
of total chromium as a surrogate for hexavalent chromium could result in an
underestimate of the risk from hexavalent chromium by no more than 7 times; on
the other hand, underestimation of plant exposures and of smoking habits in
the workers could lead to an overestimate of the risk by roughly 4 times. On
balance, the estimate based on the Mancuso data is judged to be the best possible
estimate of the risk from hexavalent chromium.
Data from three studies on ferrochroraium plants in Norway, Sweden, and the
Soviet Union have also been used to calculate the carcinogenic potency of hexa-
valent chromium (Cr VI). Each of these studies has at least one characteristic
(e.g., underestimation of ambient chromium concentration, concurrent exposure
to asbestos, and ill-defined cohort) that makes it less adequate than the
Mancuso study for purposes of risk assessment. Most of these characteristics
tend to overestimate the risk. The deficiences of these foreign studies are
addressed in detail in the epidemiology section. As may be expected, the
carcinogenic potency (and hence, the unit risk) estimated from the Mancuso
data is consistently smaller (but within a factor of 5) than the lower-limit
potency estimates calculated from the three foreign studies.
7.2.5. Conclusions. The following conclusions can be made:
1. Using the IARC criteria, the epidemiologic studies of chromate production
workers would be classified as sufficient evidence of carcinogenicity.
2. Using the IARC criteria, the evidence of the carcinogenicity of hexavalent
chromium (Cr VI) in animal bioassay studies would be considered sufficient.
The results in animals appear to be determined to some extent by the
solubility of hexavalent chromium compounds. Trivalent chromium (Cr III)
7-106
-------
has not yet been found to be carcinogenic in animal studies.
3. Accepting the findings as reported elsewhere in this document, hexavalent
chromium (Cr VI) is mutagenic. The GAG considers this to be supportive
of the finding that hexavalent chromium is carcinogenic in animal bioassays.
The lifetime cancer risk due to air containing 1 ug/nH of hexavalent
chromium compounds is estimated to be 1.2 x 10~^. This would place hexavalent
chromium (Cr VI) in the first quartile of the 53 compounds evaluated by the
CAG for relative carcinogenic potency.
Using the IARC classification scheme, the level of evidence available for
the combined animal and human daLa would place hexavalent chromium (Cr VI)
compounds into Group 1, meaning that there is decisive evidence for the
carcinogenicity of those compounds in humans.
7-107
-------
7.3. GENOTOXICITY
7.3.1. In Vitro Mutagenicity. In an attempt to understand the fundamental
biological activity of metals and its relationship to carcinogenesis, numerous
in vitro experiments have been conducted. Many of these studies attempt to
exploit the strong relationships between molecular events involved in
mutagenesis and carcinogenesis. In particular, the interaction of xenobiotics
with nucleic acids is believed to be a critical event in mutagenesis and/or cell
transformation. Cultures of mammalian cells and bacteria, as well as cell-free
systems, have been used to explore the potential mutagenicity/carcinogenicity of
chromium salts. Although mutagenicity assays employing bacterial tester strains
have received widespread use in screening compounds for mutagenicity (e.g., Ames
assay), these assay systems have not been frequently employed for screening
metals. Nevertheless, positive results have been obtained with several metals in
both in vitro mutagenicity assays and assays using DNA repair deficient bacteria.
A summary of the results obtained from in vitro mutagenicity assays of chromium
compounds are presented in Table 7-22.
In an early study of the mutagenicity of chromium in bacteria, Venitt and
Levy (197*0 tested three soluble Cr(VI) salts (Na-CrO., K-CrOj., and CaCrOJ for
mutagenic activity using a spot test. Following application of each compound to
the plate at levels of 0.05, 0.10, or 0.20 pmol, a 3-fold increase was observed
in the number of reversion colonies in the tester organism E. coli WP2 (try").
Similar results were observed with both E. coli WP2 and WP2 uvrA (absence of
excision-repair) were exposed to Na CrO^ in suspension, followed by removal of
the test compound and plating on minimal plates. Repair deficient E. coli WP2
(exrA, error prone DNA repair) and WP2 uvrA were compared with wild type E. coli
WP2 for mutagenic response in the spot test to K CrO^. Since all tester
7-108
-------
TABLE 7-22
The In Vitro Mutagenicity Bioassay of Chromic Compounds
Test
Reverse mutation E.
Reverse mutation E.
E.
Reverse mutation E.
E.
E.
Reverse mutation E.
E.
^ E.
1
1 — i
o Reverse mutation E.
Reverse mutation E.
Reverse mutation E.
E.
Reverse mutation S.
Indicator
Organism
coli WP2 (try")
coli HP2 (try")
coli HP2 uvrA
coli WP2 (try")
coli WP2 uvrA
coli WP2 exrA
coli WP2
coli WP2 uvrA
coli CM571
coli Hs30R
coli WP2
coli WP2
coli B/r WP2
typhimurium
TA92
Reverse mutation S. typhiraurium
TA1978, TA92
Compound
Tested
K_CrOa
cicroz
Na^K^
K CrO,
K2Cr°l)
K Cr 0
K2Cr2°7
K2Cr2S"
K2Cr2°7
chromate*
di chromate*
chromate*
di chromate*
chromate*
dichromate*
Valence
State
+6
+6
+6
+6
+6
+6
+6
+6
+6
+6
+6
+6
+6
+6
+6
+6
-1-6
+6
+6
+6
+6
Metabolic
Activation
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
yes
yes
Dose
0.05, 0.10, and
0.20 pmol/plate
0.5 to 5 rag/ml
0.5 to 5 rag/mil
0.05 pmol/plate
0.05 pmol/plate
0.05 pmol/plate
1.0 x 10"^M
1.2 x 10"^M
1.2 x 10"^
13.0 x 10"^M
13.0 x 10'^M
0.5 pg/mfc
NR
NR
NR
NR
100 and 200
nraol/plate
100 and 200
nmol/plate
100 and 200
nmol/plate
100 and 200
nmol/plate
Application Response
spot test +
suspension assay +
suspension assay +
spot test •»•
spot test +
spot test +
suspension assay +
suspension assay +
suspension assay
suspension assay +*
suspension assay +*
fluctuation test +
spot test +
spot test
plate +
incorporation
plate +
incorporation
plate +
incorporation
plate +
incorporation
plate
incorporation
plate
incorporation
Reference
Venitt and
Levy, 1971
Venitt and
Levy, 1971
Venitt and
Levy, 1971
Nishioka, 1975
Nakamuro et al . ,
1978
Green et al . ,
1976
Kanematsu
et al., 1980
Lofroth and
Ames, 1978
Lofroth, 1978
-------
TABLE 7-22 (oont.)
Test
Reverse mutation
Reverse mutation
Reverse mutation
Forward mutation
Forward mutation
to 8-azaguanine
resistance
Indicator
Organism
S. typhimurium
TA1535, TA1537,
TA1538, TA100,
TA98
S. typhimurium
S. typhimurium
TA1535, TA1537
TA98, TA100
Schizosaccharomyces
pombe
Chinese Hamster
V79 cells
Compound
Tested
K Cr 0
c t (
Na Cr,07
t C. \
Na Cr_07
£• C. 1
Na Cr 0
t C f
CrO,
J
CaCrOj.
K,Cr,07
^ £ i
ZnCrOu*Zn(OH)_
Na Cr 0?
C, C. \
CrO
CaCrO,,
KCr 0
*• ^ I
ZnCr01(«Zn(OH)2
K2Cr2°7
^Cr6°7
PbCro!!
Valence
State
+6
+6
+6
+6
+6
+6
+6
+6
+6
+6
+6
+6
+6
+6
+6
+6
+6
Metabolic
Activation
no
no
yes
no
no
no
no
no
yes
yes
yes
yes
yes
no
no
no
no
Dose
NR
10 to 40 pg/plate
10 to 80 pg/plate
10 to 200 pg/plate
10 to 200 pg/plate
10 to 200 pg/plate
10 to 200 pg/plate
10 to 200 pg/plate
10 to 200 pg/plate
10 to 200 pg/plate
10 to 200 pg/plate
10 to 200 pg/plate
10 to 200 pg/plate
102pM
0.35 to 0.78 pg/m«.
1 to 4 pg/mH
5 to 10 pg/mi
Application Response
spot test
plate •»•
incorporation
plate
incorporation
plate +
incorporation
plate •»•
Incorporation
plate +
incorporation
plate +
incorporation
plate +
incorporation
plate
incorporation
plate
incorporation
plate
incorporation
plate
incorporation
plate
incorporation
suspension +'
in culture medium +
in culture medium +
in culture medium
Reference
Kanematsu
et al., 1980
DeFlora, 1978;
Petrilli and
DeFlora, 1978
Pe trelli and
DeFlora, 1977,
1978
Bonatti et al. ,
1976
Newbold et al.,
1979
-------
TABLE 7-22 (cont.)
Test
Gene conversion
Gene conversion
Reverse mutation
-a
i
_. Reverse mutation
— *
Reverse mutation
Reverse mutation
Indicator
Organism
Schizosacoharomyces
pombe
Saccharemyces
cerevisiae D7
E. coli (try") Cr
E. coli H.-30R
S. typhimurium
TA98, TA1537,
TA1535, TA100
S. typhimurium
TA100
Compound
Tested
K2Cr20?
Cr03
_(S01.).1K,SO,-2l»H,0
Cr(CH,COO),
3 3
CrK(SO^)2«12H20
CrCl3'6H20
CrK( SO,^ -12H20
Cr(N03)3.9H20
Valence
State
+6
+6
+3
+3
+3
+3
+3
+3
+3
Metabolic
Activation Dose
no 102 to 105pM
no 10~2 to 10"3
no NR
no 130 x 10"3M
no/yes 20 mg/plate
no /yes
no/yes 800 pg/plate
no
no
Application Response
suspension +
suspension +
spot test*
suspension assay +•
plate
incorporation
plate
incorporation
plate
Incorporation
-
Reference
Bonatti et al. ,
1976
Fukunaga et al. ,
1982
Venitt and
Levy, 197t
Nakamuro et al. ,
1978
Petrilli and
DeFlora, 1977
Petrilli and
DeFlora, 1978a,b
* = Refer to text for further information.
NR = Not reported
-------
organisms yielded approximately the same mutagenic response regardless of the
presence or absence of DNA repair pathways, the authors concluded that chromium
directly interacted with DNA, with subsequent mispairing of bases occurring
during cell division. Other possible explanations for these results were not
considered. The soluble Cr(III) compound, Cr2S0ltK2SO]|»2H20 [This was the
formula given in the report, although the test compound was probably
Cr;?(S01.),"K2S01.«21lH20.] , gave negative results (specifics in experimental
protocol were not given), along with soluble salts of tungsten, molybdenum, zinc,
cadmium, and mercury.
Nishioka (1975) obtained positive mutagenic results with K Cr20_(VI) in a
suspension assay using the tester strain E. coli WP2 and WP2 uvrA, but not in the
tester strain E. coli CM,.-.. Since strain CM..™, is a recombination-deficient
DM
strain, it was postulated that metal mutagenesis needed some component of the
recA allele. Positive results were also reported for K2Cr20? by Kanematsu et al.
(1980) in the spot test using strain E. coli WP2; however, E. coli B/r WP2 was
negative (no dose information was presented in this study). Nakamuro et al.
(1978) used E. coli HsSOR, a uvrA minus mutant requiring arginine, to study the
mutagenicity of KgCr^tVI) and KgCrO^VI), and Cr(CH3COO)3(III) in a suspension
assay. The respective mutagenic frequency of these compounds was 13**, 45, and 30
o
mutants x 10 viable cells. Although positive mutagenic responses were reported
for all test substances, the low survival of between 4 and 10$ at the doses which
were reported as mutagenic may lead to artifacts, particularly if spontaneous
revertants arise during growth on the trace levels of arginine used to supplement
the top agar, as suggested by Green (1976).
Green et al. (1976) also observed mutation in E. coli WP2 using a
fluctuation test. At levels of 0.5 pg/m£ KpCrOj., 148 of 250 tubes were positive
as compared to 61 in controls. At a high dose of 2.5 pg/mJl, only 10 of 150 tubes
7-112
-------
were positive as compared with 34 tubes in the controls. The apparent anti-
mutagenic effect at the higher dose was attributed to the toxicity of the K CrOj..
The mutagenicity of chromium compounds has been assessed in the Ames assay,
which uses specially constructed histidine dependent strains of Salmonella
typhimurium; however, in the original report on the validation of this assay,
McCann et al. (1975) stated that the test was not suitable for metals as a
result of the high concentration of magnesium, citrate, and phosphate salts in
the media. Chromate and dichromate but not the chromic ion (the specific salt
was not mentioned) were reported in an abstract to cause frameshift mutations in
S. typhimurium strains containing the his D3052 and his C3076 mutations, as well
as his G 46 strains containing the R factor. The greatest response was observed
in TA92 with five revertants/nmol Cr. Kanematsu et al. (1980) has reported
negative results with K-Cr 0_ in the spot test using strains TA1535, TA100, TA98,
TA1537, and TA1538; however, the lack of exposure information in this report
makes it difficult to evaluate. In a recent study of the mutagenic activity of
chromic chromate (18.5$ Cr(VI), and 50% total chromium) in S. typhimurium strain
TA1535, Witmer et al. (1982) reported that mutagenic activity, expressed as
number of revertants on test plates T number of revertants on spontaneous
plates, was increased as the standard salt concentration of the minimal media was
reduced. At least a portion of this observed increased activity resulted from
the poor survival of the cells in the low salt media and the associated decrease
in the number of spontaneous colonies. Although the author suggested that
variation in the S. typhimurium assay may be useful in further evaluation of the
mutagenic activity of inorganic compounds such as chromium, further validation
of this method will be required before these results can be compared with the
other available data on mutagenicity of chromium.
7-113
-------
In further studies, Lofroth (1978) found that chromate and dichromate were
mutagenic in strain TA1978 as well as TA92. On addition of a mammalian
microsomal activation system (S-9 mix) prepared from the livers of Aroclor
pretreated male rats, the number of revertants decreased from 206 and 357 (assays
without activation system) to 89 and 89 (assays with activation system) for
TA1978 and TA92, respectively. In the absence of NADP in the S-9 mix, the
inhibition of chromium mutagenicity did not occur. It was also reported that
addition of Cr(VI) to an aqueous solution of NADP did not change the valence
state of chromium. It was concluded that NADP plus additional factors in the S-9
mix were necessary to inactivate Cr(VI), possibly by reduction to Cr(III).
Similarly, Petrilli and DeFlora (1977) obtained positive responses with
Na2Cr2°7' Cr°3» CaCr°n» and K2Cr2°4 wnen assayed in strains TA1537, TA1535, TA98,
and TA100 at levels between 10 and 200 pg/plate. When the response was expressed
as revertants/pg of Cr(VI), there was no statistically significant difference in
the activity of each of these compounds. The Cr(III) compounds, CrK(SO.)_ and
CrCl,, gave negative results.
DeFlora (1978) and Petrilli and DeFlora (1978a,b) further studied the
effect of metabolic activation on the mutagenicity of Cr(VI) and (III). DeFlora
(1978) observed a decrease in the mutagenicity of Na Cr 0 in S. typhimurium
strain TA100 in the presence of S-9 mix prepared from rat liver. When NapCr 0
was added at 40 pg/plate, the number of revertant colonies decreased from 705 to
420, 370, 283, 228, and 221 with the incorporation, respectively, of 0, 10, 20,
30, 40, and 50 pI of S-9 fraction/plate. Petrilli and DeFlora (1978b) extend
these observations to the Cr(VI) compounds Na Ca 0_, CrO-,, KpCrCL,
ZnCrOjj'Zn(OH)2 (zinc yellow also contained 10$ CrO,), and PbCrOj*PbO, with muta-
genic activity in strains TA1535, TA1537, TA98, and TA100 detected in the absence
of S-9 mix but not in the presence of S-9 mix, prepared from rat liver. Little or
7-114
-------
no loss of mutagenic activity occurred when the added S-9 was prepared from lung
or muscle. Also, human plasma added to the top agar had no effect on chromium
mutagenicity, but lysates of erythrocytes did result in the loss of mutagenic
activity. The loss of mutagenicity was associated with reduction potential of
components in the S-9 mix, with reduced GSH, TPNH, and G6PD plus S-9 mix (to
generate TPNH) all inhibiting the mutagenicity of Cr(VI). In a study using
varying levels of TPNH, the mutagenic response was correlated with the amount of
Na2Cr207 (originally 52 pg Cr(VI)/plate) that remained as Cr(VI). Conversely,
Cr(III) as CrKtSOj.)-, CrCl,, and CrtNO,)^ which were inactive in the Ames assay,
were converted to active mutagens by the addition of nontoxic levels of the
strong oxidizer KMnO^ (Petrilli and DeFlora, 1978a). The presence of this
oxidizing agent reversed the effect of S-9 mix on the mutagenicity of Cr(VI).
These data supported the conclusions that the valence state of chromium was an
important factor in producing a mutagenic response in the Ames assay, and that
naturally occurring biological reducing agents were capable of reducing Cr(VI)
to Cr(III). Although the studies show that Cr(VI) compounds produce a positive
response, these studies do not indicate whether Cr(III) or (VI) is the ultimate
mutagen which interacts with the DNA of the cell.
Further work on mutagenicity has also been reported by Warren et al. (1981),
and De Flora (1981). Warren et al. tested 17 hexacoordinate Cr(III) compounds
for DNA damage with the E. coli differential repair assay, and, for mutagenicity,
with S. typhimurium. The compounds showed a parallel between mutagenicity and
differential inhibition in the two bioassays. The greater the differential
inhibition in the repair test, the more likely the compound was mutagenic. The
genetically most active complexes were those that had a charge of +1 with two
relatively stable ligands, along with four more stable aromatic amine ligands.
The Cr(III) complexes were similar in size, reactivity and cellular transport to
7-115
-------
many ambient cellular molecules. They concluded that the reduction of Cr(VI) in
the cell could result in a mutagenically active Cr(III) species.
De Flora (1981) tested 18 chromium compounds in the Salmonella microsome
test designed to provide dose response curves and a potency rating. Ten of the
eleven Cr(VI) compounds showed a mutagenic response, and the dose response range
for the seven most reactive compounds was 1.4 to 5.1 net revertance per nmole.
None of the pure Cr(III) compounds showed positive results. Chromite, which
contained traces of Cr(VI), was positive only for a 2 mg spot test.
Petrilli and De Flora (1982) investigated whether the administration of
dichromate in rats may modify the lung and S-9 fractions by decreasing Cr(VI)
mutagenicity. They noted that daily intratracheal administration of sodium
dichromate (0.25 mg/kg daily, 5 times/week for 4 weeks) increased Cr(VI)
deactivation. On the other hand, S-9 fractions of muscle tissue had no effects
on deactivation. From these preliminary results,they concluded that sub-chronic
exposure to Cr(VI) could enhance the cytoplasmic defense mechanisms and thus
raise the threshold of susceptibility.
Cr(VI) compounds have also been demonstrated to be mutagenic in vitro in
eucaryotic test systems, while Cr(III) was inactive. Bonatti et al. (1976)
2 5
exposed yeast, Schizosaccharomyces pombe, to 10 to 10 pM of K?Cr507 in a
forward mutation assay and a test for gene conversion. Forward mutation was
2
observed in 7 of 480,05*1 colonies exposed to 10 pM K?Cr_07 for 7 hours, which
was not statistically increased from controls in which no mutants were observed
in 84,546 colonies; however, if comparison was made to the historical spontaneous
mutation rate (no mutants in 10 colonies), then chromium significantly
increased the mutation rate. In the study of gene conversion, there was a dose
related increase in the four allelic combinations examined. The authors noted
that there were limitations in this study, but that the data suggest that KpCr?07
was mutagenic under the test conditions.
7-116
-------
Using another strain of yeast, Saccharomyces cerivisiae D7, Fukanaga et al.
(1982) examined the genetic activity of chromium trioxide. The compound was
-2 -3
incubated with the cells at concentrations of 10 to 10 J M for a period of
24 hours. Following incubation, the cells were plated to determine viability and
recombination in the ade locus. The highest concentration tested caused nearly
100$ cell death while viability at the lowest concentration was 77$. The cross-
_p
over frequency in treated cells was dose dependent with the 10 M concentration
resulting in a frequency of 1.4$ as compared to 0.02$ for untreated cells. It
was also reported that only growing and not resting cells were susceptible to the
effects of chromium, suggesting to the authors that chromium may affect the
fidelity of DNA polymerase.
Forward mutagenesis to 8-azaguanine resistance in cultured Chinese hamster
V79/4 cells was used by Newbold et al. (1979) to assess the genetic activity of
both Cr(VI) and (III). The highly soluble and slightly soluble Cr(VI) salts,
K_Cr 0_ and ZnCrO^, respectively, both produced dose related increases in
8-azaguanine resistant cell colonies, while the insoluble Cr(VI) compound,
PbCrOjj, and the soluble Cr(III) compound, Cr(CH-COO)., were inactive. The more
soluble K_Cr_07 was =5-fold more active than the ZnCKk. It was stressed that
both the valence state and solubility of a chromium compound affected its muta-
genic potential. Although high solubility was necessary for a mutagenic
response, the authors suggested that low solubility would facilitate a carcino-
genic response, since compounds of lower solubility might remain in the body
longer, providing a low dose continuous exposure to chromium.
7.3.2. . Effects On DNA And DNA Replication. It is apparent from the results of
bioassays in both prokaryotic and eucaryotic systems that some chromium
compounds are mutagenic. In general, soluble Cr(VT) compounds were
7-117
-------
positive in reverse and forward mutagenicity assays, while insoluble salts and
Cr(III) compounds are inactive. Metabolic activation inhibited the mutagenicity
of Cr(VI), probably by the reduction of the Cr(VI) to Cr(III) by cellular
reducing agents such as GSH. Other studies have been performed which indicate
that chromium compounds interact with DNA and decrease the fidelity of DNA
polymerase, and this mechanism may participate in chromium induced mutagenesis.
The difference in the sensitivity of recombination deficient strains of
bacteria as compared to wild type to the toxic effects of chromium compounds has
been used as an indicator of DNA damage. Nishioka (1975) exposed both rec~ and
rec+ strains of Bacillus subtilis to K^r^UD, KgCrO^VI), and CrCl^III) by
placing 0.05 mi of 0.05 M solution on a filter disc placed on the agar surface.
For the two Cr(VI) compounds, the zone of growth inhibition was greater for the
rec~ strains as compared with the rec , indicating DNA damage, while there was no
difference in the size of the zone of inhibition in the assay of CrCl_. The
positive rec-effect with K^Cr^O- was diminished sharply when the reducing agent,
Na_SO,, was mixed with the test compound, indicating that the Cr(VI) oxidation
state was necessary to cause DNA damage. Similarly, Kanematsu et al. (1980), in
a survey of 127 metal compounds in the B. subtilis rec assay, reported that three
Cr(VI) compounds (K CrOj., K-Cr-Oy, and CrO.) were positive at 0.05 mi of a 0.005
and 0.1 M solution, respectively, for the salts and oxide. Cr(III), Cr?(SOj.).,,
was negative, and it was noted that the sample was toxic to the bacteria at the
dose tested.
Nakamuro et al. (1978) examined three Cr(VI) compounds (K Cr 0_, K CrO^,
and CrO-) and three Cr(III) compounds (Cr(CH C00)_, Cr(NO-)., and CrCl,) in the
rec~ assay with B. subtilis. The compounds were tested using 0.02 mi aliquots of
-1 -2
solutions between 3.2 x 10 to 1.6 x 10 M for the Cr(VI) compound and 1.3 to
1.6 x 10~ M for the Cr(III) compound. All compounds tested were positive except
7-118
-------
for CrClo(III). Although DNA damage as indicated by the rec effect was observed
with two of the Cr(III) compounds, the order of activities was
K2Cr207 > KgCrOj, > Cr03 > Cr(CH300)3 > Cr(N03)3> with the Cr(VI) compounds being
more active than the Cr(III) compounds.
In an attempt to determine the mechanism of DNA damage induced by chromium,
Levis et al. (1978) examined the effects of K2Cr207 exposure on DNA synthesis in
cultured BHK cells. The cells were treated with 10~3, 10~\ 10~5, and 10~6 M
KpCr?07 for 1 to U hours, followed by determination of 3H-thyraidine uptake into
DNA and the intracellular pools. At the two intermediate doses, increased
specific activity was observed in both the nucleotide pool and DMA, which was
attributed to chromium effects on the membrane transport of the exogenous labeled
thymidine. When DNA synthesis was adjusted for the changes in the specific
activity of the pools, transient and reversible inhibitions of DNA synthesis were
observed at 10~ M, and complete irreversible inhibition occurred at 10 M., It
was reported that inhibition of DNA synthesis occurred prior to, and to a greater
extent than, inhibition of protein or RNA synthesis. Although these effects on
DNA synthesis may be partly related to the effects of chromium on the cellular
membrane, Raffetto (1977) reported evidence of direct interaction with DNA in
AlSBcR cultured cells, as indicated by unscheduled DNA synthesis following
treatment with K Cr_07 (VI). In cells treated for 1 hour with 16, M, or 1 pg
Cr(III)/m& (as CrCl,), no increase was produced. This evidence suggests that at
least Cr(VI) can alter DNA synthesis by more than a single mechanism.
Loeb and coworkers (Sirova and Loeb, 1976; Loeb et al., 1977) have demon-
strated that chromium compounds adversely affect the fidelity of DNA transcrip-
tion. The compounds, CrCl.. and CrO,, were incubated with avian myeloblastosis
virus DNA polymerase, a template with restricted base composition, and coraple-
00
mentary deoxynucleoside radiolabeled with P and noncomplementary deoxynuc-
7-119
-------
leoside labeled with -*H. Following incubation, the error frequency was deter-
mined to be greater when either metal compound was added to the incubation system
as compared to the controls. The error prone avian myeloblastosis virus DNA
polymerase lacks exodeoxynuclease active for removal of noncomplementary base,
thus providing optimum conditions for detecting the nonfidelity of DNA
synthesis. In later studies, Tkeshelashvili et al. (1980) demonstrated a
similar increase in misincorporation of complementary bases using synthetic
polynucleotides and CrO., and CrClo in an assay system containing E. coli DNA
polymerase. Both compounds appeared to have approximately the same activity.
Nearest neighbor analysis indicated that the noncomplementary bases were
incorporated as single base substitution. Noncomplementary nucleotides were
also incorporated with native DNA from bacteriophage $x17t (am3) and E. coli DNA
polymerase in vitro in the presence of CrO,. Infidelity in DNA synthesis was
~~"~ ~""^^~""~ 3
detected by loss of an amber mutation which was assessed by growth of infected
strains of E. coli nonpermissive and permissive for this mutation.
Tamino et al. (1981) examined the in vivo and in vitro binding of both
Cr(VI) (K2Cr20_) and (III) (CrCO to DNA from cultured BHK cell, and in vitro to
commercial calf thymus DNA and synthetic polynucleotides. The interactions of
Cr(III) with these nucleic acids in vitro was dependent on the G and C content,
with synthetic poly(G) and poly(C) binding 1.05 chromium ions per 100 nucleo-
tides. In vitro and in vivo reaction also changed the ultraviolet (UV) absorp-
tion spectra and thermal stability of the DNA. When Cr(VI) was used there was no
observed change in the UV spectra whether exposure occurred in vitro or in vivo.
The manner of treatment with Cr(VI) did affect the thermal stability of the DNA
with in vitro exposure resulting in the same types of changes observed with
Cr(III), while little change in thermal stability occurred with in vivo exposure
to Cr(VI). The data indicate that Cr(III) has some capacity to cross biological
7-120
-------
membranes and interact with cellular DNA, and that the types of interactions
between chromium and polynucleotides may be different for Cr(VI) and (III).
Although in vitro mutagenicity indicated that Cr(VI) compounds were
generally active and Cr(III) compounds nonactive, the assays of chromium induced
DNA damage in the rec assay, unscheduled DNA synthesis in mammalian cultured
cells, and increased in vitro infidelity of DNA synthesis indicated that at least
some activity is associated with both valence states (Nishioka, 1975; Nakamuro et
al., 1978; Raffeto, 1977; Loeb et al., 1977; and Tkeshelashvili et al., 1980).
7.3.3. Chromium Induced Chromosomal Aberrations and Cell Transformations. In
vitro, chromium has been shown to result in the appearance of chromosomal
aberrations and cell transformation (Table 7-23). Fradkin et al. (1975)
observed morphologic changes and loss of anchorage dependent growth in BHK21
cells treated with 0.25 and 0.5 pg of CaCrCv»2H 0. Chromate transformed cells
maintain the ability to grow independent of anchorage even after the cells were
re-isolated and freed of exposure to CaCrO^»2H 0. Tsuda and Kato (1977) observed
morphologic transformation in primary hamster embryo cells after exposure to
K2Cr_07(VI) at a level of 0.1, 0.2, and 0.5 pg/mA for 24 hours. The transforma-
tion frequency increased from 0.179$ for the control cultures to 0.760, 2.86, and
2.70* for cells exposed to 0.1, 0.2, and 0.5 pg/m£, respectively. A transforma-
tion frequency of 2.10$ was achieved with the positive control, N-methyl-N'-
nitro-N-nitrosoguanidine (MNNG) at a level of 0.5 pg/mS,. Conflicting results of
exposure to CrCl-(III) and K Cr 0 (VI) on morphologic transformation of BALB/c
mouse "fetus cells in culture was obtained in two experiments by Raffetto (1977).
Although both experiments showed an increase in the number of transformed foci,
this increase was not dose-related, and there was a 10-fold difference between
the two experiments performed under similar conditions. The authors suggest
7-121
-------
TABLE 7-23
Chromium Produced Clastogenic Effects and Cell Transformation
Test
Cell transformation
Cell transformation
Cell transformation
Cell transformation
— j
I Host mediated cell
fij transformation
ru
Clastogenic
Clastogenic
Clastogenic
Clastogenic
Clastogenic
Indicator
Cells
BHK
C. \
primary embryo
hamster cells
BALB/c cells
primary embryo
hamster cells
primary embryo
hamster cells
BALB/c cells
CHO cells
mouse FM3A cells
hamster embryo
cells
V79 cells
Compound
Tested
CaCr
-------
TABLE 7-23 (cont.)
OJ
Test
Clastogenic
,
Clastogenic
Clastogenic
Clastogenic
Clastogenic
Clastogenic
Clastogenic
Clastogenic
Clastogenic
Indicator
Cells
cultured human
leukocytes
cultured human
lymphocytes
CHO cells
Don Chinese
hamster cells
cultured human
lymphocytes
polychromatic
erythrocytes from
NMRI mice
gel cells from
Boleophthalmus
dussumierl
human lymphocytes
human lymphocytes
Compound
Tested
K Cr 0
KCrJL7
CNCH^COO)-
Cr(NO )
CrCI 3 3
K Cr 0
3
K CrO
NS Cr 6
K CrO^ '
Na-CrO,,
Cro,
CrCI
Cr(NO )
KCHSOJ
CKCOOCfC),
CrCl,-6H.O
Cr (SO^)g'
CrO2
K C?0y
K2Cr207
K CrO
CaCrO. '
CrO,
K CrO.
Ha Cr 0
CrO
CrO,
Valence
State
+6
+6
+3
+3
+3
+6
+6
+6
+6
+6
+6
+6
+3
+3
+3
+3
+3
+3
+6
+6
+6
+6
+6
+6
+6
+6
+6
+6
Dose
0.125 to 1.0x10"6M
0.5 to 8.0x10" M
1 to 32 x 10"bM
*
32.0 x 10"?M
32.0 x 10 M
-8 -"5
10 , to 10 ,M
10 to 10"^
0.1 and 0.3 pg/ml
0.1 and 0.3 pg/mfc
0.25 pg/mJ,
0.25 pg/mfc
0.1 and 0.25 pg/ml
5 and 50 pg/m«,
50 and 150 pg/mJl
150 pg/mfc
5 and 20 pg/mA
32 pg/mfc
6 pg/mH
0.32 pg/mJl
0.8 pg/mJl
0.8 pg/md
0.025 to 0.1 pg/mH
0.01 to 0.02 pg/mJ.
0.025 to 0.1 pg/m£
2 x H8.5 mg/kg, IP
2 x 21.25 mg/kg, IP
2 x 12.12 mg/kg, IP
1 or 5 mg/kg IM
21) or 30.5 ppm in
the aquarial water
Occupational exposure
Occupational exposure
End point
Chromosomal aberra-
tions
Chromosomal abbe ra-
tions
Sister chroma t id
exchange
Sister chroma t id
exchange
Sister chroma t id
exchange
Micronucleus test
Chromosomal aberra-
tions
Sister chromatid
exchange
Chromosomal aberra-
tions
Response Reference
+ Hakamuro
•«• et al.,
* 6 1978
at >16 x 10 ° M
-
+ Stella
et al. ,
1982
+ Lev is and
+ Ma Jone ,
+ 1979
+
-
—
-
+ Ohno et al.,
1982
+
+
•f
+ Gomez -Arroyo
+ et al., 1981
+
+ Wild, 1978
+
-
+ Krlshnaja
and Rege,
+
+ Stella et al.
1982
+ Sarto et al. ,
1982
IP = Intraperitoneal Injection
IM = Intramuscular injection
-------
caution in interpreting these findings. Casto et al. (1979), using a different
experimental system, demonstrated that the Cr(VI) compounds, CaCrOu, K?Cru, and
ZnCrOjj, at concentrations of 0.006, 0.01, and 0.01 mM enhanced the virally
induced transformation frequency of primary embryo hamster cells. The increase
in transformations was -2 for the Ca and K salts, and 4 for the Zn salt. Other
metals assayed that had no indication from the literature of genotoxic properties
did not enhance the rate of viral transformation.
Increase in cell transformation has also occurred in cells isolated from
hamster embryos exposed in utero to Cr(VI) (DiPaolo and Casto, 1979). Initially,
it was demonstrated that NaCrO^'^O at levels of 1.0, 2.5, and 5.0 pg/mfc
increased the transformation frequency of isolated hamster embryo cells in
culture to 0.7, 2.1, and 3-48$, respectively. Following this initial study,
pregnant Syrian golden hamsters (number not reported) were given intraperitoneal
injections of NaCrOj. at 2.5 mg/100 g body weight on day 11 of gestation. It was
reported that cells isolated from embryos excised 2 days later produced elevated
numbers of transformed colonies; however, the data were not presented for these
studies. It was concluded that transplacental exposure to chromate resulted in
alterations of fetal cells.
A number of studies have also demonstrated that exposure of cells in culture
to either Cr(VT) or (III) compounds produces chromosomal aberrations. Raffetto
(1977) exposed BALB/c mouse cells to CrCl-(III) and K2Cr20,,(VT) for either 48 or
96 hours. Increase in the number of chromosomal aberrations was noted for Cr(VI)
at a level of 0.1 pg/mS,, but not for Cr(III) at the same level after a 48 hour
exposure; however, with 96 hours of exposure, Cr(III) at 0.4 pg/mfc (0.04 pg/mS,
was also tested) and Cr(VI) at 0.1 and 0.015 pg/mJ, produced significant (P<0.05)
increases in the number of chromosomal aberrations. Newbold et al. (1979) also
detected dose dependent chromosomal damage with K-Cr^O™ at levels between 0.35
7-124
-------
and 0.8 pg/mfi, in V79 cells. Similar results were reported by Levis and Majone
(1979), in which both Cr(III) compounds [CrClg, CrCNO,)., KCHSO)^, and
Cr(COOCH-K] and Cr(VI) compounds (KgCr-O™, Ua^CrJdy, K^CrO^, Na.CrO!,, CrOo, and
CaCrOj.) produced increases in chromosomal aberrations in CHO cells in culture.
Again, Cr(VI) compounds were more active, with positive responses obtained at
doses of 0.1 to 0.5 pg/mJl, while marginal positive responses with Cr(III)
occurred at doses of 5 to 150 pg/mJL Although both valence states produced
chromosomal aberrations, only Cr(VI) induced increases in sister chromatid
exchange, and even the increase observed with Cr(VI) was small in comparison to
the response elicited by the positive control, Mitomycin C. Sister-chromatid
exchange was also observed by Ohno et al. (1982) in Don Chinese hamster cells
following treatment with the Cr(VI) compounds CrO_, K-CrO^, and K^Cr 0- at con-
centrations of 0.32, 0.8, and 0.8 pg/mJl, respectively. The Cr(III) compounds
CrCl-'SHpO and Cr (SO^) »i»H 0 were less active than the Cr(VI) compounds at
levels of 32 and 6 pg/mfc, respectively, with the increases due to the sulfate not
significantly elevated over control values. Again, as in the previous study, the
levels of chromium induced sister-chromatid exchange was 10 times less than that
of the positive control, mitomycin C. However, in the study by Umeda and
Nishimura (1979), only the Cr(VI) compounds, KpCr 0_ and CrO_, produced chromo-
somal aberrations in mouse FM3A cells, while K CrO^(VI) was negative even at high
concentrations. It is not clear why Cr(III) was negative in this study, although
these results may reflect a difference in sensitivity of these cells.
The effectiveness of chromium compounds in producing chromosomal aberra-
tions' was affected by the valence state of the compound. Tsuda and Kato (1977)
demonstrated that 0.5 ug/m& of K^Cr 0. produced chromosomal aberrations in 51$
of hamster embryo metaphases examined; however, addition of the reducing agent,
Na2SO_, at a level of 0.6^5 pg/mJl resulted in a decrease in abnormal metaphases
7-125
-------
to 6%. This provides further evidence that Cr(VI) was more active than Cr(III)
in producing chromosomal aberrations.
Nakamuro et al. (1978) used peripheral blood human leukocytes in short-term
culture to elevate the clastogenic effects of a number of Cr(VI) and Cr(III)
salts. Following isolation and 24 hours of culture, the cells were exposed to
Cr(VI) (K2Cr207, KgCrOjj) and Cr(III) (Cr(CH3COO)3, CrCNO^, CrCl3) chromium at
the respective concentrations of 0.125 to 4, 0.5 to 8, 4 to 32, 32, and
32 x 10" M. Significant increases (P<0.01) in total chromosomal aberrations
were observed at the higher doses with all compounds except Cr(NO_),(III) and
CrCl.,(III). It was apparent that human leukocytes responded similarly to other
cultured cells in regard to sensitivity to chromium induced chromosomal aberra-
tions.
In vivo clastogenic activity of KCrCL has been observed in mice by the
micronucleus test (Wild, 1978). NMRI mice (4 animals of each sex) were treated
by two intraperitoneal injections of 0.0, 12.12, 24, 25, or 48.5 mg/kg of K CrCL
in saline, followed in 6 hours by examination of bone marrow for micronucleus-
containing erythrocytes. Of the 1000 cells examined per mouse, there were 3.1,
4.8, 9.6, and 15.0$ micronucleated polychromatic erythrocytes in the control and
treated animals, respectively. The responses in the two highest dose groups
represent significant increases over control values. Chromosomal aberrations,
including breaks, fragments, rings, exchanges and unclassified makers, were
observed in the gills of the fish Boleophthalmus dussumieri after exposure to
sodium dichromate(VI) (Kushnaja and Rege, 1982). The number of total aberrations
increased from 1 to 6 and 16 after intramuscular injection of Na-Cr 0_ at doses
of 0.0, 1, and 5 mg Cr/kg, respectively. Increases in total aberrations of 0, 7,
and 7 were also observed when the fish were maintained in water for 96 hours
containing chromium at levels of 0.0, 24.0, or 30.5 ppm (the high dose level
7-126
-------
killed 50% of the fish). Other heavy metals and mitomycin C also produced
chromosomal aberrations in this test system.
Human lymphocytes exposed either in vitro to chromium or isolated from
workers occupationally exposed to chromium have been used to investigate
chromium induced chromosomal damage. Increase in both sister chromatid exchange
and chromosomal aberrations were observed when cultured human lymphocytes were
exposed to the Cr(VI) compounds K2Cr207> CaCrO^, and Cr03 (Gomez-Arroyo, 1981;
Stella etal., 1982; Littorin et al., 1983); however, the Cr(III) compound,
CrCl.,, was totally inactive (Stella et al., 1982) in this assay. In blood
samples taken from workers in the chromium plating industry (Stella et al., 1982;
Sarto et al., 1982) there were increases in chromosomal changes in the peripheral
lymphocytes. Stella et al. (1982) obtained blood samples from 12 male workers
and 10 control donors not exposed to chromium. There was no significant
difference in the incidence of sister chromatid exchange in the 5 older workers
(24 to 47 years of age); however, in the 7 youngest workers (17 to 23 years of
age) there was a significant increase. The control subjects had an age
associated increase in sister chromatid exchange, while the level in the exposed
population appeared constant. Sarto et al. (1982) studied the peripheral
lymphocytes of workers in two "hard" plating (one of the plants was the same as
that studied by Stella et al., 1982) and two "bright" plating industries. In the
"hard" plating industry, exposure was to chromium alone, while in the "bright"
plating industry the workers were also exposed to nickel. As in the previous
study, the younger workers had significant increases in chromosomal aberrations
while the number of aberrations was not significantly elevated from control
levels in the older workers. There was also a positive association between
urinary chromium levels and the number of chromosomal aberrations in the two
"hard" plating plants. The authors concluded that these data support the
genotoxicity of the soluble Cr(VI) ion.
7-127
-------
Venier et al. (1982) tested 10 Cr(III) compounds used in leather tanning,
and one Cr(III)-chromite compound used as a pigment, for cytotoxicity
(inhibition and growth and survival of cultured hamster cells), mutagenicity (S.
typhimuriam), and clastogenic activity (chromosomal aberrations and SCEs).
Generally, the trivalent compounds were only slightly cytotoxic. The hexavalent
reference compounds, potassium dichromate, had a cytotoxic activity at 0.5 ng
Cr(VI)/m£ (ID50 for growth) and 0.2 pg Cr(VI)/mA (LD50 for cell survival). For
the trivalent compounds, 100 to 500 times greater amounts were required to
inhibit growth or destroy the cells. Only two trivalent compounds (chromite and
chromium nitrate) were mutagenic; however, both compounds were contaminated with
Cr(VT). Again, the frequency of SCEs was increased only for the Cr(III)
contaminated with Cr(VI). Only the chromium nitrate reference compound and the
weakly contaminated Cr(III) compounds caused an increase in chromosomal
aberrations. The authors concluded that hexavalent chromium compounds were
primarily responsible for the genotoxic effects of chromium.
In contrast, Elias et al. (1983) found that a range of 9 to 136 fig
Cr(III)/mJl caused a significant increase in SCEs, but, as seen by Venier et al.
(1982) and Ohno (1982), this only occurred at trivalent chromium levels 300 to
1000 times higher than that required by Cr(VT) compounds.
7.3.4. Summary. In vitro and in vivo assays of genotoxicity have tried to
clarify the mechanism of chromium carcinogenicity and have supported the
potential for Cr(VI) to be the active carcinogenic species. Cr(VT) has demon-
strated consistently postive mutagenic activity in a number of bacterial
systems. This activity was demonstrated in the absence of a metabolic activation
system. In the presence of a metabolic activation system, the mutagenic activity
of Cr(VI) disappeared. Because chromium shows only marginal, if any, mutagenic
7-128
-------
activity after metabolic activation, it was suggested that the mammalian enzymes
or cofactors in the activation system reduced Cr(VI) to Cr(III). Both Cr(III)
and Cr(VI) have been demonstrated to interact with DNA in bacterial assays, and
Cr(VI) has inhibited DNA synthesis and increased unscheduled DNA synthesis in
mammalian cells in culture. In in vitro studies, both Cr(III) and Cr(VI) have
increased the infidelity of DNA replication. As observed with interaction with
DNA, both valences of chromium have been demonstrated to produce clastogenic
effects in mammalian cells with Cr(VI) being more active than Cr(III). The
effects observed included a variety of chromosomal aberrations, sister chromatid
exchange, and the appearance of micronuclei in polychromatic erythrocytes.
Increased chromosomal damage also has been observed in human lymphocytes
cultured from subjects occupationally exposed to chromium. For all the observed
genotoxic effects, it has been suggested that Cr(III) may be the predominant
intracellular species as a result of the reduction of absorbed Cr(VI) by cellular
components.
7.4. DEVELOPMENTAL TOXICITY AND OTHER REPRODUCTIVE EFFECTS
7.4.1. Developmental Toxicity. Chromium salts have been shown to be teratogenic
and embryotoxic in mice and hamsters following intravenous or intraperitoneal
injection.
In one study, golden hamsters were given single intravenous CrO.(vi)
injections of 5, 7.5, 10, or 15 mg/kg on the eighth day of gestation (Gale,
1978)! Treatment groups were as follows: 15 dams exposed (5 mg/kg), 18 dams
exposed (7.5 mg/kg), 21 dams exposed (10 mg/kg), and 4 dams exposed (15 mg/kg).
In all but the 5 mg/kg group, dams exhibited signs of chromium toxicity,
including weight loss and tubular necrosis of the kidneys. The highest dose was
lethal to 75$ of the dams.
7-129
-------
Chromium given on the eighth day of gestation resulted in increased fetal
wastage at doses of >7.5 mg/kg.(statistics not reported). The major external
anomaly noted was cleft palate. Incidence was significantly elevated above the
control level for all treated groups (method not stated). The primary internal
abnormality recorded was hydrocephaly, with an increased incidence noted in all
treatment groups. In addition, a wide range of skeletal defects was noted.
In a second study, golden hamsters were given intravenous CrO_ of 8 mg/kg on
day 7, 8, 9, 10 or 11 (Gale and Bunch, 1979). Six pregnant female hamsters were
treated in each time-group with three concurrent controls. Fetal death was
greatest in litters treated on the seventh day of gestation when there was an 8U$
incidence of resorptions. Resorption incidence was elevated to a small degree on
days 8 and 9 (20 and 13?, respectively), but was not affected on days 10 and 11.
As in the previous study, cleft palate was the primary external abnormality
recorded, occurring most frequently in litters from dams exposed on day 7 of
gestation and at elevated incidence rates with respect to controls on days 8 and
9, but not on days 10 and 11 (determined based on tables of binomial confidence
limits). The only internal abnormalities noted were kidney defects in groups
exposed on days 7 or 8 of gestation. Group sizes were too small to establish a
significant difference from controls. As in the previous study, treated females
exhibited signs of chromium toxicity which included body weight loss and tubular
necrosis of the kidneys.
In a continuation of this work, Gale (1982) tested Cr(III) in noninbred
(LVG) and inbred (CB, LHC, SDH, MHA, and PDU) strains of hamsters for reproduc-
tive and teratological effects. The dams (10 animals/group) were treated by
intravenous injection on the 8th day of gestation with a single dose of 8 mg/kg
of Cr(III). On termination at day 15 of gestation, it was determined that
strains CB, LHC, and PDU were resistant to any adverse effects of treatment. In
7-130
-------
the other strains, LVG, LSH, and MHA there was an increased incidence of cleft
palate (33, 18, and 15 animals, respectively, while control groups had no more
than 1 pup/group with cleft palate) and other external abnormalities<. The dams
of the strains with the highest incidences of fetal abnormalities also lost
weight during the period of gestation following treatment. Strain differences
were reported for the reproductive effects of other metal compounds in these
animals; however, the genetic mechanism which results in these differences in
susceptability was not understood.
Matsumoto et al. (1976) administered chromium chloride (+3), 19.52 mg/kg
(as Cr), intraperitoneally to ICR mice on day 7, 8, or 9 of gestation. Eight
control dams received a saline injection. Treatment group sizes were 5,6, and 7
on days 7, 8, and 9, respectively. The embryonic and fetal death rates were
elevated following treatment on days 8 and 9 (P<0.001, method not stated). The
incidence of external malformations was significantly greater than in controls
in litters of dams exposed on day 8 of gestation (P<0.001, method not stated).
Recorded anomalies included exencephaly and open eyelids. In addition, there was
a small increase in skeletal defects in litters from dams treated on days 8 and
9.
In a second experiment, pregnant dams were given intraperitoneal injections
of chromium chloride as dose levels of 9-76, 14.64, 19.52, or 24.4 mg/kg as Cr on
day 8 of gestation. Group sizes were: control (11 dams), 9.76 mg/kg (13 dams),
14.64 mg/kg (13 dams), 19.52 mg/kg (7 dams), 24.4 mg/kg (9 dams). Fetal weights
were significantly reduced in all treatment groups (P<0.05, method not stated).
The incidence of external malformations was significantly greater than control
values for the 14.64 mg/kg dose group and all higher dose groups (P<0.05, method
not stated). Anomalies with greatest incidence in rats were open eyelids,
exencephaly, and acephalia. In addition, skeletal defects were noted in these
same groups.
7-131
-------
lijima et al. (1979) dosed pregnant ICR mice with CrO,(VI) by subcutaneous
injection. Mice were injected with a single 10 or 20 mg/kg dose on day 7, 8, 9,
10, or 11 of gestation. The 20 mg/kg dose was lethal to one-third of the dams.
In this group, a significant increase in external anomalies was recorded in
litters from dams exposed on day 8 of gestation. Cleft palate was the primary
abnormality. In addition, increased fetal and embryonic death rates were noted
in groups exposed to 20 mg/kg on day 8 or day 11.
The studies discussed in the text above are summarized in Table 7-24.
lijima et al. (1983) found that radiolabeled chromium levels in fetal
tissues increased, while levels in maternal blood decreased after single
intraperitoneal injections of 9-8 mg Cr ( CrCl-)/kg body weight at gestation day
8. Their findings indicated that CrCl could affect embryos directly and cause
neural tube defects. They noted also that pyknotic cells on the neural plate
could indicate the development of exencephaly.
Danielsson et al. (1982) found considerable differences between Cr(III) and
Cr(VI) in their distribution in embryonic and fetal uptake. Forty-two pregnant
mice received single doses (5 pg/kg i.v.) of CrCl, or Na_Cr_07 on days 8 to 18 of
gestation. On day 13 of gestation, embryonic concentrations were 12$ Cr(VI) and
0.4$ Cr(III). Fetal concentration of both compounds increased with gestational
age, probably binding to fetal calcified bone, which develops on day 14.
According to the authors, it appeared that Cr(VI) occurred at sufficiently high
fetal concentrations to cause direct effects on embryonic sturctures, i.e.,
delay in skeletal development by inhibition of cartilage formation. Cr(III), on
the other hand, was not detected in embryonic structures during early gestation,
but it .could act on placental structures and thereby affect fetal development.
As discussed previously (Section 5.2.1.), there are limited data
demonstrating transfer across the placenta. It should be pointed out that most
7-132
-------
TABLE 7-21
Teratogenic and Fetotoxic Effects of Chromium
u>
UJ
Compound Route Species
CrO- i.v. hamster
CrO, i.v. hamster
CrCl, i.p. mouse
CrO. i.v. hamsters
j (strain
LVG)
hamsters
(strain
CB)
Dose
5, 7.5, 10,
or 15 mg/kg
on day 8 of
gestation
8 mg/kg on
day 7, 8, 9,
10, or 11 of
gestation
9.76, 14.61,
19.52, or 21.1
mg/kg on day
8 of gestation
8 mg/kg on day
8 of gestation
8 mg/kg on day
8 of gestation
Fetal Effects
increased fetal death in
7.5, 10, and 15 mg/kg
groups, increased incidence
of cleft palate in all
groups, hydrocephaly and
skeletal defects
increased fetal death
following administration
on day 7, increased
Incidence of cleft palate
following administration on
days 7, 8, or 9
depression of fetal weights
in all Cr treated groups,
increase in rate of external
abnormalities for groups
treated with 11.61, 19.52,
or 21.1 mg/kg
increased incidence of cleft
palate
no effect
Maternal Effects Reference
depressed weight gain and Gale, 1978
kidney tubular necrosis
at all doses above 5 mg/kg
weight loss, tubular necrosis Gale and Bunch, 1979
of kidneys
not reported Matsumoto et al., 1976
body weight loss Gale, 1982
no effect
-------
TABLE 7-21 (oont.)
-4
I
oo
.«=•
CrO.
ound Route Species
hamsters
(strain
LHC)
hamsters
(strain
LSH)
hamsters
(strain
PD1)
hamsters
(strain
MHA)
s . c . mouse
Dose
8 rag/kg on day
8 of gestation
8 mg/kg on day
8 of gestation
8 mg/kg on day
8 of gestation
8 mg/kg on day
8 of gestation
10 or 20 mg/kg
on day 7, 8, 9,
10, or 11 of
gestation
Fetal Effects Maternal Effects
no effect no effect
increased incidence of cleft body weight loss
palate
no effect no effect
increased incidence of cleft body weight loss
palate
increase in external tnalfor- lethal to 1/3 of dams
mations in 20 mg/kg group
when dosed on day 8, as well
as increase fetal death when
dosed on day 8 or 11
Reference
lijima et al., 1979
-------
TABLE 7-21 (cont)
U)
Ul
Compound
Na Cr 0
(cKvi)S
Cr(VI)
Route
CrCl i.p.
CrCl i.p.
CrCl i.v.
i.v.
Species
Dose
mouse
mouse
[in
vitro]
[in
vitro]
9.8 mg/kg on
day 8 of ges-
tation
mouse 19.5 mg/kg/day
10 mg/kg on
days 13 and
16 of ges-
tation
0 to 15 pg/mfc
0.1 to 0.28
pg/mi.
Fetal Effects
Cr increased gradually and
peaked at 21 hr, exceeding
maternal blood Cr level.
Pyknotic cells in neuro-
epithelium of neural ecto-
derm in 2 of 5 embryos
after 1 hr; in all 5, after
8 hr.
Fetal Cr(III) was 0.1* of
maternal serum Cr 1 hr post-
i.v.; high accumulation of
Cr in yolk sac placenta. In
late gestation, Cr accumu-
lated in calcified areas of
fetal skeleton.
No overt cytotoxicity at
pg/mX, in embryonic cell
cultures (chick cells).
15
Fetal Cr(VI) was 12* of
maternal serum Cr 1 hr post-
i.v. In late gestation, Cr
accumulated in calcified
areas of fetal skeleton.
Affected cartilage produc-
tion at =0.1 pg/rai in em-
bryonic cell cultures
(chick cells).
i.v. = intravenous; i.p. = intraperitoneal; s.c. = subcutaneous
NR = Not reported; NA = Not applicable
Maternal Effects
Maximum blood Cr at 1 hr
post-i.p. and gradually de-
creased
NR
NR
NA
NR
NA
Reference
lijima et al., 1983
lijima et al., 1983
Danielsson et al., 1982
Danielsson et al., 1982
Danielsson et al., 1982
Danielsson et al., 1982
-------
studies have sampled fetuses near the end of gestation, not early in
organogenesis when the fetus may be more susceptible to teratogens. Further
research is clearly indicated to further define the nature and extent of possible
chromium teratogenesis and embryo toxicity.
7.4.2. Other Reproductive Effects. Behari et al. (1978) examined the effects of
chromium on testicular tissue in rabbits. Animals were injected
intraperitoneally with chromium nitrate or potassium dichromate at doses of
2 mg/kg for 3 or 6 weeks. Both forms of chromium resulted in a decrease in
testicular succinic dehydrogenase at 3 and 6 weeks. ftdenosine triphosphatase
was inhibited at both time points with both compounds, but the Cr(III) compound
resulted in more severe depression. Acid phosphatase was significantly
depressed at 6 weeks in animals given the trivalent salt.
The administration of Cr(III) resulted in thickening of the tunia albuginea
and congestion of blood vessels following 3 weeks of treatment. In addition,
cells in the seminiferous epithelium showed degenerative changes. At the 6 week
time point, these degenerative changes were more pronounced, and there was a
complete absence of spermatocytes in the lumen. Cr(VI) produced mild edema of
the interstitial tissue of the testes at 3 weeks. Following 6 weeks of exposure,
edema was more marked and congestion of the blood vessels was noted. Although
the seminiferous epithelium appeared normal, the tubules were devoid of sperma-
tocytes. ^
Hopkins (1965) postulated that incorporation of chromium into sperm and
subsequent passage into the epididymis may be a problem. This hypothesis is
based .upon observation of chromium uptake by testis, subsequent decline of tes-
ticular radioactivity, and subsequent increase in radioactivity in the
epididymis (see Section 5.2.2.). Gross and Heller (19^6) have reported
7-136
-------
sterility in rats exposed to zinc chromate and potassium chromate in the diet
(Section 7.2.5.).
7.4.3. Summary. Chromium has adversely affected fetal development and male
reproduction in experimental animals. Hamsters administered chromium trioxide
intravenously on day 8 of gestation had an increased incidence of cleft palates
in the young when examined on day 15 of gestation. The malformations were strain
specific and associated with maternal toxicity. Studies on the mouse indicated
that while some skeletal effects were present, increased incidence of cleft
palate or fetal death were not observed. While several of the studies reported
fetal malformations only where maternal toxicity was also present, not all
studies reported data on maternal effects, so definitive conclusions concerning
the correlation if any between fetal and maternal effects cannot be made at this
time.
Other reproductive effects of chromium include testicular degeneration in
rabbits receiving 2 mg/kg/day for 6 weeks of either Cr(III) or Cr(VI) compounds
by intraperitoneal injection. The Cr(III) compound produced more severe effects
in this study than did the Cr(VI) compound (Behari et al., 1978). The relevance
of these observations to effects observed after environmental exposure is
questionable, as is that of the previously discussed teratogenicity studies,
since the routes of exposure were not natural. There is clearly a need for more
relevant studies on the route of exposure than are presently available.
7.5. OTHER TOXIC EFFECTS OF CHROMIUM
The literature concerning chronic and subacute exposure to chromium
consists primarily of reports of no observable effect levels (NOEL). Although
7-137
-------
the kidney and liver have been shown to be targets following acute exposures,
reports of pathology in these organs following long-term exposure via relevant
routes were not found. There are two reports which suggest effect levels
following oral exposure; these studies are summarized in the following text. The
numerous studies which suggest free standing NOAELs are of limited value as a
result of experimental design and are summarized in Table 7-25.
Gross and Heller (1946) reported that 0.125$ K^rOjj in the feed of rats was
tolerated without observable effects, 0.25$ resulted in "subnormal condition",
including rough coat and "subnormal" young born to treated animals. Doses of 0.5
and 1$ resulted in diarrhea, rough dirty coats, and sterility. ZnCrCK adminis
tered in the feed at levels of 0.125, 0.25, 0.5, and 1.0$ resulted in subnormal
appearance, rough and dirty coats, and sterility at all dose levels. Group
sizes, duration of treatment, and criteria for determining sterility were not
reported.
Ivankovic and Preussman (1975) administered Cr 0_ to rats in their feed.
The compound was prepared by incorporating it into bread dough at levels of 2 or
5$. In a 2-year study (600 feeding days), it was estimated that for the 2$
treatment level, males consumed 75 g/kg (25 g/animal) over the entire study
period and females consumed 72 g/kg (18 g/animal); at the 5$ treatment level,
males consumed 180 g/kg and females 160 g/kg for the duration of the study. Body
weights were monitored and animals were maintained on normal rations following
the treatment period. At death, animals were autopsied and "all the important
organs" were studied for micropathology. In addition, a 90-day study was conduc-
ted also using bread levels of 2 or 5$. In this study, urinary protein, sugar,
bilirubin, blood, and sediment were monitored. During the last 30 days of the
study, treated animals were mated and the number of viable young produced in each
litter was recorded. At the end of treatment, blood samples were analyzed for
7-138
-------
TABLE 7-25
Studies Suggesting NOAELS or NOELS
Species
mouse
mouse
rat
(young)
rat
(young)
— a
^ rat
OJ
vo
dog
Route
drinking
water
feed
drinking
water
feed
feed
drinking
water
Compound Dose Duration
K,CrO,. 100, 200, 300, N.S.
* 300, 100, or
500 ppm
ZnCrOy 1J N.S.
K CrO,. 300 and 500 ppm N.S.
KgCrO,, 0.125* N.S.
Cr20- 0, 1, 2, or 5> 2 years
K CrO, 0.15, 2.25, 1 years
2 4 1.5, 6.75,
or 11.2 ppm
No. at
Start
N.S.
N.S.
N.S.
N.S.
60/
group
2/group
End points
Monitored
general appearance,
reproduction
general appearance,
reproduction
general appearance,
reproduction
general appearance,
reproduction
gross and micro-
scopic pathology,
body weights
urtnalysis including
albumin, acetone,
bile pigments
Result Reference
NOEL for all Gross and Heller, 1916
doses
NOEL Gross and Heller, 1916
NOAEL; slight Gross and Heller, 1916
roughness of
coal at 500 ppm
NOEL Gross and Heller, 1916
NOEL Ivankovic and Preussman,
1975
NOEL Anwar et al., 1961
glucose, indican
erythrocytes, and
specific gravity;
gross and micro-
scopic pathology of
adrenals, bone
marrow, brain, heart,
intestine, kidney,
liver, lung,
mesenteric lymph
node, parathyroid,
pancreas, spinal cord,
spleen, stomach,
thyroid, and tonsils;
weights of liver
spleen and kidney
-------
TABLE 7-25 (cont.)
Species
rat
No. at
Route Compound Dose Duration Start
drinking K-CrOtt o, 0.45, 2.2, 12 mo highest
water * H i|.5, 7.7, 11 dose 9
and 25 ppm females,
12 males
control
10 of
each sex,
all other
groups , 8
males, 8
females
End points
Monitored Result Reference
clinical blood NOEL MacKenzie et al., 1958
chemistry, body
weights gross and
microscopic patho-
logy
I
*r
O
rat
cat
drinking CrCl.,
water
0, 25 ppm
feed
chromium 50 to 100
carbonate, mg/cat/day
chromium
phosphate
cat
inhalation
chromium
carbonate
dust
from 3.3 to
to 83 mg/m3
average 58.3
mg/m3
12 mo
1 to 3
mo
12 males, clinical chemistry
9 females body weights, gross
and microscopic
pathology
10
86 ses-
sions
which
varied from
10 to 60
min and
averaged
28 min
for one
cat and
57 min
for the
other
organ weights
macroscopic patho-
logy microscopic
pathogy of lung,
heart, liver, stomach,
spleen, pancreas,
kidney, brain,
skatal muscles
gross and micro-
scopic pathology
NOEL
NOEL
MacKenzie et al., 1958
Akatsuka and Pairhall,
1934
NOEL
Akatsuka and Fairhall,
1934
-------
blood sugar, serum protein, serum bilirubin, and haemoglobin. Erythrocytes and
leukocytes were counted. Liver, spleen, kidney, brain, and ovaries were weighed,
fixed, and sectioned at autopsy. In addition, lung, heart, pancreas, stomach,
small intestine, and urinary bladder were also fixed and sectioned.
In the 90-day study, the only treatment related effect was a depression of
spleen and liver weights. Spleen weights appeared to be depressed at both doses
in both sexes. Liver weights appeared to be depressed in both dose groups for
females and in high dose males. Statistical analyses were not reported. In the
2-year study, no treatment related effects were reported.
Berry et al. (1978) examined localization of chromium within the kidney.
Rats were dosed by intraperitoneal injection with 0.1 mg potassium dichro-
mate/100 g body weight. Doses were given 5 times per week for 8 months. Chromium
was localized within cells of the proximal renal tubules, specifically within
lysosomes. Chromium was retained throughout most of the study period, being
eliminated only when necrosis involved the entire cytoplasm of the tubule cells.
7.5.1. Respiratory Effects. Steffee and Baetjer (1965) exposed rabbits, guinea
pigs, rats, and mice to mixed chromate dust via inhalation or intratracheal
injection. Inhalation exposures were conducted 5 hours/day, 4 days/week
throughout the lifespan of the animals. The average air concentration was
estimated to be 3 to 4 mg of CrO /m . The average weekly exposure was estimated
to be 53, 44, and 49 rag/hour for rabbits, guinea pigs, and rats, respectively.
The following exposure related effects were documented in the lungs of exposed
animals. Fifteen percent of the rabbits and rats in the inhalation exposures,
and rabbits and guinea pigs in the intratracheal groups exhibited granulomata.
This lesion was found in only one control rat and in none of the rabbits or guinea
pigs. The incidence of alveolar and interstitial inflammation was greater than
7-141
-------
controls in guinea pigs exposed via both routes. Exposure related effects in
liver, kidney, and spleen were not found.
Nettesheim et al. (1971) exposed mice to calcium chromate dust 5 hours/day,
5 days/week for life. The exposure concentration was 13 mg/m of CaCrOj.. This
exposure concentration depressed weight gains. Exposed animals showed marked
alterations in the epithelium of the bronchial tree after 6 months of exposure.
These alterations were graded from epithelial necrosis and atrophy to marked
hyperplasia. In addition, bronchiolization of alveoli (bronchiolar cells lining
alveolar walls) was observed. Another frequently observed effect was alveolar
proteinosis. In addition to lung effects, morphological changes were noted in
the tracheal submandibular lymph nodes. After 2 years of exposure, spleen and
liver were atrophied. Ulcerations in the stomach and intestinal mucosa were
noted occasionally.
Early historical recognition of the ulcerative property of Cr(VI) compounds
in humans is evidenced by several studies on the subject (Becourt and
Chevallier, 1863; Delpech and Hillairet, 1869: Legge, 1902).
Bloomfield and Blum (1928) reported on 23 men employed in six chromium
plating plants in the United States. Their findings are presented in Table 7-26.
They concluded that continuous daily exposure to chromic acid at concentrations
>0.1 mg/m is likely to cause nasal tissue injury. As can be seen from
Table 7-26, no concentrations <0.12 mg/m were observed; hence, injury to nasal
tissue caused by lower concentrations could not be ruled out.
Four of 33 chromium platers were found to have septal perforation, although
the highest measured concentration of chromium trioxide in the workplace was
3
0.003 mg/m . Of the 33 workers, six had what the author considered to be normal
noses. He suggested that in view of the low chromium concentration, the lesions
7-112
-------
TABLE 7-26
Clinical Findings in Workers Employed in Chromium-Plating Plants
Time Employed in
Case
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Chromium-Plating Time Over
Occupation Room, mo Tank, h/d
Chromium plater
Chromium plater
Foreman plater
Foreman plater
Chromium plater
Chromium plater
Chromium plater
Chromium plater
Chromium plater
Chromium plater
Chromium plater
Chromium plater
Chromium plater0
Chromium plater
Nickel platerd
Backer
Backer
Backer
Wiper
6
20
7
8.5
3.5
0.75
0.25
7
3
36
5
0.75
12
0.67
1.5
8
0.75
0.75
1.5
4
4
2
3
t
7
7
7
7
1
6
6
4
2
0
0
0
0
0
Approximate
CrO, Exposure, Perforated
mg/m Septum
1.5 ++
2.8 -M-
2.5
2.5
5.6
0.12
0.12
0.12
0.12
0.2
0.12
0.12
2.8
2.8
?
7 +•
?
?
?
Ulcerated Inflamed
Septum Muoosa Nosebleed
++ Yes
+ Yes
++ ++ Yes
++ ++ Yes
+> +•+• Yes
++ Yes
++ Yes
+-•• Yes
++ No
++ No
+ Yes
+ No
No
No
+ + Yes
+ Yes
+ No
+ No
+ No
Chrome
Holes
Yes
Yes
No
No
Yes
Yes
No
No
Yes
No
Yes
No
No
No
No
No
No
No
No
-------
TABLE 7-26 (cont.)
Case
20
21
22
23
Occupation
Foreman6
Foreman6
Clerk6
Inspector6
Time Employed in
Chromium-Plating
Room , mo
0
0
0
0
Time Over
Tank, h/d
0
0
0
0
Approximate
CrO, Exposure, Perforated
mg/ffl Septum
0
0
0
0
Ulcerated
Septum
-
-
-
-
Inflamed
Mucosa Nosebleed
+• NO
+ No
No
+ No
Chrome
Holes
No
No
No
No
"Source: NAS, 1971a
c++, marked; 4-, slight; -, negative.
.Used vaseline in nose.
Cyanide burns
Worked in other departments of factory
mo = month; h = hour; d = day
-------
that resulted were due to exposure to periodic high concentrations of chromium
trioxide that occurred when ventilation of the tank failed (Lumio, 1953).
Anodizing operators exposed to concentrations of chromic acid mist ranging
from 0.09 to 1.2 mg/m3 (as CrO ) developed ulceration of the nasal passages and
atrophic rhinitis (Gresh, 1944; Zvaifler, 1944).
The United States Public Health Service conducted a study on workers in
seven chromate-producing plants in the early 1950's. The results are shown in
Table 7-27. Unfortunately, the results of the physical examinations on the
workers were not related to chromium exposures, and hence, the data are of
limited usefulness (Federal Security Agency, 1953).
Mancuso (1951) reported on physical examinations of a random sample of 97
workers from a chromate-chemical plant. It can be seen from the results which
are presented in Table 7-28, that 61 of 97 workers (63$) had septal perforation.
The data suggested to the author that Cr(III) may be partly responsible for the
perforations; however, later studies have not provided support for this theory.
The results of examinations of nine workers in a chrome-plating plant are
shown in Table 7-29. Analyses of air samples showed chromium concentrations of
0.18 to 1.4 mg/m . Some degree of nasal septal ulceration was seen in 7 of the 9
men, with 4 of 7 demonstrating frank perforations (Kleinfeld and Russo, 1965).
Unfortunately, the effects of chromium for a specific length of time at a fixed
concentration were not studied.
In a Russian study conducted by Kuperman (1964), 10 apparently normal
persons were exposed to Cr(VI) aerosol concentrations of non-reported composi-
tion 'ranging from 0.0015 to 0.04 mg/m . Air containing Cr(VI) at 0.01 to
0.024 mg/m sharply irritated the nose when inhaled for short periods of time.
The most sensitive person responded at a chromium concentration of 0.0025 to
7-145
-------
TABLE 7-27
Perforation of Nasal Septum in Chromate Workers*
All Workers
—5
i
jr
Time Worked
in Chromate
Industry
<6 months
6 months to
3 years
3 to 10 years
>10 years
TOTAL
Total
No.
41
117
370
369
897
Workers with
Perforation
No. £
2..
46 39.3
205 55.4
257 69.6
509 56.7
White Workers
Nonwhite Workers
Workers with
Total
No.
32
89
235
297
653
Perforation
No.
0
28
104
190
322
%
0
31.5
44.3
64.0
49.3
Total
No.
9
28
135
72
244
Workers with
Perforation
No.
1
18
101
67
187
*
11.1
64.3
74.8
93.1
76.6
•Source: NAS, 1974
-------
TABLE 7-28
Perforation of Nasal Septum in Chromate Workers*
Ratio of
insol Cr*-: to
+6
sol Cr
Workers in plant
Chromium Concentration,
mg/m (as Cr)
•Source: NAS, 1974
insol = insoluble; sol = soluble
No. Workers
Examined
Workers with
Septal Perforation
No. %
<1.0:1 <0.25
0.26 to 0.51
<0.52
1.1 to 4.9:1 <0.25
0.26 to 0.51
>0.52
>5.0:1 <0.25
0.26 to 0.51
>0.52
TOTAL
Office workers 0.06
4
7
8
9
32
15
7
2
13
97
4
2
3
4
7
20
11
2
1
11
61
0
50
43
50
78
63
73
29
50
85
63
0
7-147
-------
TABLE 7-29
Nasal Medical Findings in a Chromium-Plating Plant*
Case
1
2
3
H
5
6
7
8
9
Age, yr
30
19
19
18
47
45
23
20
48
Duration of
Exposure, mo
6
2
12
9
10
6
1
0.5
9
Findings
Perforated septum
Perforated septum
Perforated septum
Perforated septum
Ulcerated septum
Ulcerated septum
Ulcerated septum
Moderate injection of septum and turbinates
Moderate injection of septum
•Source: NAS, 1974
7-148
-------
0.004 mg/nr; however, it was not known if this was a reaction to chromium or to
the acidity of the aerosol.
Vigliani and Zurlo (1955) reported nasal septal perforation in workers
exposed to chromic acid and chromates in concentrations of 0.11 to 0.15 mg/m •
The lengths of exposure were not known. Otolaryngologic exeiminations of 77
persons exposed to chromic acid aerosol during chrome plating revealed 19? to
have septal perforation and 48$ to have nasal mucosal irritation. These people
averaged 6.6 years of exposure to an air chromium concentration of 0.4 mg/m . In
14 persons, papillomas of the oral cavity and larynx were found. The diagnosis
of papilloma was confirmed by histologic examination. There were no signs of
atypical growth or malignant degeneration (Hanslian et al., 1967).
Cohen et al. (1974) have identified a serious health hazard among workers in
a nickel-chrome plating area. Thirty-five of 37 (95/t) employees exposed to
atmospheric concentrations averaging 0.0071 mg/m as total chromium were shown
to have developed significant nasal pathology and skin lesions characteristic of
exposure to chromic acid. The authors attributed the high incidence of adverse
health effects to the lack of emphasis on the implementation of good industrial
hygiene and personal hygiene. The mechanism postulated for the occurrence of the
observed nasal damage resulted from either long-term exposure to levels of Cr(VI)
below prescribed "safe" levels, or direct contact of the affected tissue resul-
ting from inadequate personal hygiene practices.
The literature suggests that chromium compounds are responsible for a wide
variety of other respiratory effects. Studies done by German investigators
demonstrate mixed results from exposure to chromium compounds. Fischer (1911)
and L^hmann (1914) reported that there were no marked clinical symptoms in
persons exposed to chromate dust. Other German investigators (Alwens and Jonas,
1938; Fischer-Wasels, 1938; Koelsch, 1938; Lehmann, 1932; Mancuso, 1951) have
7-149
-------
reported that prolonged inhalation of chromate dust caused chronic irritation of
the respiratory tract and resulted in such manifestations as congestion and
hyperemia, chronic rhinitis, congestion of the larynx, polyps of the upper
respiratory tract, chronic inflammation of the lungs, emphysema, tracheitis,
chronic bronchitis, chronic pharyngitis, and bronchopneumonia. X-ray findings
included enlargement of the hilar region (often on only one side), enlargement of
the lymph nodes, increase in peribronchial and perivascular lung markings, and
adhesions of the diaphragm. Letterer et al. (1974) and Lukanin (1930) stated
that a characteristic pneumonoconiosis resulted from exposure to some chromates.
correlation between workers exposed to a given airborne concentration of
chromium (VI) and the development of harmful effects could not be made.
Cohen and Kramkowski (1973) and Cohen et al. (1974) examined 37 workers
employed by a chromium-plating plant. Within 1 year of being employed, 12
workers experienced nasal ulceration or perforation. The airborne chromium (VI)
concentrations ranged from <0.71 to 9.12 pg/m .
In a chromium plating plant where the maximum airborne chromium (VI) concen-
tration was 3 pg/m , no ulcerated nasal mucosa or perforated nasal septa were
found; however, half of the 32 employees had varying degrees of mucosal irrita-
tion (Markel and Lucas, 1973). This was not thought to be significant by the
investigators, because the survey was carried out at the peak of the 1972 to 1973
influenza epidemic. The length of employment for the workers was as follows: >8
years, 15 workers; 4 to 8 years, 7 workers; 1 to 4 years, 4 workers; <1 year, 6
workers.
Machle and Gregorius (1948) reported an incidence of nasal septal perfora-
tion of 43.5$ in 354 employees who worked in a chromate-producing plant that
manufactured sodium chromate and bichromate. At the time of the study, airborne
chromate concentrations ranged from 10 to 2800 pg/m . The plant has been in
7-150
-------
operation for at least 17 years, and some employees had probably worked in the
plant when reverberatory furnaces, a prominent source of high chromate exposure,
were used.
Chromium exposure in Australian shipyards, resulting from welding opera-
tions, has been reported by Bell (1976) to cause irritation to the nose and
throat at welding fume ranging from 0.006 to 0.05 mg/ . Lung biopsy specimens
from welders confirmed the presence of severe pneumoconiosis similar to that
reported by Stettler et al. (1977).
Various other disease states such as asthma have been attributed to
chromium, but, in most cases, the etiologic relation to chromium is obscure
because of the presence of other chemicals (WAS, 1974). Asthma was reported as a
complaint among workers employed for >10 years in a ferrochromium plant in Norway
(Broch, 1950). Bronchial asthma was also reported in Russian bauxite workers,
where Cr(VI) was involved in the bauxite caking process (Budanova, 1976). A
clinical test for the diagnosis of actual or imminent bronchial allergic reac-
tions to Cr(VI) has been developed by Budanova and Makarova (1979). A correla-
tion was established between the concentration of leukocyte agglomerates in
blood stimulated by Cr(VI), with 3% for a healthy control group versus =16$ for
workers with heavy bronchial asthma. It appears, then, that chromium sensitiza-
tion can occur after inhalation exposure as well as dermal exposure. In one
chromium-alloy plant, four cases of pulmonary disease with nodular fibrosis and
ventilatory impairment were reported by Princi et al. (1962), but no such cases
were found in another similar plant (Pierce and Scheel, 1965).
Bovet et al. (1977) have suggested that workers exposed to chromium in
electroplating operations have increased frequencies of obstructive respiratory
disease. The lowest dynamic values of pulmonary function (e.g., vital capacity,
forced vital capacity, forced expiratory volume, and forced expiratory flow)
7-151
-------
were reported for those workers displaying the higher urinary chromium levels.
The effect of tobacco smoke on pulmonary function was minor compared to the
effect attributed to chromium exposure. Forced expiratory flow and forced
expiratory volume were the pulmonary functions decreased to the greatest extent
as a result of occupational exposure to chromium. The total number of workers
examined was 44, and no data on specific compounds, levels or duration of expo-
sure were reported.
Capodaglio et al. (1975) demonstrated alterations in respiratory function
among bichromate and chromic acid production workers. Observed changes in chest
X-ray and respiratory function parameters were hypothesized to be the result of
exposure to chromate, and the extent of the reduced pulmonary function was
correlated to length of exposure. Data on the specific compounds and levels of
exposure were not reported in the only available review of this study (NIOSH,
1975).
Workers in a chromite mine and concentration plant developed pulmonary
markings (ground-glass types 1 and 2) that were attributed to chromite dust.
However, free silica dust was also present in the air. No clinical or roentgeno-
logic evidence of fibrosis was found in the chromate workers (Federal Security
Agency, 1953). Royle (1975) has investigated the occurrence of various
respiratory symptoms among British electroplaters exposed to chromic acid. A
total of 997 platers and 1117 controls completed a Medical Questionnaire on
Respiratory Symptoms. There was a significant (p<0.025) occurrence of attacks of
bronchitis in platers (28.2$) compared to controls (23.7$). Platers also had
significantly higher incidences of haemoptysis, perennial nasal catarrah, and
Grade .2 habitual winter cough and Grade 2 winter phlegm production. Putative
asthma was indicated in 13.1$ of the platers at a 2.5$ level of significance.
Nasal and skin ulcers were significantly higher in platers. Smoking histories
7-152
-------
were comparable in intensity and longevity for both populations. No clinical
testing or medical examinations were performed to verify the findings of the
Medical Questionnaire. Controls were more frequently exposed to a variety of
"dusts" in present or previous employment situations or both, and were also
exposed to asbestos much more so than was the study population. Air samples
taken between the years of 1969 and 1970 were <0.03 mg/m in all but two cases
(these were reported as <0.1 mg/m ). Dust samples generally ranged from 0.3 to
97.0 mg/g, but were reported as high as 298 mg/g. Members of both the study and
control populations were exposed to a variety of additional compounds that could
cause similar symptoms (asbestos, cadmium, nickel), a circumstance that detracts
from the overall quality of the study as does the use of medical questionnaires
without any medical examinations to substantiate these subjective findings.
Lucas and Kramkowski (1975) have examined the occurrence of abnormal medi-
cal findings in a "hard" chromium electroplating processing plant. A total of 11
workers were screened for various medical complaints associated with exposure to
chromic acid in electroplating operations. Average age of the workers was 39
years (range 22 to 5^), with an average occupational exposure of 7.5 years (range
3 to 16 years). Average airborne chromium (VI) levels were reported as
O.OOM mg/m , with a range of >0.001 to 0.020 mg/m , well within acceptable limits
of the current NIOSH standards. Ventilation systems appeared to be functioning
adequately, and work surfaces adjacent to the electroplating site were free from
contamination by chromium trioxide, CrO_. A summary of medical findings is found
in Table 7-30. Despite the degree of protection offered by ventilation and
protective clothing and equipment, workers displayed a significant number of
adverse medical effects at presumably "safe" levels of exposure.
In addition, a fine nodular pneumoconiosis has been reported in a few
chromite miners in South Africa (Sluis-Cremer and duToit, 1968). The available
7-153
-------
TABLE 7-30
a
Medical Complaints of Workers in "hard" Chromium Electroplating Plant
Frequency Symptom
1 Nasal irritation
4 Nasal soreness 'c
6 Runny nose (chronic) '°
4 Frequent nose bleeds
2 Ulceration of nasal septum
9 Scars indicating previous
Ulceration of the nasal septum
4 Perforated nasal septum
5 Chronic coughing episodes
2 Pulmonary distress indicative
of emphysema
7 Current skin sores
9 Scars indicating healed chrome
Ulceration on skin
5 Gastric distress
1 Chemical diabetes
1 Ocular pterygium on corneal
conjunctiva
1 Advanced renal carcinoma (1967
diagnosis; 1968, one kidney
removed; 1973, metastasized
to second kidney; 1974, cobalt
treatment
Q _ ._..__ L mT*m • n__M . _w - — _ — _. u_ T- IIM - _m_ - -1-. - . _- TI. ._ -- -| -.• _, ; _-
.Source: Lucas and Kramkowski, 1975
Previous to occurrence of nasal perforation.
.Two occurences were attributed to cold outdoor temperatures.
Three individuals were reported as moderate to heavy smokers.
7-154
-------
evidence suggests that the pneumoconiosis is due to deposition of chromite dust
in the lung tissue, and that the condition is benign and caused no fibrosis.
Other components of the ore, such as iron, may also have been responsible for the
observed fine nodular pneumoconiosis.
7.5.2. Renal Effects Of Chromium. Several authors have reported kidney damage
following the deliberate ingestion or therapeutic application of chromium
compounds (Brieger, 1920; Godlman and Karotkin, 1935; Major 1922; Partington,
1950; Rambousek, 1913). Twelve persons died after the application of anti-
scabietic ointment in which sulfur had been replaced with Cr(VI). Necrosis of
the skin developed at sites of application and was followed or accompanied by
nausea, vomiting, shock, and coma. Urinalysis revealed albumin and blood.
Post-mortem findings included tabular necrosis and hyperemia of the kidneys
(Brieger, 1920).
Only one of the above studies (Goldman and Karotkin, 1935) was available for
review. It is difficult to determine if chromium has a direct toxic effect on
the kidneys or alters the normal homeostasis of the body, thereby exerting its
effect. Evidence to support the first view comes from a study conduted by Mutti
et al. (1979). They examined welders and chromium platers, and found that
workers with a higher degree of exposure to chromium showed a pattern of nephro-
toxicity, as evidenced by increases in the indices for renal tubular damage. The
length of exposure and concentration of chromium were not specified. It is
interesting to note that the workers who had a higher degree of exposure to
chromium also demonstrated higher urinary chromium values.
Pederson and Mersch (1978) reported the incident of a woman who ingested 10
mH of a 50-% chromic acid solution with 150 mJl of Coca-Cola. Vomiting occurred
several times in the hospital, where gastric lavage was performed and activated
7-155
-------
charcoal, magnesia, and milk were administered. Hemodialysis was initiated 18
hours after ingestion. Blood samples were consistent with anemia, which was most
pronounced by the eighth day of hospitalization. Slight granulocytopenia was
observed on the third day. There was significant proteinuria (1.9 g) on the
first day.
Bilirubin levels rose during the first 3 days. One day after ingestion, the
patient was admitted to the renal unit; the serum chromium was 1.37 pg/m£ (2.94
mg Cr/A whole blood). Dialysis treatment was instituted, which accelerated the
removal of chromium from the serum. During the patient's stay, no signs of
gastrointestinal or cerebral disturbance were noted. Hepatic function was
normal upon discharge, following a hospital stay of 11 days.
7.5.3. Miscellaneous Toxic Effects. Mancuso (1951) reported that chromate
workers frequently showed excessive susceptibility to inflammatory and
ulcerative conditions of the gastrointestinal tract caused by ingestion of
chromium.
Hepatic injury, apparently due to exposure to chromic acid mist from plating
baths, has been reported (Pascale et al., 1952). One woman who had been employed
for 5 years at a chromium plating factory was hospitalized with jaundice and was
found to be excreting significant amounts of chromium. A liver biopsy specimen
showed microscopic changes resembling those found in toxic hepatitis. Eight
coworkers were screened for urinary chromium excretion in an effort to investi-
gate the possibility that the hepatic damage was of occupational origin. Four of
the workers were found to be excreting significant amounts of chromium. In three
workers who had been exposed to chromic acid mists for 1 to 4 years, liver biopsy
specimens and a series of 12 hepatic function tests showed mild to moderate
7-156
-------
abnormalities. The correlation of degree of hepatic injury with the concentra-
tion of chromic acid mist and information on controls were not available.
Frenkiel and Albert (1976) reported that traumatic injury of the tympanic
membrane and external auditory canal, and changes in middle ear mucosa were
suffered by a worker who had fallen into a vat of chromic acid. Extensive
chemical burns were sustained, along with a 20 decibel conductive hearing loss
stemming from an acid burn to the right tympanic membrane.
7.6. SUMMARY OF TOXIC EFFECTS OTHER THAN CANCER FOLLOWING EXPOSURE TO CHROMIUM
COMPOUNDS
Inhalation is both the most predominant route of exposure to chromium
compounds in industry, and the route most extensively investigated. Local
effects on the respiratory system are the primary toxic effects observed in
workers exposed to chromium in the atmosphere. Cr(VI), in the form of chromic
acid, has been associated for many years with the development of perforations of
the nasal septum. The implication of chromic acid as the causative agent results
from the common occurrence of this disorder in the chromium-plating industry,
where exposure is restricted to this Cr(VI) compound. Other Cr(VI) compounds may
also participate in the etiology of perforated nasal septums, since this disorder
has been reported in the chromate manufacturing industry, where the predominate
exposures are to Cr(III) and the Cr(VI) compounds, sodium chromate and sodium
dichromate; however, chromic acid mist may also be present in these plants. It
is interesting to note that nasal septum perforation has not been reported as an
occupa'tional hazard in the chrome leather tanning industry or the chrome pigment
industry, both of which exclusively use Cr(VI), although these industries are
associated with severe chromium dermatitis. The lack of perforated nasal septums
in these industries may result from differences in the physical or chemical form
7-157
-------
I
of the chromium, droplets in the tanning industry and particulates in the pigment
industry as compared to the chromic acid mist generated in the plating industry.
The measurements of chromic acid associated with perforated septums in the
chrome-plating industry is >0.1 mg/m^ (see Table 7-26); however, it is not known
if lower concentrations are also effective. Also, severe irritation of the
throat and lower respiratory tract have been associated with chromium compounds
at concentrations as low as 0.12 mg/m . Again, as with perforated nasal septum,
this respiratory tract irritation is primarily associated with Cr(VI). Hyper
sensitivity may result from dermal or inhalation exposure to either Cr(VI) or
Cr(III); however, there is little information available on the levels of exposure
necessary to induce an allergic response.
Little information is available on systemic effects of inhalation of
chromium compounds, although Pascale et al. (1952) and Mutti et al. (1979)
reported liver injury in a chromate worker and kidney injury in a welder exposed
to chromium, respectively. Acute exposure of animals using a variety of routes
of administration (Section 7.1.2) have indicated that both Cr(VI) and Cr(III)
compounds can produce kidney and liver damage, although the dose levels employed
were relatively high. From the evidence available from both human case reports
and animals studies, it can only be speculated whether the kidneys and liver may
be target organs following chronic exposure to chromium compounds.
Although inhalation studies of occupational exposure to chromium indicate
that exposure to some chromium compounds can result in perforation of the nasal
septum, irritation of the respiratory tract, pneumoconiosis, bronchitis, chronic
lung congestion, and possible liver and kidney damage (as supported by target
organ toxicity in acute animal studies), there are insufficient data available to
make a quantitative risk assessment for either chromium as a class or individual
chromium compounds from these inhalation studies. The only studies that provide
7-158
-------
any exposure data are the studies of the occurrence of perforated nasal septums.
However, these are of limited and questionable quality, since measurements were
not made contemporary with exposure and personal habits, such as picking of the
nose, may result in high local concentrations of chromium. Since perforated
nasal septum results from the local destruction of the mucous membrane of the
nose, this does not represent a systemic effect of chromium. Also, in the study
by Nettesheim et al. (1971), inhalation exposure of mice to chromium resulted in
marked effects to the respiratory tract. These effects, including hyperplasia
and necrosis, were likely to have resulted from the severe local irritation of
the cells lining the air ways. It would not be expected that exposure to
chromium by any other route would result in this disorder, and although these
data can be used to derive an acceptable inhalation exposure level, they are
inadequate for determining safe levels of chromium by all routes of exposure.
The limited information on other systemic effects of inhaling chromium, liver and
kidney damage, contains both insufficient exposure data and too few case reports
to form a firm association between exposure and effect.
There are only a few instances of human exposure to overtly toxic levels of
chromium compounds by ingestion, and these represent acute exposure to massive
doses which provide little information on the safe levels of chromium following
chronic exposure. A number of animal studies have been performed in which the
chromium compound was administered in the food, water, or by gavage. The acute
oral toxicity data indicate that Cr(VI) is approximately 2 or 3 orders of magni-
tude more toxic than Cr(III), with the latter toxic at the level of g/kg body
weight (Section 7.1.2). The difference in valence state may be less relevant
following chronic or subchronic ingestion of chromium, since it is suggested that
Cr(VI) is reduced to Cr(III) under the acid conditions of the stomach. The
determination as to whether Cr(III) or Cr(VI) is more toxic after chronic expo-
7-159
-------
sure, however, cannot be made, since none of the studies employed a sufficiently
high dose to produce a toxic effect.
The only ingestion study in which an effect was observed was that of
Ivankovic and Preussman (1975) in which rats were fed diets containing 2 or 5%
Cr^Q (Cr+ ), 5 days/week for 90 days. The only observed effect was a reduction
in the weight of the liver and spleen in the treated male rats as compared with
liver and spleen weights of control animals. Similar results were observed in
female rats maintained on the same diet. Neither organ showed macroscopic or
microscopic abnor malities, and the authors concluded that these changes were not
toxicologically important. In a larger 2-year study using the same experimental
procedure and 60 animals of each sex per group, Ivankovic and Preussman (1975)
did not mention any treatment-related changes in organ weight, although it was
mentioned that no signs of chronic toxicity were observed. Therefore, it is
unclear whether the slight change in organ weight observed in the small number of
animals in the 90-day study was the result of spurious observation due to the
small group size, or whether the 5% exposure level represents a true NOAEL (no-
observed-adverse-effect-level). No matter whether this 90-day study represents
a NOAEL or an NOEL (no-observed-effect-level), the small group size makes this
study very tenuous as the basis for quantitative risk assessment.
While chromium compounds have been shown to cause developmental toxicity in
experimental animals, the reproductive effects (e.g., fetal malformation) were
observed only where maternal toxicity was also present. Because of the unnatural
routes of exposure in these studies (e.g., intravenous and intraperitoneal
injection), the relevance of these developmental effects to environmental
exposures is very uncertain; more research is indicated to better understand the
implications of different levels and routes of exposure.
7-160
-------
In the absence of any data on effect levels following chronic exposure to
chromium, the U.S. EPA, in the Ambient Water Quality Criteria Document for
Chromium (U.S. EPA, 1980a), derived acceptable daily intake values (ADIs) of
0.175 and 125 mg/day/man for Cr(VI) and Cr(III), respectively. These ADIs were
derived by using the highest NOEL available for each valence state. For Cr(VI),
the study of MacKenzie et al. (1958) was used, in which rats were exposed to
several levels of chromium in the form of K CrOj. up to 25 ppm in the drinking
water for 1 year, while for Cr(III), the chronic study of Ivankovic and Preussman
(1975) was used, in which rats were fed diets containing up to 5% C^O, for 2
years. The ADIs for both Cr(VI) and Cr(III) were expressed as mg/d of the
compound administered. Although these ADIs were derived, it is not apparent from
the toxicologic information available whether the ADIs are more appropriate for
the specific chromium compounds tested, K Cr(V and Cr-O-, rather than the general
classes of Cr(VI) and Cr(III). This is particularly hard to determine, since no
toxic effects were observed in these chronic animals studies.
7-161
-------
-------
8. CURRENT REGULATIONS AND STANDARDS
A number of recommended standards presently exist for permissible levels of
chromium in both air and water. The National Academy of Science (NAS, 1980) and
the U.S. Environmental Protection Agency (U.S. EPA, 1980a), in regard to ambient
water quality criteria, have stated that a distinction between the Cr(III) and
(VI) forms should be made when considering the formulation of regulations and
exposure criteria.
8.1. OCCUPATIONAL EXPOSURE
NIOSH (1973, 1975), OSHA (1978), and ACGIH (1981) have recommended various
exposure limits for chromium. These values are based on the chemical form of the
chromium compounds or their solubilities. Table 8-1 outlines the various United
States occupational standards for chromium compounds.
8.2. EXPOSURE TO CHROMIUM IN AMBIENT WATER.
A number of standards exist for chromium in ambient water or drinking water.
In general, the standards for occupational exposure to chromium in the air allow
for a greater uptake of chromium than by the uptake of chromium from drinking
water. The U.S. EPA (1980a) has estimated the daily intake of chromium from
drinking water to be 5 pg/day. Table 8-2 outlines various recommended or esta-
blished standards for chromium in the United States.
The U.S. EPA (1980a) recently proposed several ambient water quality cri-
teria for chromium. Based on methodology outlined in the Federal Register
(45 FR 79353), acceptable daily intakes (ADIs), no-observable-adverse-effect
levels (NOAELs), and bioconcentration factors (BCFs) obtained from experimental
animal studies, separate water quality criteria were proposed. Table 8-3
8-1
-------
TABLE 8-1
Recommended Occupational Standards and Recommended Criteria for
Chromium Compounds in the United States
Chemical Form
Non-carcinogenic chromium (VI)a
Carcinogenic chromium (VI)C
Chromic acid (as chromium trioxide)
Soluble chromic or chromous salt
Insoluble salts or chromium metal
Chromium metal
Chromium (II) compounds
Chromium (III) compounds
Chromium (VI) compounds
water soluble
water insoluble
r>
Chromite ore
Chromium: soluble chromic and
chromous salts
Standard (pg/m^)
25 TWAb
50 Maximum
1
50 TWA
100 Maximum
500 TWA
1000
500 TWA
500 TWA
500 TWA
50 TWA
50 TWA
50
500 TWA
Reference
NIOSH, 1975
NIOSH, 1975
NIOSH, 1973
OSHA, 1978
OSHA, 1978
ACGIH, 198ld
ACGIH, 1981
ACGIH, 1981
ACGIH, 1981
ACGIH, 1981
ACGIH, 1981
ACGIH, 1981
NIOSH listed "non-carcinogenic" chromium VI compounds as the monochromates and
dichromates (bichromates) of: hydrogen, lithium, potassium, rubidium, cesium,
ammonium, and chromic oxide (chromic acid anhydride).
DTime Weighted Average
'NIOSH listed "carcinogenic" chromium VI compounds as all other chromium compounds not
included in the "non-carcinogenic" chromium VI listed above.
These'are also known as the TLV values established by ACGIH.
a
'ACGIH listed these classes of chromium compounds as substances associated with
industrial use that have been recognized as carcinogens.
8-2
-------
TABLE 8-2
Recommended Standards for Chromium
In Ambient Waters in the United States
Chemical Form
Chromium (VI)
Total Chromium
Total Chromium
Chromium (VI)
Medium
drinking water
total
domestic water
supply
freshwater
(aquatic life)
livestock water
Criteria (ng/£)
50
50
100
1000
Reference
U.S. Public Health
Service (USPHS), 1962
U.S. EPA, 1976
U.S. EPA, 1976
National Academy
of
Chromium
community water
systems and non-
community water
systems
50
Science and National
Academy of Engineering
(NAS/NAE), 1972
40 CFR
8-3
-------
TABLE 8-3
Ambient Water Quality Criteria for
the Protection of Hunan Health3
00
1
Chemical
Form
Chromium (III)b
Chromium (VI)
NOAEL
(rag/ JO
50,000
25
Rat
NOAEL
(mg/d/kg)
1786
2.50
ADI
for man
(mg/d/man)
125
0.175
BCF
16
16
Calculated
Criteria (ng/fc)
59,000
83
1
Source: U.S. EPA, 1980a
Revised ADI and criterion. Published values (45 FR 79331) were incorrect. If exposure to trivalent chromium
results only from the eating of fish and shellfish, then the calculated ambient water criterion for chromium
(III) is proposed as 1200 mg/i, (1.2 x 10 jig/i)
-------
summarizes the proposed U.S. EPA (1980a) ambient water quality criteria for the
protection of human health.
The U.S. EPA (1980a) also has proposed several ambient water quality cri-
teria for the protection of aquatic life. Table 8-4 summarizes the proposed
criteria for the protection of aquatic life.
8.3. EXPOSURE TO CHROMIUM IN AMBIENT AIR.
No federal or state (other than the state of Maine, i.e., 0.05 pg/nr for
mean annual and 0.3 pg/m for mean daily limits) ambient air chromium standards
have been proposed. No United States emission standards for chromium were found
in the available literature. The U.S.S.R. recommends a "sanitary clearance zone"
of 1000 m for plants discharging 200 kg Cr(VI) per day, and 2000 m for plants
discharging 1000 kg per day (NAS, 1974).
8-5
-------
TABLE 8-4
Calculated Ambient Water Quality Criteria
for the Protection of Aquatic Life*
Chemical
Form
Chromium (VI)
Chromium (III)
Freshwater
24 -hour Average
(ng/a)
0.29
44
(chronic value
toxicity)
Life
Maximum
(us/ JO
21
NR
Marine Life
24-hour Average
(jig/A)
18
10,300 (acute
toxicity value)
Maximum
(ng/n)
1260
NR
•Source: U.S. EPA, 1980a
NR = Not recorded
8-6
-------
9. REFERENCES
Aaseth, J., J. Alexander and T. Norseth. 1981. Uptake of Cr-Chromate by
Human Erythrocytes—A Role of Glutathione. Acta Pharmacol. Toxicol. 50: 310-
315.
Aaseth, J., J. Alexander and T. Norseth. 1982. Uptake of 51Cr-chromate by human
erythrocytes - a role of glutathione. Acta. Pharmacol. Toxicol. 50: 310-315.
Abu-Samra, A., J.S. Morris and S.R. Koirtyohann. 1975. Wet ashing of biological
samples in a microwave oven. Anal. Chem. 47: 1475-1477.
ACGIH (American Conference of Governmental Industrial Hygienists). 1981. TLVs:
Threshold Limit Values for Chemical Substances and Physical Agents in the
Workroom Environment with Intended Changes for 1981. Cincinnati, Ohio.
Agemian, H., and A.S.Y. Chau. 1976. No title provided. Analyst. 101: 91, 761.
(Cited in Tinsley, 1983.)
Ahearn, A.J. (Ed.) 1972. Trace Analysis by Mass Spectrometry. Academic Press,
NY. pp. 161, 206 and 290. (Cited in Towill et al., 1978).
Akatsuka, K. and Fairhall. 1934. The toxicology of chromium. J. Ind. Hyg.
16: 1-24.
Alderson, M.R., N.S. Rattan and L. Bidstrup. 1981. Health of workmen in the
chromate-producing industry in Britain. Br. J. Ind. Med. 38: 117-124.
9-1
-------
Alexander, J., J. Aaseth and T. Norseth. 1982. Uptake of chromium by rat liver
mitochondria. Toxicology. 24: 115-122.
Alkezweeny, A.J., D.W. Glover, R.N. Lee, J.W. Sloot and M.A. Wolf. 1975.
Measured chromium distributions resulting from cooling-tower drift. In;
Cooling Tower Environment. S.R. Hanna and J. Pell, eds. U.S. Energy Research
and Development Administration, pp. 558-572. (Cited in Towill et al., 1978).
Altshuler, B. 1961. The Role of the Mixing of Intrapulmonary Gas Flow in the
Deposition of Aerosols. In; Particles and Vapours. II. C.N. Davies, Ed.
Pergamon Press, Oxford, p. 47. (Cited in U.S. EPA, 1982).
Alwens, W. and W. Jonas. 1938. Der Chromat-Lungenkrebs. Acta Un. Int. Cancr.
3: 103-118.
Andersen, R.A. 1981. Nutritional role of chromium. The Science of the Total
Environment. 17: 13-29.
Andersen, R.A. 1981. Chromium as naturally occurring chemical in humans.
Proceedings of Chromate Symposium Industrial Health Foundation, Inc.
Pittsburgh, PA. pp. 332-341.
Andersen, R.A., M.M. Polanksy, N.A. Bryden et al. 1982. Urinary chromium
excretion of human subjects: effects of chromium supplementation and glucose
loading. Amer. J. Clinical Nutr. 36: 1184-1193-
9-2
-------
Andersen, R.A., M.M. Polansky, N.A. Bryden, K.Y. Patterson, C. Veillon and
W.H. Glinsmann. 1982. Effects of chromium supplementation in urinary Cr excre-
tion of human subjects and correlation of Cr excretion with selected clinical
parameters. J. Nutrition. (In press.)
Anonymous. 1976. Metal Sensitivity. Lancet, March 26. p. 677-678.
Anwar, R.A., C.F. Langham, C.A. Hoppert, B.V. Alfredson and R.I'. Byerrum. 1961.
Chronic toxicity studies. III. Chronic toxicity of cadmium and chromium in
dogs. Arch. Environ. 3: 456.
Armitage, P. and R. Doll. 1954. The age distribution of cancer and a multistage
theory of carcinogenesis. Br. J. Cancer. 8: 1-12.
Axelsson, G., R. Rylander and A. Schmidt. 1980. Mortality and incidence of
tumours among ferrochromium workers. Br. J. ind. Med. 37: 121-127.
Baetjer, A.M. 1950a. Pulmonary carcinoma in chroroate workers. I. A review of
the literature and report of cases. Arch. Ind. Hyg. Occup. Med. 2: 487-50U.
(Cited in NIOSH 1975.)
Baetjer, A.M. 1950b. Pulmonary carcinoma in chromate workers. II. Incidence
on basis of hospital records. Arch. Ind. Hyg. Occup. Med. 2: 505-516. (Cited
in NIOSH 1975.)
Baetjer, A.M., C. Damron and V. Budacz. 1959a. The distribution and retention
of chromium in men and animals. Arch. Ind. Health. 20: 136-150.
9-3
-------
Baetjer, A.M., J.F. Lowney, H. Steffee and V. Budacz. 1959b. Effect of chromium
on incidence of lung tumors in mice and rats. Arch. Ind. Health. 20: 124-135.
Baines, A.D.W. 1965. Cell renewal following dichromate induced renal tubular
necrosis. Amer. J. Pathol. 47: 851.
Bartsch, P., A. Collignon, G. Weber, G. Robaye, J.M. Delbrouck, I Roelandts and
J. Yujie. 1982. Distribution of Metals in Human Lung: Analysis by Particle
Induced X-Ray Emission. Arch. Environ. Health. 37: 111-117.
Becourt, M.M. and A. Chevallier. 1863. Memoire sur les accidnets qui atteignent
les ouvriers qui travaillent le bichromate de potasse. Ann. d'Hyg. Publ. Med.
Leg. 20: 83-95.
Behari, J., S.V. Chandra and S.K. Tandon. 1978. Comparative toxicity of triva-
lent and hexavalent chromium to rabbits. III. Biochemical and histological
changes in testicular tissue. Actu. Biol. Med. Germ. 37: 463-468.
Bell, A. 1976. Industrial hygiene and occupational health studies in Australian
(New South Wales) shipyards. Environ. Res. 11: 198-212.
Berndt, W.O. 1976. Renal chromium accumulation and its relationship to chromium
induced nephrotoxicity. J. Toxicol. Environ. Health. 1: 449-459.
Berry,. J.P., J. Hourdry, P. Galle and G. Laquie. 1978. Chromium concentration
by proximal renal tubule cells: An ultrastructural microanalytical and cyto-
chemical study. J. Histochem. Cyctochem. 26: 651-657.
9-4
-------
Bhagat, S.K., W.H. Funk, R.H. Filby and K.R. Shah. 1971. Trace element analysis
of environmental samples by neutron activation method. J. Water Pollut. Control
Fed. 43: 2414-2423.
Bhargava, O.P., H.E. Bumsted, F.I. Grunder, B.L. Hunt, G.E. Manning, R.A.
Riemann, J.K. Samuels, V. Tatone, S. Jeff Waldschmidt and P. Hernandez. 1983.
Study of an analytical method for hexavalent chromium. Am. Tnd. Hyg. Assoc. J.
44(6): 433-436.
Bidstrup, P.L. 1951. Carcinoma of the lung in chromate workers. Br. J. Med.
8: 302-305.
Bidstrup, P.L. and R.A.M. Case. 1956. Carcinoma of the lung in workman in the
bichromates-producing industry in Great Britain. Br. J. Tnd. Med. 13: 260-264.
(Cited in NIOSH 1975.)
Bittersohl, G. 1971. Epidemiological research of cancer cases in the chemical
industry. Arch. Geschwulstforsch. 38(3-4): 198-209.
Black, M.S. and R.F. Browner. 1981. Volatile metal-chelate sample introduction
for inductively coupled plasma-atomic emission spectrometry. Anal. Chem.
53: 249-253.
Block, C. and R. Dams. 1976. Study of fly ash emission during combustion of
coal.' Environ. Sci. Technol. 10: 1011-1017.
9-5
-------
Bloomfield, J.J. and W. Blum. 1928. Health hazards in chromium plating. Public
Health Rep. 43: 2330-2351.
Bohning, D.E., R.E. Albert, M. Lippman and W.M. Foster. 1975. Tracheobronchial
particle deposition and clearance. Arch. Environ. Health. 30: 457. (Cited in
U.S. EPA, 1982).
Bonatti, S., M. Meini and A. Abbondandolo. 1976. Genetic effects of potassium
dichromate in Schizosaccharomyces pombe. Mutat. Res. 38: 147-150.
Borghetti, A., A. Mutti, A. Cavatorta, M. Falzoi, F. Cigala and I. Franchini.
1977. Indices renaux d'exposition aigue et d'impregnation chronique par le
chrome. Med. Lav. 68: 355-363.
Boumanne, P.W.J.M. and F.J. deBoer. 1972. Studies of flame and plasma torch
emission for simultaneous multi-sleraent analysis. I. Preliminary investiga-
tions. Spectrochim. Acta. 27B: 391-414.
Bourne, H.G., Jr. and H.T. Yee. 1950. Occupational cancer in a chromate plant -
An environmental appraisal. Ind. Med. Surg. 19(12): 563-567.
Bovet, P., M. Lob and M. Grandjean. 1977. Sphirometric alterations in workers in
the chromium electroplating industry. Int. Arch. Occup. Environ. Health.
40(1): 25-32.
Brieger, H. 1920. Zur Klinik der akuten Chromatvergiftung. Z. Exp. Path.
Therap. 21: 393-408.
9-6
-------
Brinton, H.P., E.S. Frasier and A.L. Koven. 1952. Morbidity and mortality
experience among chromate workers. Public Health Rep. 67: 835-847.
Broch, C. 1949. Bronchial asthma caused by chromium trioxide fumes. Nord. Med.
41: 996-997. (In Swedish). Arch. Ind. Hyg. Occup. Med. 1: 588.
Broekaert, J.A.C., B. Wopenka and H. Puxbaum. 1982. Inductively coupled plasma
optical emission spectrometry for the analysis of aerosol samples collected by
cascade impactors. Anal. Chem. 54: 2174-2179.
Brown, G.C., M.D. Lockshin, E.A. Salvati and P.O. Bullough. 1977. Sensitivity
to metal as a possible cause of sterile loosening after cobalt-chromium total
hip-replacement arthroplasty. J. Bone Joint Surg. 59(2): 164-168.
Bruland, K.W., K. Bertine, M. Koide and E.D. Goldberg. 1974. History of metal
pollution in Southern California coastal zone. Environ. Sci. Technol.
8: 425-432.
Bryson, W.G. and C.M. Goodall. 1981. Improved spectrophotometric determination
of chromium in animal tissue digests with diphenylcarbazide. Anal. Chim. Acta.
124: 391-401.
Budanova, L.F. 1976. On occupational bronchial asthma in alumina production
workers. Ter. Arkh. 48(2): 30-33.
Budanova, L.F. and V.N. Marakova. 1979. Leukocyte agglomeration index in
occupational bronchial asthma resulting from exposure to hexavalent chromium.
Serd.-sousd. i. Legoch. Patol. v Klinike Prof. Zabolev., M. 48-50.
9-7
-------
Buhler, D.R. 1977. Tissue accumulation and enzymatic effects of hexavalent
chromium in rainbow trout, Salmo gajrdneri. J. Fish. Res. Board Can. 31*: 9.
Burry, J.N. and J. Kirk. 1975. Environmental dermatitis: Chrome cripples.
Med. J. Australia. 2(18): 720-721.
Cairns, R.J. and C.D. Calnan. 1962. Green tattoo reactions associated with
cement dermatitis. Brit. J. Dermatol. 71: 288-291. (Cited in NIOSH, 1975.)
Calabrese, E.J. 1978. Methodological approaches to deriving environmental and
occupational health standards. New York: John Wiley and Sons.
Camp, D.C., A.L. Van Lehn, J.R. Rhodes and A.H. Pradzynski. 1975. Intercom-
parison of trace element determination in simulated and real air particulate
samples. X-ray Spectrometry. 4: 123-137.
Capodaglio, E., G. Catenacci and G. Pezzagno. 1975. Functional conditions of
the respiratory system and worksite pollution. II. Prevalence of respiratory
function changes in workers employed in bichromate and chromic acid production.
Lav. Umano. 27(6): 175-187.
Cappuzzo, J.M. and J.J. Sasner. 1977. The effect of chromium on filtration
rates and metabolic activity of Mytilu edulis L. and Mya arenaria L. In:
Physiological Responses of Marine Biota to Pollutants. Academic Press, New York.
225 pp..
9-8
-------
Gary, E.E. 1982. Chromium in air, soil and natural waters. In: Biological and
Environmental Aspects of Chromium. S. Langard, Ed. Elsevier Biomedical Press.
Casto, B.C., J. Meyers and J.A. DiPaolo. 1979. Enhancement of viral transforma-
tion for evaluation of the carcinogenic or mutagenic potential of inorganic metal
salts. Cancer Res. 39: 193-198.
Cawse, P.A. 1974. A survey of atmospheric trace elements in the U.K.
(1972-1973). Atomic Energy Research Establishment, Harwell, England. (Cited in
Towill et al., 1978).
Chemical Marketing Reporter. May 31, 1982. 221(22): 55.
Chen, K.Y., C.S. Young, T.K. Jan and N. Rohatgi. 1974. Trace metals in waste-
water effluents. J. Water Pollut. Contr. Fed. 46: 2663-2675.
Chiu, C.H. and J.C. Hilborn. 1979. Analyst. 104: 1159. (Cited in Tinsley,
1983.)
Cikrt and Benco. 1979. J. Hyg. Epidemiol. Microbiol. Immunol. 23: 241-246.
Clausen, J. and S.C. Rastogi. 1977. Heavy metal pollution among autoworkers.
II. Cadium, chromium, copper, manganese, and nickel. Brit. J. Ind. Med.
34: 216-220.
9-9
-------
Cline, S.J., J. Glerup, D.J. Hodgson, G.S. Jensen and E. Pedersen. 1981.
Synthesis and structural and magnetic characterization of trans-aquatitraammine-
chromium(III)-p-hydroxo-pentaammine chromium(III) chloride trihydrate
[[(NH3)ll(H20)Cr(OH)Cr(NH3)5]Cl5«3H20]. Inorganic Chem. 20: 2229-2233.
Cohen, S.R. and R.S. Kramkowsi. 1973- Health Hazard Evaluation Determination,
Report No. 72-118-104. Cincinnati, Ohio, U.S. Department of Health, Education,
and Welfare, National Institute for Occupational Safety and Health, Division of
Technical Services, Hazard Evaluation Services Branch. (Cited in NIOSH, 1973.)
Cohen, S.R., D.M. Davis and R.S. Kramkowski. 197**. Clinical manifestations of
chromic acid toxicity—Nasal lesions in electroplate workers. Cutis.
13: 558-568. (Cited in NIOSH, 1975.)
Coleman, R.F., J. Herrinton and J.T. Scales. 1973. Brit. Med. J. 1: 527.
(Cited in Anonymous, 1976.)
Collins, R.J., P.O. Fromm and W.D. Collings. 1961. Chromium excretion in the
dog. Amer. J. Physiol. 201: 795.
Cotton, F.A. and G. Wilkinson. 1972. Advanced Inorganic Chemistry. Wiley
Interscience Publishers, New York. 1145 pp.
Cranston, R.E. and J.W. Murray. 1978. The determination of chromium species in
natural waters. Anal. Chem. Acta. 99: 275-282.
9-10
-------
Treason, J.P., T.A. Hinners, J.E. Bumgarner and C. Pinkerton. 1975. Trace
elements in hair, as related to exposure in metropolitan New York. Clin. Chem.
21: 603-612.
Criaui, M.H., M. Austin and E. Barrett-Conner. 1979. The effect of non-response
on risk ratios in a cardiovascular disease study. J. Chronic Dis. 32: 633-638.
Crosmun, S.T. and T.R. Mueller. 1975. The determination of chromium(VI) in
natural waters by differential pulse polargraphy. Anal. Chira. Acta.
75: 199-205.
Dams, R., K.A. Rahn and J.W. Winchester. 1972. Evaluation of filter materials
and impaction surfaces for nondestructive neutron activation analysis of
aerosols. Environ. Sci. Technol. 6: 441-450.
Danielsson, B.R.G., E. Hassoun and L. Oencker. 1982. Embryotoxicity of
Chromium: Distribution in Pregnant Mice and Effects on Embryonic Cells in vitro.
Arch. Toxicol. 51: 233-245.
Davidson, I.W.F., R.L. Burt and .J.r, Parker. 1974. Renal excretion of trace
elements: Chromium and copper. Proc. Soc. Exp. Biol. Med. 147: 720-725.
Davidson, I.W.F. and W.L. Secrist. 1972. Determination of chromium in biolo-
gical materials by atomic absorption spectrometry using a graphic furnace
atomizer. Anal. Chem. 44: 1808-1813.
9-11
-------
Davies, C.N. 1972. An algebraical model for the deposition of aerosols in the
human respiratory tract during steady breathing. J. Aerosol. Sci. 3: 297.
(Cited in U.S. EPA, 1982).
Davies, J.M. 1978. Lung-cancer mortality of workers making chrome pigments.
Lancet. 1: 38U.
Davies, J.M. 1979. Lung cancer mortality in workers in chromate pigment manu-
facture: An epidemiological survey. J. Oil Chem. Assoc. 62: 157-163.
Davis, J.M. 1972. The fibrogenic effects of mineral dusts injected into the
pleural cavity of mice. Br. J. Exp. Pathol. 53: 190-201.
Davison, R.L., D.F.S. Natusch, J.R. Wallace and C.A. Evans, Jr. 1971*. Environ.
Sci. Technol. 8: 1107-1113. (Cited in Gary, 1982.)
Day, J.P., M. Hart and M.S. Robinson. 1975. Nature. 253: 3M3. (Cited in
Tinsley, 1983.)
DeFlora, S. 1978. Metabolic activation of mutagens in the Salmonella/microsome
test. Nature. 271: 455-M56.
DeFlora, S. 1981. Study of 106 organic and inorganic compounds in the
Salmonella/Microsome Test. Carconogenesis. 2U): 283-298.
9-12
-------
DeGroot, A.J. and E. Allersma. 1973. Field observations on the transport of
heavy metals in sediments. Presented at the Conference on Heavy Metals in the
Aquatic Environment. Nashville, Tennessee, Dec. 4-7, 1973, pp. 1-16. (Cited in
Towill et al., 1978).
Delpech, M.A. and M. Hillairet. 1869. Memoire sur les accidents auxquels sont
soumis les ouvriers: Employes a la fabricationdes chromate. Ann. d'Hyg. Publ.
Med. 31: 5-30.
deMeester, P. and D.J. Hodgson. 1977. Synthesis and structural characterization
of L-histidinate-D-penicillaminato chromium(III) mohohydrate. J. Chem. Soc.,
Dalton Trans, pp. 1604-1607.
deMeester, P., D. Hodgson, H.C. Freeman and C.J. Moore. 1977. Tridentate
coordination by the L-cysteine dianion. Crystal and molecular structure of
sodium bis(L-cryteinato)chromate dihydrate. Inorg. Chem. 16: 1494-1498.
Demuynck, M., K.A. Rahn, M. Janssens and R. Dams. 1976. Chemical analysis of
airborne patriculate matter during a period of unusually high pollution. Atmos.
Environ. 10: 21-26.
Denton, C.R., M.S. Keenan and D.J. Birmingham. 1954. The chromium content of
cement and its significance in cement dermatitis. J. Invest. Dermatol. 23: 189-
192.
Deutman, R., T.J. Mulder, R. Brian and J.P. Nater. 1977. Metal sensitivity
before and after total hip arthroplasty. J. Bone Joint Surgery. 59(2):863-865.
9-13
-------
DiPaolo, J.A. and B.C. Casto. 1979. Quantitative studies of in vitro morpho-
logical transformation of Syrian hamster cells by inorganic metal salts. Cancer
Res. 39: 1008-1013.
Doisy, R.J., D.H.P. Streeten, R.A. Levine and R.B. Chodos. 1969. Effects and
metabolism of chromium in normals, elderly subjects, and diabetic. In; Trace
Substances in Environmental Health. II. D.D. Hemphill, ed. University of
Missouri, Columbia, Missouri, pp. 75-81. (Cited in U.S. EPA, 1978.)
Doisy, R.J., D.H.P. Streeten, M.L. Souma, M.E. Kalafer, S.I. Rekant and T.G.
Dalakos. 1971. Metabolism of chromium-51 in human subjects - Normal, elderly,
and diabetic subjects. In: Newer Trace Elements in Nutrition. W. Mertz and
W.E. Cornatzer, eds. Marcel Dekker, Inc., New York. pp. 155-168.
Doll, R. 1971. The age of distribution of cancer: implications for models of
carcinogenesis. J. Roy. Stat. Soc. A134: 133-166.
Doll, R. and A.B. Hill. 1964. Mortality in relation to smoking: Ten years
observation of British doctors. Br. Med. J. 1: 1399-1410.
Donaldson, R.M. and R.F. Barreras. 1966. Intestinal absorption of trace quanti-
ties of chromium. J. Lab. Clin. Med. 68: 481-193-
DuBois, A.B. and R.M. Rogers. 1968. Respiratory factors determining the tissue
concentration of inhaled toxic substances. Respir. Physiol. 5: 34. (Cited in
U.S. EPA, 1982).
9-14
-------
Duce, R.A. and W.H. Zoller. 1975. No title provided. Science. 187: 59-61.
(Cited in Gary, 1982.)
Dunstan, L.P. and E.L. Garner. 1977. Chemical preparation of biological
materials for accurate chromium determinations by isotope dilution mass spectro
metry. In: Proceeding of the Eleventh Annual Conference on Trace Substances in
Environmental Health, Columbia, MO. (Cited in Towill et al., 1978).
Edmundson, W.F. 1951. Chrome ulcers of the skin and nasal septum and their
relation to patch testing. J. Invest. Dermatol. 17: 17-19. (Cited in NIOSH,
1975).
Eisenrich, S.J. 1982. Atmospheric role in trace metal exchange at the air-water
interface. J. Great Lakes Res. 8(2): 243-256.
Elias, Z., 0. Schneider, F. Aubry, M.C. Daniere and 0. Poirot. 1983. Sister
chromatid exchanges in Chinese Hamster V79 cells treated with the trivalent
chromium compounds chromic chloride and chromic oxide. Carcinogenesis. 4(5):
605-611.
Elser, B.C. 1976. Trace Analysis: Spectroscopic Methods for the Elements.
J.D. Winefordner (Ed.). John Wiley and Sons, Inc., NY. p. 409. (Cited in
Towill et al., 1978).
Engebrigtsen, J.K. 1982. Some investigations on hypersensitiveness to bichro-
mate in cement workers. Acta Derm. Venereol. 32: 462-468.
9-15
-------
Engel, L.A., L.D. Wood, G. Utz and P.T. Macklem. 1973. Gas mixing during
inspiration. J. Appl. Physiol. 35: 18. (Cited in U.S. EPA, 1982).
Engle, H.O. and C.D. Calnan. 1963. Chromate dermatitis from paint. Brit. J.
Ind. Med. 20: 192-198. (Cited in NIOSH, 1975).
Enterline, P.E. 1972*. Respiratory cancer among chromate workers. J. Occup.
Med. 16: 523-526. (Cited in NIOSH 1975.)
Evan, A.P. and W.G. Dail. 1974. The effects of sodium chromate on the proximal
tubules of rat kidney. Lab. Invest. 30: 70M-715.
Evans, W. 1977a. Health Hazard Evaluation Determination, Report No. 76-86-389.
Allied Chemical Corporation. Cincinnati, Ohio. U.S. Department of Health,
Education, and Welfare, Public Health Service, Center for Disease Control,
NIOSH.
Evans, W.A. 1977b. Health Hazard Evaluation Determination, Report No.
76-87-389. Hercule's Inc., Cincinnati, Ohio. U.S. Department of Health, Educa-
tion, and Welfare, Public Health Service, Center for Disease Control, NIOSH.
Evans, W.A. 1977c. Health Hazard Evaluation Determination, Report No.
76-88-380. Minnesota Mining and Manufacturing Co. Cincinnati, Ohio. U.S.
Department of Health, Education, and Welfare, Public Health Service, Center for
Disease Control, NIOSH.
9-16
-------
Parkas, T.G. and S.L. Roberson. 1965. The effect of Cr^+ on the glucose
utilization of isolate lenses. Exp. Eye Res. 1: 121-126.
Federal Security Agency. 1953. Health of Workers in Chromate Producing
Industry. A Study. U.S. Public Health Service Publication No. 192. Washington,
D.C., U.S. Government Printing Office. 131 pp.
Feldman, F.C., B.C. Knoblock and W. Purdy. 1967. The determination of chromium
in biological materials by atomic absorption spectroscopy. Anal. Chim. Acta.
38: 489-^97.
Fischer, R. 1911. Die industrielle Herstellung and Verwendung der Chromver-
bindugen, die dabei entstehenden Gesundheitsgefahren fur die Arbeiter und.
(Cited in NIOSH, 1975).
Fischer-Wasels, B. 1938. Das primare Lungencarcinom. Acta Un. Int. Cancr. 3:
1MO-152.
Fiscus, D.E., P.O. Gorman, M.P. Schrag and L.J. Shannon. 1978. Assessment of
bacteria and virus emissions at a refuse derived fuel plant and other waste
handling facilities. EPA Contract No. 68-02-1871, prepared for Municipal
Environmental Research Laboratory, Office of Research and Development, U.S. EPA,
Cincinnati, Ohio.
Fisher., A.A. 1977. Allergen replacement of allergic dermatitis. Internat. J.
Dermatol. 16(5): 319-328.
9-17
-------
Fornace, A.J., Jr., D.S. Seres, J.F. Lechner and C.C. Harris. 1981. DNA-protein
cross-linking by chromium salts. Chem. Biol. Interact. 36: 315-351.
Fradkin, A., A. Janoff, B.P. Lane and M. Kuschner. 1975. In vitro transforma-
tion of BHK 21 cells grown in the presence of calcium chromate. Cancer Res. 35:
1058-1063.
Franchini, I., A. Mutti, A. Cavatorta et al. 1978. Nephrotoxicity of chromium.
Remarks on an experimental and epidemiological investigation. Contr. Nephr. 10:
98-119. (Cited in Mutti et al., 1979).
Franchini, I., A. Mutti, F. Gardini and A. Borghetti. 1975. Excretion et
clearance renale du crhome par rapport au degre et a duree de le exposition.
Rein et Toxiques 271-278. Paris, Masson. (Cited in Mutti et al., 1979).
Fraser, J.L. and K.R. Lum. 1983. Availability of elements of environmental
importance in incinerated sludge ash. Environ. Sci. Technol. 17: 52-51.
Fregert, S. 1961. Chromate eczema and matches. Acta Dermatol. Venereol. 11:
133-1^2.
Fregert, S. and B. Gruvberger. 1976. Chromate dermatitis from zinc chromate
used for marking textiles. Contact Dermatitis. 2(2): 121.
Fregert, S. and P. Ovrum. 1963. Chromate in welding fumes with special
references to contact dermatitis. Acta Dermatol. Venereol. 13: 119-121.
9-18
-------
Fremm, P.O. and P.M. Stokes. 1962. Assimilation and metabolism of chromium by
trout. J. Water Pollut. Contr. Fed. 34: 1151.
Frenkiel, S. and P.W. Albert. 1976. Traumatic injury of the middle ear. J.
Otolargyngology. 6(1): 17-22.
Frentzel-Beyme, R. 1983. Lung cancer mortality of workers employed in chromate
pigment factories. J. Cancer Res. Clin. Oncol. 105:183-188.
Freund, H., S. Atamian and J.E. Fisher. 1979. Chromium deficiency during total
parenteral nutrition. J. Am. Med. Assoc. 211: 196-498.
Fukai, R. 1967. Valency state of chromium in seawater. Nature. 213: 901.
Fukunaga, M., Y. Kurachi and Y. Mizuguchi. 1982. Action of Some Metal Ions on
Yeast Chromosomes. Chem. Pharm. Bull. 30: 3017-3019.
Furst, A., M. Schlauder and D.P. Sasmore. 1976. Tumorigenic activity of lead
chromate. Cancer Res. 36: 1779-1783.
Gafafer, W.M. 1953- U.S. Fed. Public Health Serv. Pub. 192. Washington, DC.
(Cited in Cary, 1982.)
Gale, T.F. 1978. Embryotoxic effects of chromium trioxide in hamster. Environ.
Res. 16: 101-109.
9-19
-------
Gale, T.F. and J.D. Bunch. 1979. The effect of the time of administration of
chromium trioxide on the embryo toxic response in hamster. Teratology.
19: 81-86.
Gale, T.F. 1982. The embryotoxio response to maternal chromium trioxide expo-
sure in different strains of hamsters. Environ. Res. 29: 196-203.
Gallorini, M., M. DiCasa, R. Stella, N. Geneva and E. Orvini. 1976. Multiele-
ment trace analysis by atomic absorption spectometry and neutron activation
analysis in biological matrices. J. Radioanal. Chem. 32: 17-23.
GCA Corporation. 1973. National Emissions Inventory of Sources and Emissions of
Chromium. EPA U50/3-74-012, U.S. EPA, Research Triangle Park, North Carolina,
p. 33. [Cited in IRPTC (International Register of Potentially Toxic Chemicals),
1979. United Nations Environment Programme, Geneva, Switzerland.]
Gibbs, R.J. 1977. No title provided. Geol. Soc. Am. Bull. 88: 829-8H3. (Cited
in Gary, 1982.)
Gilbert, T.R. and A.M. Clay. 1973. Determination of Chromium in Sea Water by
Atomic Absorption Speotrometry. Anal. Chiro. Acta. (Netherlands). 67: 289-295.
Glinsmann, W.H. and W. Mertz. 1966. Effect of trivalent chromium on glucose
tolerance. Metabolism. 15: 510-520.
Goldberg, A.J. 1973. A survey of emissions and controls for hazardous and other
pollutants, EPA-Rl-73-021, U.S. Environmental Protection Agency, Washington,
D.C., 168 pp.
9-20
-------
Golden, H.H. and B.E. Golden. 1981. Trace elements: Potential importance in
human nutrition with particular reference to zinc and vanadium. Brit. Med. J.
37(1): 31-36.
Goldman, M. and R.H. Karotkin. 1935. Acute potassium bichromate poisoning.
Amer. J. Med. Sci. 189: 400-103.
Gomes, E.R. 1972. Incidence of chromium-induced lesions among electroplating
workers in Brazil. Ind. Med. 41: 21-25.
Gomez-Arroyo, S., M. Altamirano and R. Villalobos-Pietrini. 1981. Sister-
chromated exchanges induced by some chromium compounds in human lymphocytes in
vitro. Mutation Research. 90: 425-431.
Gosink, T.A. 1976. Rapid simultaneous analysis of ficogram quantities of
aluminum and chromium from water by gas pahse chromatography. Anal. Chem.
47: 165-168.
Gray, S.J. and K. Sterling. 1950. The tagging of red cells and plasma proteins
with radioactive chromium. J. Clin. Invest. 29: 1604-1613.
Greathouse, D.G. and G.F. Craun. 1978. Cardiovascular disease study - Occurence
of inorganics in household tap water and relationships to cardiovascular
mortality rates. Trace substances in Environ. Health - XII. 1978. A Symposium.
D.D. Hempill (Ed). University of Missouri, Columbia, pp. 31-39.
9-21
-------
Green, M.H.L., W.J. Muriel and B.A. Bridges. 1976, Use of a simplified fluctua-
tion test to detect low levels of mutagens. Mutat. Res. 38: 33-42.
Greenberg, R.R., W.H. Zoller and G.E. Gordon. 1978. Composition and site
distribution of particles released in refuse incineration. Environ. Sci.
Technol. 12: 566-573.
Gresh, J.T. 1944. Chromic acid poisoning resulting from inhalation of mist
developed from 5% chromic acid solution. II. Engineering aspects of chromic
acid poisoning from anodizing operations. J. Ind. Hyg. Toxicol. 26: 127-130.
(Cited in NAS, 1974.)
Gross, P.R., S.A. Katz and M.H. Samitz. 1968. Sensitization of guinea pigs to
chromium salts. J. Invest. Dermatol. 50: 424-427.
Gross, W.G. and U.G. Heller. 1946. Chromium in animal nutrition. J. Ind. Hyg.
and Toxicol. 28: 52.
Gurson, C.T. and G. Saner. 1971. Effect of chromium on glucose utilization in
marasmic protein-calorie malnutrition. Amer. J. Clin. Nutr. 24: 1313-1319.
Guthrie, B.E., W.R. Wolf and C. Veillon. 1978. Background correction and
related problems in the determination of chromium in urine by graphite furnace
atomic absorption spectrometry. Anal. Chem. 50: 1900-1902.
Gylseth, B., N. Gundersen and S. Langard. 1977. Evaluation of chromium exposure
based on a simplified method for urinary chromium determination. Scand. J. Work
Environ. Health. 3: 28-31.
9-22
-------
Hall, A.F. 194M. Occupational contact dermatitis among aircraft workers. J.
Amer. Med. Assoc. 125: 179-185. (Cited in NIOSH, 1975.)
Hambidge, K.M. 1971. Chromium nutrition in the mother and growing child. In:
Newer Trace Elements in Nutrition. W. Mertz and W.E. Cornatzer, eds. Marcel
Dekker, Inc., New York. pp. 169-191*. (Cited in U.S. EPA, 1978.)
Hambidge, K.M. and W. Droegnueller. 1974. Changes in plasma and hair concentra-
tions of zinc, copper, chromium and manganese during pregnancy. Obstet. Gynecol.
M: 666-672.
Hambidge, K.M. and D.O. Rodgerson. 1969. Comparison of hair chromium levels of
nulliparous and parous women. Amer. J. Obstet. Gynecol. 103(3): 370-371.
(Cited in U.S. EPA, 1978.)
Hambidge, K.M., M.L. Franklin and M.A. Jacobs. 1972. Changes in hair chromium
concentrations with increasing distances from hair roots. Amer. J. Clin. Nutr.
25: 380-383. (Cited in U.S. EPA., 1978.)
Hambidge, K.M. 1981. "Chromium." In: Disorders of Mineral Metabolism. F.
Bronner and J.W. Coburn, Eds. Acad. Press, p. 271-295.
Hamilton, E.I., M.J. Minski and J.J. Cleary. 1973. The concentration and
distribution of some stable elements in healthy human tissues from the United
Kingdom. Sci. Total Environ. (Netherlands). 1: 341-37M. (Cited in U.S. EPA,
1978.)
9-23
-------
Hanslian, L., J. Navratil, J. Jurak and M. Kotrle. 1967. Upper respiratory
tract lesions from chromic acid aerosol. Pracovni. Lekar. 19: 29*1-298.
(Czech.) (Cited in NAS, 197U.)
Hartford, W.H. 1979. Chromium compounds. In; Kirk-Othmei- Encyclopedia of
Chemical Technology, 3rd ed., volume 6. M. Grayson, ed. John Wiley and Sons,
1979, PP. 82-120.
Hayes, B.B., A.M. Lilienfeld and L.M. Snell. 1979. Mortality in chromium
chemical production workers: A prospective study. Int. J. Epidemiol.
8: 365-374.
Hoyt, S.D. and J.D. Ingle, Jr. 1976. A chemiluminescence photometer from trace
chromium(IIT) determinations. Anal. Chim. Acta. 87: 168-175.
Hicks, R., P.J. Hewitt and H.F. Lava. 1979. An investigation of the experimen-
tal induction of hypersensitivity in the guinea pig by material containing chro-
mium, nickel, and cobalt from arc welding fumes. Internat. Arch. Allergy Appl.
Tmmunol. 59(3): 265-272.
Hill, W.J. and W.S. Ferguson. 1979. Statistical analysis of epidemiological
data from chromium chemical manufacturing plant. J. Occup. Med. 21: 103-106.
Hopkins, H.F. 1965. Distribution in the rat of physiological amounts of
injected Cr-51 (III) with time. Amer. J. Physiol. 209: 731-735.
9-2H
-------
Hopkins, L.L. and K. Schwarz. 1964. Chromium (ITT) binding to serum proteins,
specifically siderophilin. Biochem. Biophys. Acta. 90: 481-191.
Hopkins, L.L., Jr., 0. Ransome-Kuti and A.S. Majaj. 1968. Improvement of
impaired carbohydrate metabolism by chromium(III) in malnourished infants.
Amer. J. Clin. Nutr. 21: 203-211.
Hueper, W.C. 1955. Experimental studies in metal carcinogenesis. VII. Tissue
reactions to parenterally introduced powdered metallic chromium and chromite
ore. J. Natl. Cancer Instit. 16(2): 447-469.
Hueper, W.C. 1958. Experimental studies in metal cancerigenesis. X. Canceri-
genic effects of chromite ore roast deposited in muscle tissue and pleural cavity
of rats. Amer. Med. Assoc. Arch. Ind. Health. 18: 284-291.
Hueper, W.C. 1961. Environmental carcinogenesis and cancers. Cancer Res.
21: 842-857.
Hueper, W.C, and W.W. Payne. 1959. Experimental cancers in rats produced by
chromium cotnpunds and their significance to industry and public health. Ind.
Hyg. J. 20: 274-280.
Hueper, W.C. and W.W. Payne. 1962. Experimental studies in metal caroinogene-
sis: 'Chromium, nickel, iron, arsenic. Arch. Environ. Health. 5: 445-462.
Hunter, W.C. and J.M. Roberts. 1933. Experimental study of the effects of
potassium bichromate on the monkey's kidney. Amer. J. Pathol. 9: 133-147.
9-25
-------
Husain, S.L. 1977. Contact dermatitis in the West of Scotland. Contact
Dermatitis. 3(6): 327-332.
Hyado, K., S. Suzuki, N. Furuga and K. Meshizuka. 1980. An analysis of
chromium, copper, and zinc in organs of a chromate worker. Int. Arch. Occup.
Environ. Health. 46: 141-150.
IARC (International Agency for Research on Cancer). 1980. IARC monographs on
the evaluation of the carcinogenic risk of chemicals to humans: Some metals and
metallic compounds. Lyon, France: World Health Organization. 23: 205-323.
lijima, S., N. Matsumoto, C.C. Lu and H. Katsunuma. 1979. Embryotoxic and
fetotoxic effects of chromium trioxide in mice. Teratology. 20: 152.
lijima, S., N. Matsumoto and C. Lu. 1983. Transfer of chromic chloride to
embryonic mice and changes in the embryonic mouse neuroepithelium. Toxicology.
26: 257-265.
Imbus, H.R., J. Cholak, L.H. Miller and T. Sterling. 1963. Boron, cadmium,
chromium, and nickel in blood and urine. Arch. Environ. Health. 6: 112-121.
(Cited in U.S. EPA, 1978.)
Ivankovic, S. and R. Preussman. 1975. Absence of toxic and carcinogenic effects
after administrations of high doses of chromic oxide pigment in subacute and long
term feeding experiments in rats. Food Cosmet. Toxicol. 13: 347-351.
9-26
-------
Iwata, Y., K. Matsumoto, H. Haraguchi, K. Fuwa and K. Okamoto. 1981. Proposed
certified reference material for pond sediment. Anal. Chera. 53: 1136-1138.
Jaeger H. and E. Pelloni. 1950. Skin test to bichromates, positive with cement
eczema. Dermatologica. 100: 207-216. (Cited in NIOSH, 1975.)
James, B.R and R.J. Bartlett. 1983a. Behavior of chromium in soils: V. Fate of
organically complexed Cr(III) added to soil. J. Environ. Qual. 12(2): 169-172.
James, B.R. and R.J. Bartlett. 1983b. Behavior of chromium in soils: VII.
Adsorption and reduction of hexavalent forms. J. Environ. Qual. 12(2): 177-181.
Jansen, L.H. and L. Berrens. 1968. Sensitization and partial desensitization of
guinea pigs to trivalent and hexavalent chromium. Dermatologica (Switzerland).
137: 65-73.
Jeejeebhoy, K.N., R.C. Chu, E.B. Marliss, G.R. Greenberg and A. Bruce-Robertson.
1977. Chromium deficiency, glucose intolerance, and neuropathy reversed by
chromium supplementation in a patient receiving long-term total parenteral
nutrition. Amer. J. Clin. Nutr. 30: 531-538.
Jennette, K.W. 1982. Microsomal reduction of the carcinogen chromate produced
chromium(V). J. Am. Chem. Soc. 104: 871-875.
Jennings, M.E. and J.M.H. Howard. 1980. Chromium, wine, and uschaemic heart
disease. The Lancet. July 12, 1980. pp. 90-91.
9-27
-------
Jennrich, R.I. and R.H. Moore. 1975. Maximum likelihood estimation by means of
nonlinear least squares. Proceedings of Statistical Computing, Sect. 1,
American Statistical Association.
Jervis, R.E., S. Landsberger, S. Aufreiter, J.C. Van Loon, R. Lecomte and S.
Monaro. 1983. Trace elements in wet atmospheric deposition: Application and
comparison of PIXE, INAA, and Graphite-Furnace AAS techniques. Intern. J.
Environ. Anal. Chem. 15: 89-106.
John, W.R., K. Kaifer, K. Rahn and J.J. Wesolowski. 1973. Atmosph. Environ.
7: 107-118.
Jovovic, D., A. Sofronic and G. Trobozic. 1977. Allergic sensitivity in shoe-
makers. Dermatovenerol. Ingoslav. 4(1): 24.
Kaaber, K. and N.K. Deien. 1977. Significance of chromate ingestion in patients
allergic to chromate. Acta Dermato-Venereologica. 57: 321-323.
Kamiya, N., Y. Sakamoto, T. Nakao, and K. Hiraga. 1981. Inhalation experiment
of chromite ore residue dust to rats for 7 days (II) analysis of chromium in rat
tissues. Ann. Rep. Tokyo Metr. Res. Lab. P.H. 32(2): 85-88.
Kanematsu, N., M. Hara and T. Kada. 1980. Rec assay and mutagenicity studies on
metal, compounds. Mutat. Res. 77: 109-116.
Kayne, F.J., G. Komar, H. Laboda and R.E. Vanderlinde. 1978. Atomic absorption
spectrophotometry of chromium in serum and urine with a modified Perkin-Elmer 603
atomic absorption spectrophotometer. Clin. Chem. 24: 2151-2154.
9-28
-------
Kiilunen, M., H. Kivisto, P. Ala-Laurila, A. Tossavainen, A. Aito. 1983.
Exceptional pharmacokinetics of trivalent chromium during occupational exposure
to chromium lignosulfonate dust. Scand. J. Work Env. Health. 9: 265-271.
Kirschbaum, B.B., P.M. Sprinkel and D.E. Oken. 1981. Proximal tubule brush
border alterations during the course of chromate nephropathy. J. Toxicol. Appl.
Pharmacol. 58: 19-30.
Kitagawa, S., H. Seki, F. Kametani and H. Sakurai. 1982. Uptake.of hexavalent
chromium by bovine erythrocytes and its interaction with cytoplasmic components;
the role of glulathione. Chem.-Biol. Interactions. 40: 265-271.
Kleinfeld, M. and A. Rosso. 1965. Ulcerations of the nasal septum due to
inhalation of chromic acid mist. Ind. Med. Surg. 31: 212-213.
Kneebone, B.M. and H. Freiser. 1975. Determination of chromium(VI) in indus-
trial atmospheres by a catalytic method. Anal. Chem. 17: 595-598.
Kobayashi, H. 1976. Toxicological studies on chromium. Acute toxicities of
chromium trioxide. Tokyo Toritsu Eisei Kenkyusho Kenkyu Nempo. 27: 119-123.
Koelsch, F. 1938. Lungenkrebs und Beruf. Acta Un. Int. Cancr. 3: 213-252.
(Cited in NAS, 1971.)
Kopp, J.F. and R.C. Kroner. 1967. Trace metals in waters of the United States.
U.S. Department of the Interior, Cincinnati, Ohio, 28 pp. (Cited in Towill et
al.f 1978).
9-29
-------
Korallus, V., H. Lange, A. Ness, E. Wuestefeld, and T. Zwingers. 1982. Rela-
tionships between precautionary measures and bronchial carcinoma mortality in
the chromate-producing industry. Arbeitsmedizin, Socialmedizin, Preventiv-
medizin 17(7): 159-167.
Kowalczyk, G.S., G.E. Gordon and S.W. Rheingrover. 1982. Identification of
atmospheric particulate sources in Washington, D.C., using chemical element
balances. Environ. Sci. Technol. 16: 79-90.
Krishna, G., J.S. Mathur and R.K. Gupta. 1976. Health hazard amongst chrome
workers with special reference to nasal septum perforation. Indian J. Med. Res.
64(6): 866-872.
Kuennen, R.W., K.A. Woinik, F.L. Fricke and J.A. Caruso. 1982. Pressure disso-
lution and renal sample matrix calibration for multielement analysis of raw
agricultural crops by inductively coupled plasma atomic emission spectrometry.
Anal. Chem. 54: 2146-2150.
Kuhn, J.K. 1973. Trace elements in whole coal determined by X-ray fluorescence.
Norelco Rep. 20(3): 7-10. (Cited in Towill et al., 1978).
Kumpulainen, J., E. Vuori, S. Mukinen and R. Kara. 1980. Dietary chromium
intake of lactating Finnish mothers: effect on the Cr content of their breast
milk. Br. J. Nutr. 44: 257-263.
Kumpulainen, J.T., W.R. Wolf, C. Veillin and W. Mertz. 1979. Determination of
chromium in selected United States diets. J. Agric. Food Chem. 27: 490-494.
9-30
-------
Kushnaja, A.P. and M.S. Fege. 1982. Induction of chromosomal aberrations in
fish Boleophthalmus dussumieri after exposure in vivo to mitomycin C and heavy
metals, mercury, selenium, and chromium. Mutat. Res. 102: 71-82.
Lane, B.P. and M.J. Mass. 1977. Carcinogenicity and cocarcinogenicity of
chromium carbonyl in heterotopic tracheal grafts. Cancer Res. 37: 1476-1479.
Langard, S. 1979. The time-related subcellular distribution of chromium in the
rat liver cell after intravenous administration of disodium chromate (chromium-
51). Bio. Trace Elem. Res. 1: 15-54. (Chem. Abst. 72: 887395).
Langard, S. and T. Norseth. 1975. A cohort study of bronchial carcinomas in
workers producing chromate pigments. Br. J. Ind. Med. 32: 62-65.
Langard, S., A. Andersen and B. Gylseth. 1980. Incidence of cancer among
ferrochromium and ferrosilicon workers. Br. J. Ind. Med. 114-120.
Langard, S., N. Gundersen, D.L. Tsalev and B. Gylseth. 1978. Whole blood
chromium level and chromium excretion in the rat after zinc chromate inhalation.
Actu. Pharmacol. Toxicol. 42: 142-149.
Lantzy, R.J. and F.T. Mackenzie. 1979. Atmospheric trace metals: Global cycles
and assessment of man's impact. Geochim. Cosraochim. Acta. 43: 511-525.
Larochelle, J.H. and D.C. Johnson. 1978. Chromatographic determination of
chromium(VI) with coulometric detection based on the electrocatalysis by
adsorbed iodine of the reduction at platinum electrodes in acidic solutions.
Anal. Chem. 50: 240-243.
9-31
-------
Laskin, S. 1972. Research in Environmental Sciences, Washington D.C. Institute
of Environmental Medicine, pp. 92-97.
Laskin, S., M. Kuschner and R.T. Drew. 1970. Studies in pulmonary carcino-
genesis. Pages 327-350 in M.G. Hanna, Jr., P. Nettesheim, and J.R. Gilbert,
Eds. Inhalation carcinogenesis. U.S. Atomic Energy Comm. Symp. Series.
18: 327-350.
Lautner, G.M., J.C. Carver and R.B. Konzer. 1978. Measurement of chromium(VI)
and chromium(III) in stainless steel welding fumes with electron spectroscopy
for chemical analysis and neutron activation analysis. Am. Ind. Hyg. J. 39: 8.
Lee, R.E., Jr. and D.J. von Lehmden. 1973. Trace metal pollution in the
environment. J. Air Pollut. Contr. Assoc. 23: 853-857.
Legge, T.M. 1902. The lesions resulting from the manufacture and uses of
potassium and sodium bichromate. pp. 447-454. In: Dangerous Trades.
T. Oliver, Ed. London, J. Murray, 1902. (Cited in HAS, 1974.)
Lehmann, K.B. 1914. Die Bedeutung der Chromate fur die Gesundheit der Arbeiter.
Zentralbl. Gewerbehyg. 2: 193-195. (Cited in NAS, 1974.)
Lehmann, K.B. 1932. 1st Grund zu einer besonderen Beunruhigung wegen des
Autftretens von Lungenkrebs bei Chromatargeitern vorhanden? Zentralbl.
Gewerbehyg. 9: 168-170. (Cited in NAS, 1974.)
9-32
-------
Leonard, A. and R.R. Lauwery. 1980. Carcinogenicity and mutagenicity of
chromium. Mutat. Res. 76: 227-239.
Letterer, E., K. Neidhardt and H. Klett. 1944. Chromatlungenkrebs und Chromat-
staublunge. Bine klinische, pathologisch-anatomische und gewerbe-hygienische
Studie. Arch. Gewergepath. Gewerbehyg. 12: 323-361. (Cited in NAS, 1974.)
Levantine, A.V. 1974. Sensitivity to metal dental plates. Proc. Royal Soc.
Med. 67(10): 1007.
Levin, H.M., M.J. Brunner and H. Rattner. 1959. Lithographer's dermatitis. J.
Amer. Med. Assoc. 169: 566-569. (Cited in NIOSH, 1975.)
Levine, R.A., D.P.H. Streeten and R.J. Doisy. 1968. Effects of oral chromium
supplementation on the glucose tolerance of elderly human subjects. Metabolism.
17: 114-125.
Levis, A.G. and F. Majone. 1979- Cytotoxic and clastogenic effects of soluble
chromium compounds on mammalian cell cultures. Brit. J. Cancer. 40: 523-533.
Levis, A.G., V. Bianchi, G. Tamino and B. Pegoraro. 1978. Effects of potassium
dichromate on nucleic acid and protein syntheses and on precursor uptake in BHK
fibroblasts. Cancer Res. 38: 110-116.
Levy/L.S., and P.A. Martin. 1983 (July). The effects of a range of chromium
material on rat lung (draft). Sponsored by Dry Color Manufacturers Association
and others. Unpublished.
9-33
-------
Levy, L.S., and S. Venitt. 1975. Carcinogenic and mutagenic activity of
chromium containing materials. Br. J. Cancer 32:254-255.
Leyden, D.E., R.E. Channell and C.W. Blount. 1972. Determination of microgram
quantities of chromium(VI) and/or chromium(III) by X-ray fluorescence: Applica-
tion of a chelating ion-exchange resin as sample matrix. Anal. Chem.
44: 607-610.
Lim, T.H., T. Sargent III and N. Kusubov. 1983. Kinetics of trace element
Chromium(III) in the human body. Am. J. Physiol. 244: R445-R454.
Lin, Shan-Meei. 1983. Determination of trace elements in human whole blood by
instrumental neutron activation analysis. Radioisotopes. 32: 155-162.
Littorin, M., B. Hogstedt and B. Stromback, A. Karlsson, H. Welinder, F. Mitelman
and S. Skerfving. 1983. No cytogenetic effects in lymphocytes of stainless
steel welders. Scand. J. Work Environ. Health. 9: 259-264.
Loeb, L.A., M.A. Sirover and S.S. Agarwal. 1977. Infidelity of DNA synthesis as
related to mutagenesis and carcinogenesis. Adv. Exp. Biol. Med. 91: 103-115.
Loewenthal, L.J.A. 1960. Reactions in green tattoos—The significance of the
valence state of chromium. Arch. Dermatol. 82: 237-243. (Cited in NIOSH,
1975.)
Lofroth, G. 1978. The mutagenicity of hexavalent chromium is decreased by
microsomal metabolism. Naturivissenschaften. 65: 207-208.
9-34
-------
Lovett, R.J. and G.F. Lee. 1976. Analysis of chromium in natural waters by gas
chromatography. Environ. Sci. Technol. 10: 67-71.
Lu, J.C.S. and K.Y. Chen. 1976. Migration of trace metals in interfaces of
seawater and polluted surficial sediments. Environ. Sci. Technol. 11: 17*1-182.
Lucas, J.B. and R.S. Kramkowski. 1975. Health Hazard Evaluation Report No. 74-
87-221. Cincinnati, Ohio. Health Hazard Evaluation Branch, U.S. Department of
Health, Education, and Welfare, Public Health Service Center for Disease
Control. National Institute for Occupational Safety and Health.
Lukanin, W.P. 1930. Zur Pathologie der Chromat-Pneumokoniose. Arch. Hyg. Bakt.
101: 166-174. (Cited in NAS, 1974.)
Lumio, J.S. 1953. Lesions of the upper respiratory tract in chromium platers.
Nord. Hyg. Tidskr. 5-6: 86-91. (in Swedish) [Amer. Med. Assoc. Arch. Ind. Hyg.
8: 387, 1953 (abstract)]. (Cited in NAS, 1974.)
Machle, W. and F. Gregorius. 1948. Cancer of the respiratory system in the
United States chromate-producing industry. Public Health Rep.
63(35): 1114-1127. (Cited in National Academy of Sciences 1974.)
MaeKenzie, R.D., R.A. Anwar, R.U. Byerrum and C.A. Hoppert. 1959. Absorption
and distribution of Cr in the albino rat. Arch. Biochem. Biophys.
79: 200-205.
9-35
-------
MacKenzie, R.D., R.U. Byerrum, C.F. Decker, C.A. Hoppert and F.L. Langham. 1958.
Chronic toxicity studies. II. Hexavalent and trivalent chromium administered in
drinking water to rats. Amer. Med. Assoc. Arch. Ind. Health. 18: 232-231.
Maenhout, W. and W.H. Zoller. 1977. No title provided. J. Radioanal. Chem. 37:
637-650. (Cited in Cary, 1982.)
Major, R.H. 1922. Studies on a case of chromic acid nephritis. Bull. John
Hopkins Hosp. 33: 56-61. (Cited in NAS, 1974.)
Mali, J.W.H. 1963. Some aspects of the behavior of chromium compounds in the
skin. J. Invest. Dermatol. 11: 111-122.
Maltoni, C. 1974. Occupational carcinogenesis. Excerpta Med. Int. Congr. Ser.
322: 19-26.
Maltoni, C. 1976. Predictive value of carcinogenesis bioassays. Ann. NY Acad.
Sci. 271: 131-443.
Mancuso, T.F. 1951. Occupational cancer and other health hazards in a chromate
plant. A medical appraisal. II. Clinical and toxicologic aspects. Ind. Med.
Surg. 20: 393-407. (Cited in NAS, 1974.)
Mancuso, T.F. 1975. International Conference on Heavy Metals in the Environ-
ment. Toronto, Ontario, Canada. Oct. 27-31. (Cited in Towill et al., 1978).
9-36
-------
Mancuso, T.F. and W.C. Hueper. 1951. Occupational cancer and other health
hazards in a chromate plant: A medical appraisal. I. Lung cancers in chromate
workers. Ind. Med. Surg. 20: 358-363. (Cited in NIOSH, 1975.)
Marino, D.F. and J.D. Ingle, Jr. 1981. Determination of chromium(VI) in water
by lophine cherailumirescence. Anal. Chem. 53: 29^-298.
Markel, H.L., Jr. and J.B. Lucas. 1973. Health Hazard Evaluation Report 72-106.
Cincinnati, Ohio, U.S. Department of Health, Education, and Welfare, National
Institute for Occupational Safety and Health, Division of Technical Services,
Hazard Evaluation Services Branch, 1973. (Cited in NIOSH, 1975.)
Mathur, A.K., S.V. Chandra and S.K. Tandon. 1977. Comparative toxicity of
trivalent and hexavalent chromium to rabbits. III. Morphological changes in some
organs. Toxicology. 8: 53-61.
Matsumoto, N., S. lijima and H. Katsunuma. 1976. Placental transfer of chromic
chloride and its teratogenic potential in embryonic mice. J. Toxicol. Sci.
2: 1-13.
McCabe, L.J., J.M. Symons, R.D. Lee and G.G. Robeek. 1970. Survey of community
water supply systems. J. Amer. Water Works Assoc. 62: 670-687.
McCann, J., E. Choi, E. Yanasaki and B.N. Ames. 1975. Detection of carcinogens
as mutagens in the Salmonella/microsome test: Assay of 300 chemicals. Proc.
Natl. Acad. Sci. 72: 5135-5139.
9-37
-------
McClendon, L.T. 1974. Method measures trace chromium. Chem. Eng. New. 52: 23.
McCord, C.P., H.G. Higginbotham and J.C. McGuire. 1930. Experimental chromium
dermatitis. J. Amer. Med. Assoc. 94: 1043-1014. (Cited in NIOSH, 1975.)
McKeague, J.A. and M.S. Wolynetz. 1980. No title provided. Geoderma. 21: 299-
307. (Cited in Gary, 1982.)
Meranger, J.C. 1970. The heavy metal content of fruit juices and carbonated
beverages by atomic absorption spectrophotometry. Bull. Environ. Contam.
Toxicol. 5: 271-275.
Meranger, J.C. and E. Somers. 1968. Determination of the heavy metal content of
sea-foods by atomic absorption spectrophotometry. Bull. Environ. Contam.
Toxicol. 3: 360-365.
Meranger, J.C., K.S. Subramanian and C. Chalifoux. 1979. A national survey for
cadmium, chromium, copper, lead, zinc, calcium, and magnesium in Canadian
drinking water supplies. Environ. Sci. Technol. 13: 707-711.
Meranger, J.C., K.S. Subramanian and C. Chalifoux. 1981. Survey of cadmium,
chromium, copper, nickel, lead zinc, calcium, and magnesium in Caladian drinking
water supplies. J. Assoc. Off. Anal. Chem. 64: 44-53.
Mertz, W. 1969. Chromium occurrence and function in biological systems.
Physiol. Rev. 49(2): 163-239.
9-38
-------
Mertz, W. 197M. The newer essential trace elements, chromium, tin, vanadium,
nickel and silicon. Proc. Nutr. Soc. 33: 307-313.
Mertz, W. and E.E. Roginski. 1971. Chromium metabolism. The glucose tolerance
factor. In: New Trace Elements in Nutrition. W. Mertz and W.E. Cornatzer, eds.
Marcel Dekker, Inc. New York. pp. 123-153. (Cited in NAS, 1974.)
Mertz, W., E.E. Roginski, F.J. Leldmand and D.E. Thurman. 1969. Dependence of
chromium transfer into the rat embryo on the chemical form. J. Nutr. 99: 363-
367.
Mertz, W., E.E. Roginski and R.C. Reba. 1965. Biological activity and fate of
trace quantities of intravenous chromium (III) in the rat. Amer. J. Physiol.
209:
Mertz, W. and K. Schwarz. 1959. Relation of glucose tolerance factor to
impaired glucose tolerance in rats on stock diets. Amer. J. Physiol.
196: 611-618.
Metternich, P., H.W. Georgii and K.O. Groeneveld. 1981. Elemental Particle Size
Distributions of the Tropospheric Aerosol, Determined by PIXE. In: J. Herbert,
Ed. Atmospheric Trace Constituents. Proceedings of a Colloqium in Mainz,
Federal German Republic.
Meyers, J.B. 1950. Acute pulmonary complications following inhalation of chro-
mic acid mist. Preliminary observations of two patients who inhaled massive
amounts of chromic acid. Amer. Med. Assoc. Arch. Ind. Hyg. Occup. Med.
2: 712-717. (Cited in NAS, 1971.)
9-39
-------
Mineral Commodity Summaries. 1980. U.S. Bureau of Mines. Department of the
Interior, p. 34. (Cited in Towill et al., 1978).
Moattar, P., H. Rahimi and M. Abedini. 1982. Determination of some toxic trace
elements in airborne particulate matter using neutron activation analysis.
Radiochem. Radioanal. Letters. 50(5): 269-274.
Morning, J.L, 1977. Chromium - 1977. U.S. Bureau Mines Report, MCP-1, May
1977. (Cited in Towill et al., 1978).
Morris, G.E. 1955. Chromate dermatitis from chrome glue and other aspects of
the chrome problem. Arch. Ind. Health. 11: 368-371. (Cited in NIOSH, 1975.)
Morrow, P.E., D.V. Pates, B.R. Fish, T.F. Hatch and T.T. Mercer. 1966. Inter-
national commission on radiological protection task group on lung dynamics,
deposition and retention models for internal dosimetry of the human respiratory
tract. Health Phys. 12: 173-207. (Cited in U.S. EPA, 1982).
Murphy, E.W., L. Page and B.K. Watt. 1971. Trace minerals in type A school
lunches. J. Amer. Diet. Assoc. 58: 115-122.
Murthy, O.K., U. Rhea and J.T. Peeler. 1971. Levels of antimony, cadmium,
chromium, cobalt, manganese, and zinc in institutional total diets. Environ.
Sci. Technol. 5: 436-442.
Mutti, A., A. Cavatorta, C. Pedroni, A. Borghi, C. Giaroli and I. Franchini.
1979. The role of chromium accumulation in the relationship between airborne and
urinary chromium in welders. Int. Arch. Occup. Environ. Health. 43: 123-133.
9-40
-------
Nakamuro, K., K. Yoshikawa, Y. Sayato and H. Kurata. 1978. Comparative studies
of chromosomal aberration and mutagenicity of trivalent and hexavalent chromium.
Mutat. Res. 58: 175-181.
NAS (National Academy of Sciences). 1974. Medical and Biological Effects of
Environmental Pollutants: Chromium. Washington, D.C. National Academy Press.
NAS (National Academy of Sciences). 1980. Recommended dietary allowances. 9th
rev. ed. Food and Nutrition Board. National Academy of Sciences, Washington,
D.C.
Natusch, D.F.S. and J.R. Wallace. 1971. No title provided. Science. 186: 695-
699. (Cited in Gary, 1982.)
Nettesheim, P., M.G. Hanna, Jr., D.G. Doherty, R.F. Newell and A. Hellman. 1971.
Effect of calcium chromate dust, influenza virus, and 100 R whole-body X-radia-
tion on lung tumor incidence in mice. J. Natl. Cancer Inst. 17: 1129-11M.
Newbold, R.F., J. Amos and J.R. Connell. 1979. The cytotoxic, mutagenic and
clastogenic effects of chromium-containing compounds on mammalian cells in cul-
ture. Mutat. Res. 67: 55-63.
Newhouse, M.L. 1963. A cause of chromate dermatitis among assemblers in an
automobile factory. Brit. J. Ind. Med. 20: 199-203. (Cited in NIOSH, 1975.)
9-41
-------
NIOSH (National Institute for Occupational Safety and Health). 1973. Criteria
for a recommended standard: Occupational exposure to chromium (VI). DHEW
(Department of Health, Education and Welfare)-(NIOSH)-77-1'|J9. Rockville,
Maryland.
NIOSH (National Institute for Occupational Safety and Health). 1975. Criteria
for a recommended standard - Occupational exposure to chromium (VI). U.S.
Department of Health, Education, and Welfare, Washington, D.C.
NIOSH (National Institute for Occupational Safety and Health). 1977. NIOSH
Criteria for a Recommended Standard—Occupational Exposure to Chromium(VI).
DHEW (NIOSH) Pub. No. 76-129: 171-176.
NIOSH (National Institute for Occupational Safety and Health). 1977. NIOSH
Manual of Analytical Methods. Second edition. Vol. 3- U.S. Dept. of HEW,
Public Health Service, Center for Disease Control. NIOSH DHEW(NIOSH) Publica-
tion No. 77-157-C. Method S 317 and S 323.
NIOSH (National Institute for Occupational Safety and Health). 1979. Registry
of Toxic Effects of Chemical Substances (RTECS).
Nishioka, H. 1975. Mutagenic activities of metal compounds in bacteria. Mutat.
Res. 31: 185-189.
Nomiyama, K., H. Nomiyama and M. Yotoriyama. 1982. Low-molecular-weight
proteins in urine from rabbits given nephrotoxic compounds. Ind. Health. 20: 1-
10.
9-42
-------
Norseth et al. 1982. Biliary excretion of chromium in the rat: A role of
glutathione. Acta Pharmacol. Toxicol. 51: 450-455.
Ogawa, E. 1976. Experimental study on absorption, distribution and excretion of
trivalent and hexavalent chromes. Japanese J. Pharmacol. 26: 92.
Ohno, H., F. Hanaoka and M. Yanada. 1982. Inducibility of sister-chromatid
exchanges by heavy-metal ions. Mutation Research. 104: 141-145.
Ohsaki, Y., S. Abe, K. Kimura, Y. Tsunita, H. Mikami and M. Murao. 1978. Lung
cancer in Japanese chromate workers. Thorax. 33: 372-374.
Okubo, T. and K. Tsuchiya. 1979. Epidemiological study of chromium plater in
Japan. Biological Trace Element Research. 1: 35-44.
Okuno, I., J.A. Whitehead and R.E. White. 1978. Flameless atomic absorption
spectroscopic determination of heavy metals in whole-fish samples. J. Assoc.
Off. Anal. Chem. 61: 664-667.
Onkelinx, C. 1977. Compartment analysis of metabolism of chromium (III) in rats
of various ages. Amer. J. Physiol. 232: E478-484.
OSHA (Occupational Safety and Health Administration). 1978. Air contaminants.
29 CFR. 1910.1000. pp. 99-102.
9-43
-------
Ottinger, R.S., J.L. Blumenthal, D.F. Dal Porto, G.I. Gurber, M.J. Santy and C.C.
Shin. 1973. Recommended methods of reduction, neutralization, recovery, or
disposal of hazardous waste, vol. VI, National Disposal Site Candidate Waste
Stream Constituent Profile Reports - Mercury, Chromium, and Cadmium Compounds.
EPA, Cincinnati, Ohio, pp. 143-162.
Ovrutskii, G.D. 1976. Allergy to chrome from wearing steel dental prosthesis.
Stomatologia. 55(5): 60-62.
Pankow, J.F. and G.E. Janauer. 1974. Analysis for chromium in natural waters.
Part I. Preconcentration of chromate from PPB levels in aqueous solutions by ion
exchange. Anal. Chim. Acta. 69: 97-104.
Pankow, J.F., D.P. Leta, J.W. Lin, S.E. Ohl, W.P. Shum and G.E. Janauer. 1977.
Analysis for chromium in the aquatic ecosystem. II. A study of Cr(III) and
Cr(VI) in the Susquehanna River basin of New York and Pennsylvania. The Sci. of
Total Environ. 7: 17-26.
Parkhurst, H.J. 1925. Dermatosis industrialis in a blueprint worker due to
chromium compounds. Arch. Dermatol. Syphilol. 12: 253-256. (Cited in NIOSH,
1975.)
Partington, C.N. 1950. Acute poisoning with potassium bichromate. Brit. Med.
J. 2: 1097-1098. (Cited in NAS, 1974.)
Pascale, L.R., S.S. Waldstein, G. Engloring, A. Dubin and P.B. Szanto. 1952.
Chromium intoxication with special reference to hepatic injury. J. Amer. Med.
Assoc. 149: 1385-1389. (Cited in NIOSH, 1975.)
9-44
-------
Pattle, R.E. 1961. The Retention of Gases and Particles in the Human Nose. In:
Inhaled Particles and Vapours. C.N. Davies, Ed. Pergamon Press, Oxford, p.
302. (Cited in U.S. EPA, 1982).
Payne, W.W. 1960a. The role of roasted chromite ore in the production of
cancer. Arch. Environ. Health. 1: 20-26.
Payne, W.W. 1960b. Production of cancers in mice and rats by chromium com-
pounds. Arch. Ind. Health. 21: 530-535.
Pedersen, R.S. and P.T. Mersch. 1978. Chromic acid poisoning treated with acute
hemodialysis. Nephron. 22: 592-595.
Peltonen, L. and J. Fraki. 1983. Prevalence of dichromate sensitivity. Contact
Dermatitis. 9: 190-194.
Perisic, S. and D. Jovovic. 1977. Allergic contact sensitivity to chromates in
housewives. Dermatovenerol. Ingoslav. 4(1): 15-20.
Perone, V.B., A.E. Moffitt, P.A. Possick, M.M. Key, S.J. Danzinger and G.A.
Gellin. 1974. The chromium, cobalt, and nickel contents of American cement and
their relationship to cement dermatitis. Amer. Ind. Hyg. Assoc. J. 35: 301-306.
(Cited in NIOSH, 1975.)
Petrillji, F.L. and S. DeFlora. 1977. Toxicity and mutagenicity of hexavalent
chromium on Salmonella typhimurium. Appl. Environ. Microbiol. 33: 805-809.
9-45
-------
Petrilli, F. and S. DeFlora. 1978a. Oxidation of inactive trivalent chromium to
the mutagenic hexavalent form. Mutat. Res. 58: 167-178.
Petrilli, F.L. and S. DeFlora. 1978b. Metabolic deactivation of hexavalent
chromium mutagenicity. Mutat. Res. 51: 139-117.
Petrilli, F.L. and S. De Flora. 1982. Interpretations On Chromium Mutagenicity
and Carcinogenicity. In: Mutagens In Our Environment. A.R. Liss, Inc., New
York. p. 153-161.
Phung, H.T., L.J. Lund, A.L. Page and G.R. Bradford. 1979. J. Environ. Qual. 8:
171-175. (Cited in Cary, 1982.)
Pierce, J.O. and L.D. Scheel. 1965. Toxicity of alloys of chromium.
I. Solubility of chromium in solutions of water and protein. Arch. Environ.
Health. 10: 870-876. (Cited in NAS, 1971.)
Pik, A.J., J.M. Eckert and K.L. Williams. 1981. The determination of dissolved
chromium(III) and chromium(VI) and particulate chromium in waters at pg £
levels by thin-film Xray fluorescence spectrometry. Anal. Chira. Acta.
121: 351-356.
Ping, L., K. Matsumoto and K. Fuwa. 1983. Determination of urinary chromium
levels for healthy men and diabetic patients by electrothermal atomic absorption
spectrometry. Anal. Chim. Acta. 117: 205-212.
9-16
-------
Pirila V. and 0. Kilpio. 1949. On dermatoses caused by bichromates. Acta
Dermatol. Venereol. 29: 550-563. (Cited in NIOSH, 1975.)
Pokrovskaya, L.V. and N.K. Shabynina. 1973. On carcinogenic hazard in the
production of chromium ferroalloys. Gig. Tr. Prof. Zabol. 10:23-26.
Preston, A.M., R.P. Dowdy, M.A. Preston and J.N. Freeman. 1976. Effect of
dietary chromium on glucose tolerance and serum cholesterol in guinea pigs. J.
Nutr. 106(10): 1391-1397.
Princi, P., L.H. Miller, A. Davis and J. Cholak. 1962. Pulmonary disease of
ferroalloy workers. J. Occup. Med. 4: 301-310. (Cited in NAS, 1974.)
Public Health Service, Federal Security Agency. 1953. Health of workers in
chromate producing industry. Public Health Service Publication 192.
Quinby-Hunt, M.S. 1978. A survey of instrumentation used for monitoring metals
in water. American Lab. Dec. 1978. pp. 17-37.
Radian Corporation. 1983. Estimates of Population Exposure to Ambient Chromium
Emissions. EPA Contract No. 68-02-3818, Work Assignment No. 2, p. iv.
Radian Corporation. 1984. Study of Sources of Chromium Nickel and Manganese Air
Emissions. EPA Contract No. 68-02-3818, Work Assignment 34, p. 15.
Raffetto, G. 1977. Direct interaction with cellular targets as the mechanism for
chromium carcinogenesis. Tumori. 63: 503-512.
9-47
-------
Rambousek, J. 1913. Industrial Poisoning from Fumes, Gases and Poisons of
Manufacturing Processes. Translated and edited by T.M. Legge. E. Arnold,
London. 360 pp.
Rinaldi, G.M., T.R. Blackwood, D.L. Harris and K.M. Tackett. 1980. An evalua-
tion of emission factors for waste-to-energy systems, Interagency Energy/Envi-
ronment R and D Program Report, Industrial Environmental Research Laboratory,
Office of Research and Development. U.S. EPA, Cincinnati, Ohio, Report No.
EPA-600/7-80-135.
Robertson, D.E. and R. Carpenter. 1971*. Neutron Activation Techniques for
Measurement of Trace Metals in Environmental Samples. NAS-NS-3111*, U.S. Atomic
Energy Commission Technical Information Center, Oak Ridge, TN. 79 pp. (Cited in
Towill et al., 1978).
Robertson, F.N. 1975. No title provided. Ground Water. 13: 516-527. (Cited in
Cary, 1982.)
Roe, F.J.C. and R.L. Carter. 1969. Chromium carcinogensis: Calcium chromate as
a potent carcinogen for the subcutaneous tissues of the rat. Br. J. Cancer.
23: 172-176.
Roginski, E.E. and W. Mertz. 1969. Effects of chromium(III) supplementation on
glucose and amino acid metabolism in rats fed a low protein diet. J. Nutr.
97: 525-530.
9-18
-------
Rosensteel, F.E. and J.B. Lucas. 1975. Health Hazard Evaluation Determination,
Report No. 74-28-212, Westinghouse Air Brake Co. Cincinnati, Ohio.- U.S.
Department of Health, Education, and Welfare, Public Health Service, Center for
Disease Control, NIOSH.
Royle, H. 1975. Toxicity of chromic acid in the chromium plating industry (2).
Environ. Res. 10: 111-163.
Salbu, B., E. Steinnes and A.C. Pappas. 1975. Multielement neutron activation
analysis of fresh water using Ge(Li) gamma spectrometry. Anal. Chem.
47: 1011-1016.
Samitz, M.H. and S. Gross. 1961. Effects of hexavalent and trivalent chromium
compounds on the skin. Arch. Dermatol. 84: 104-409.
Samitz, M.H. and S. Katz. 1963. Preliminary studies on the reduction and
-binding of chromium with skin. Arch. Dermatol. 88: 816-819.
Samitz, M.H. and S. Katz. 1964. A study of the chemical reaction between
chromium and skin. J. Tnvestig. Dermatol. 43: 35-43.
Samitz, M.H. and J. Shrager. 1966. Patch test reactions to hexavalent and
trivalent chromium compounds. Arch. Dermatol. 94: 304-306.
Sanderson, C.J. 1976. The uptake and retention of chromium by cells. Trans-
plantation. 21: 526-529.
9-49
-------
Saner, G. 1980. Current Topics in Nutrition and Diseases, vol. 2. Chromium in
Nutrition and Disease. Alan B. Liss, Inc. New York. 135 pp.
Sano, T. and I. Mitohara. 1978. Occupational cancer among chromium workers.
Japanese J. Chest Disorders. 37: 90-101.
Sarto, P., I. Ceminato, V. Bianihi and A.G. Levis. 1982. Increased incidence of
chromosomal aberrations and sister chromatid exchanges in workers exposed to
chromic acid (Cr03) in electroplating factories. Carcinogenesis (London).
3: 1011-1016.
Satoh, H., Y. Fukuda, K. Terii, and N. Chika Katsuno. 1981. Epidemiologic study
of workers engaged in the manufacturing of chromium compounds. J. Occup. Med.
23(12)835-838.
Sayato, Y.t K. Nakamuro, S. Matoni and M. Ando. 1980. Metabolic fate of
•chromium compounds. I. Comparative behavior of chromium in rat administered
with Na251CrO,j and CrO_. J. Pharm. Dyn. 3: 17-23.
Schafer, H.A. 1977. Characteristics of municipal wastewater discharges, 1976
Southern California Coastal Water Research Project, El Segundo, California
(Cited in International Register of Potentially Toxic Chemicals (IRPTC), 1979,
United Nations Environmental Programme, Geneva, Switzerland). (Cited in Towill
et al., 1978).
Scheiner, D.M. and S.A. Katz. 1973. Chromium complexes and allergic eczematous
chromium dermatitis. Rev. Latinoam. Quim. (Mexico) 3CO: 160-172.
9-50
-------
Schneeberger, H.W. and G. Forck. 1971. Tierexperimentelle Chromallergie-
Einfluss der Valenzen und der Chemischen Umgebung der Chromionen (Experimental
chromium allergy of animals — The influence of valences and chemical properties
of chromium ions). Arch. Dermatol. Forsch. (West Germany) 219: 71-81. (Cited
in U.S. EPA, 1978.)
Schroeder, D.C. and G.F. Lee. 1975. Potential transformations of chromium in
natural waters. Water Air Soil Pollut. 1: 355-365.
Schroeder, H.A. and A.P. Nason. 1969. Trace metals in human hair. J. Investig.
Dermatol. 53: 71078. (Cited in U.S. EPA, 1978.)
Schroeder, H.A., J.J. Balassa and I.H. Tipton. 1962. Abnormal trace metals in
man - Chromium. J. Chron. Dis. 15: 911-964.
Schroeder, H.A., J.J. Balassa and W.H. Vinton, Jr. 1961. Chromium, lead,
•cadmium, nickel, and titanium in mice: Effect on mortality, tumors, and tissue
levels. J. Nutrit. 83: 239-250. (Cited by World Health Organization 1973.)
Schroeder, H.A., J.J. Balassa and W.H. Vinton, Jr. 1965. Chromium, cadium, and
lead in rats: Effects on lifespan, tumors, and tissue levels. J. Nutrit.
86: 51-66. (Cited by World Health Organization 1973.)
Schwartz, L., L. Tulipan and D.J. Birmingham. 1957. Occupational Diseases of the
Skin. 3rd ed. Lea and Febiger, Philadelphia. 981 pp. (Cited in NAS, 1971.)
9-51
-------
Schwartz-Speck, M. and H.P. Grundsmann. 1972. Experimental sensitization of
guinea pigs with trivalent and hexavalent chromium. In vivo and in vitro reac-
tions. Arch. Dermatol. Forsch. 242(3): 273-281.
Schwarz, K. and W. Mertz. 1959. Chromium(III) and the glucose tolerance factor.
Arch. Biochem. Biophys. 85: 292-295.
Sehmel, G.A. and W.H. Hodgson. 1976. Predicated dry deposition velocities.
Proc. of the symposium on atmosphere - surface exchange of particulate and
gaseous pollutants. ERDA Symposium Series 38, Richland, WA. pp. 399-122.
(Cited in Towill et al., 1978).
Shelley, W.B. 1964. Chromium in welding fumes as cause of eczematous hand
eruption. J. Amer. Med. Assoc. 189: 170-171. (Cited in NIOSH, 1975.)
Sherman, L., J.A. Glennon, W.J. Brech, G.H. Klomberg and E.S. Gordon. 1968.
Failure of trivalent chromium to improve hyperglycemia in diabetes mellitus.
Metabolism. 17: 439-442.
Shimkin, M.B. and J. Leiter. 1940. Induced pulmonary tumors in mice. III. The
role of chronic irritation in the production of pulmonary tumors in strain A
mice. J. Natl. Cancer Instit. 1: 211-254.
Shimkin, M.B., G.D. Stoner and J.C. Theiss. 1978. Lung tumor response in mice
to medals and metal salts. Adv. Exp. Med. Biol. 91: 85-91.
9-52
-------
Shuster, C.N., Jr. and B.J. Pringle. 1969. Trace metal accumulation by the
American oyster, Grasses trea virginioa. 1968 Proe. Natl. Shellfish Assoc.
59: 91.
Siegenthaler, U., A. Laine and L. Polak. 1983. Studies on contact sensitivity
to chromium in the guinea pig; the role of valence in the formation of the
antigenic determinant. J. Inv. Dermatology. 80: 44-47.
Silverstein, M., F. Mirer, D. Kotelchuck, B. Silverstein, and M. Bennett. 1981.
Mortality among workers in a die-casting and electroplating plant. Scand. J.
Work Environ. Health. 7(4): 156-165.
Sirova, M.A. and L.A. Loeb. 1976. Infidelity of DMA synthesis in vitro;
Screening for potential metal mutagens or carcinogens. Science.
: 1434-1436.
Sjogren, B., L. Hedstrom and U. Ulfvarson. 1983. Urine chromium as an estimator
of air exposure to stainless steel welding fumes. Int. Arch. Occup. Environ.
Health. 51: 347-354.
Slavin, W. 1981. Determination of chromium in the environment and in the work
place. Atmoic Spectroscopy. 2: 8-12.
Sluis-Cremer, O.K. and R.S.J. DuToit. 1968. Pneumoconiosis in chromite miners
in Soyth Africa. Brit. J. Ind. Med. 25: 63-67.
Smith, A.R. 1931. Chrome poisoning with manifestations of sensitization. J.
Amer. Med. Assoc. 97: 95-98.
9-53
-------
Smits, J. and R. Van Grieken. 1981. chelating 2,2'-diamino diethylamine cellu-
lose filters and X-ray fluorescence for preconcentration and trace analysis of
natural waters. Intern. J. Environ. Anal. Chem. 9: 81-91.
Smyth, H.F., C.P. Carpenter, C.S. Weil, U.C. Pozzani, J.A. Striegel and J.S.
Nycum. 1969. Range finding toxicity date: List VII. Amer. Ind. Hyg. Assoc.
J. 30: H70.
Somov, B.A., T.S. Bruevich and N.N. Bogomolets. 1976. Role of chromium com-
pounds in the development of allergic occupational dermatitis. Vestu. Dermatol.
Veneral. 10: 28-31.
Spruit, D. and K.E. Martin. 1975. Occupational cobalt and chromium dermatitis
in an offset printing factory. Dermatologica. 151(1): 34-42.
Spyrou, N.M., M.E. Pricker, R. Robertson and W.B. Gilboy. 1974. Realistic
'detection limits for neutron activation analysis of biological samples. In;
Comparative Studies of Food and Environmental Contamination, Proceedings of
Symposium on Nuclear Techniques in Comparative Studies of Food and Environmental
Contamination, Otaniemi, Finland. International Atomic Energy Agency, Vienna.
pp. 471-487. (Cited in Towill et al., 1978).
SRI International. 1982. 1982 Directory of Chemical Producers, United States of
American, Menlo Park, California.
Steffee, C.H. and A.M. Baetjer. 1965. Histopathological effects of chromate
chemicals. Report of studies in rabbits, guinea pigs, rats and mice. Arch.
Environ. Health. 11: 66.
9-54
-------
Steinhoff, S., C. Gud, O.K. Hatfield, and U. Mohr. 1983 (Sept.). Testing sodium
dichromate and soluble calcium chromate for carcinogenicity in rats. Bayer A.G.
Institute of Toxicology. Unpublished.
Stella, M., A. Mentaldi, R. Rossi, G. Rossi and A.G. Levis. 1982. Clastogenic
effects of chromium on human lymphocytes in vitro and in vivo. Mutat. Res.
101: 151-164.
Stephan, C.E. 1980. Memorandum to J. Stara. U.S. EPA. July 3.
Stettler, L.W., D.H. Groth and G.R. MacKay. 1977. Indentification of stainless
steel welding fume particulates in human lung and environmental samples using
elution probe microanalysis. Amer. Ind. Hyg. Assoc. J. 38(2): 76-82.
Stoewsand, G.S., J.R. Stamer, F.V. Kosikowski, R.A. Morse, C.A. Bache and D.J.
Lisk. 1979. Chromium and nickel in acidic foods and by-products contacting
stainless steel during processing. Bull. Environ. Contain. Toxicol.
21: 600-603.
Stokinger, H.E. and R.L. Woodward. 1958. Toxicologic methods for establishing
drinking water standards. J. Amer. Water Works Assoc. 50: 515-529.
Stoner, G.D., M.B. Shimkin, M.C. Trexell, T.L. Thompson and T.S. Terry. 1976.
Test for carcinogenicity of metallic compounds by the pulmonary tumor response in
strain A mice. Cancer Res. 36: 17^-1147.
9-55
-------
Sunderman, F.W., Jr., T.J. Lau and L.J. Cralley. 1971. Inhibitory effect of
manganese upon muscle tumorigenesis by nickel subsulfide. Cancer Res.
31: 92-95.
Swanson, S.A.V., M.A.B. Freeman and J.C. Heath. 1973. Lancet. 1: 561. (Cited
in Anonymous, 1976.)
Tamino, G., L. Peretta and A.G. Levis. 1981. Effects of trivalent and hexava-
lent chromium on the physicochemical properties of mammalian cell nucleic acids
and synthetic polynucleotides. Chem.-Biol. Interact. 37: 309-319.
Tanaka, S., M. Darzi and J.W. Winchester. 1981. Elemental analysis of soluble
and insoluble fractions of rain and surface waters by particle-induced X-ray
emission. Environ. Sci. Technol. 15: 351-357.
Tandon, S.K., J.B. Behari and D.N. Kachru. 1979. Distribution of chromium in
poisoned rats. Toxicology. 13: 29-37.
Tandon, S.K., A.K. Mathur and J.S. Gaur. 1977. Urinary excretion of chromium
and nickel among electroplaters and pigment industry workers. Int. Arch.
Occupat. Environ. Health. 10(1): 71-76.
Tandon, S.K., D.K. Saxena, J.S. Gaur and S.V. Chandra. 1978. Comparative
toxicity of trivalent and hexavalent chromium. Alterations in blood and liver.
Environ. Res. 15: 90-99.
9-56
-------
Taylor, F.H. 1966. The relationship and duration of employment as reflected by
a cohort of chromate workers. Amer. J. Public Health. 56: 218-229. (Cited in
NIOSH.)
Teraoka, H. 1981. Distribution of 21 elements in the internal organs of normal
males and the metallic workers in Japan. Archives of Environmental Health.
36: 155-164.
Thomas, B., J.A. Roughan and E.D. Watters. 1974. Cobalt, chromium and nickel
content of some vegetable foodstuffs. J. Sci. Food Agricul. (England).
25: 771-776.
Thompson, K.C. and K. Wagstaff. 1979. Determination of chromium in natural
waters and sewage effluents by atomic-absorption spectrophotometry using an air-
acetylene flame. Analyst. 101: 224-231.
i
Thomsen, E. and R.M. Stern. 1979. A simple analytical technique for the
determination of hexavalent chromium in welding fumes and other complex
matrices. Report No. 79.01. the Danish Welding institute, Copenhagen.
Tinsley, D.A., A.R. Baron, R. Critchley and R. J. Williamson. 1983. Extraction
Procedures for Atomic Absorption Spectrometric Analysis of Toxic Metals in Urban
Dust. Intern. J. Environ. Anal. Chem. 14: 285-298.
Tkeshelashvili, L.K., C.W. Shearman, R.A. Koplitz, R. Marlene and L.A. Loeb.
1980. Effects of arsenic, selenium, and chromium on the fidelity of DNA synthe-
sis. Cancer Res. 40: 2455-2460.
9-57
-------
Todd, G.F. 1962. Tobacco manufacturer's standing committee research papers. No.
1. Statistics of smoking in the United Kingdom. 3rd. ed. London: Tobacco
Research Council.
Toepfer, E.W., W. Mertz, E.E. Roginski and M.M. Polansky. 1973. Chromium in
foods in relation to biological activity. J. Agricul. Food Chem. 21: 69-73.
Tola, S., J. Kilpio, M. Virtamo and K. Haapa. 1977. Urinary chromium as an
indicator of the exposure of welders to chromium. Scand. J. Work Environ.
Health. 3: 192-202. (Cited in Mutti et al., 1979.)
Towill, L.E., C.R. Shriner, J.S. Drury, A.S. Hammons and J.W. Holleman. 1978.
Reviews of the environmental effects of pollutants: III. Chromium. Prepared for
Health Effects Research Laboratory, Office of Research and Development, U.S.
EPA, Cincinnati, Ohio. Report No. ORNL/EIS-80 and EPA-600/1-78-023.
-Toxic Materials News. February 1982. Toxic Materials News in Brief, p. 55.
Trama, F.B. and R.J. Benoit. 1960. Toxicity of hexavalent chromium to blue-
gills. J. Pollut. Cont. Fed. 32: 868.
Tsapakas, M.J., T.H. Hampton and K.W. Jennette. 1981. The carcinogen chromate
induced DNA-cross-links in rat and liver and kidney. J. Biol. Chem.
256: 3623-3626.
Tsuda, H. and K. Kato. 1977. Chromosomal aberrations and morphological trans-
formation in hamster embryonic cells treated with potassium dichromate. Mutat.
Res. H6: 87-9H.
9-58
-------
Tucks, N.A. 1964. The Mechanisms of Aerosols. The MacMillan Company, New York.
(Cited in U.S. EPA, 1982).
Turekian, K.K. and M.R. Scott. 1967. Concentration of Cr, Ag, Mo, Ni, Co, and Mn
in suspended material in streams. Environ. Sci. Technol. 1: 9UO-9M2.
Omeda, M. and M. Nishimura. 1979. Inducibility of chromosomal aberrations by
metal compounds in cultured mammalian cells. Mutat. Res. 67: 221-229.
Underwood, L.J. 1971. Chromium. In: Trace Elements in Human and Animal
Nutrition. 3rd ed. Academic Press, New York. pp. 253-266. (Cited in U.S. EPA,
1978.)
U.S. Department of Health and Human Services (HHS). 1982. The health conse-
quences of smoking-cancer. A report of the Surgeon General, U.S. Department of
Health and Human Services, Public Health Service, Office on Smoking and Health,
flockville, MD.
U.S. EPA. 1975. Region V joint federal/state survey of organics and inorganics
in selected drinking water supplies. Draft Report, Chicago, Illinois.
U.S. EPA. 1978. Reviews of the environmental effects of pollutants. III.
Chromium. Towill, L.E., C.R. Shriner, J.S. Drury, A.S. Hammons and J.W.
Holleman. Publication No. EPA-600/1-78-023. Prepared by the Information Divi-
sion, Oak Ridge National Laboratory.
U.S. EPA. 1979. Chemistry Laboratory Manual for Bottom Sediments and Elutriate
Testing. U.S. EPA, Chicago, IL. NTIS Order No. PB-294596, Springfield, VA.
9-59
-------
U.S. EPA. 1980a. Ambient water quality criteria for chromium. U.S. Environ-
mental Protection Agency, Washington, DC. NTIS No. PB 81-117467.
U.S. EPA. 1980b. Seafood consumption data analysis. Stanford Research Institute
International, Menlo Park, California. Final Report, Task 11. Contract No.
68-01-3997.
U.S. EPA. 1980c. Unpublished laboratory data. Environmental Research Labora-
tory, Narragansett, Rhode Island.
U.S. EPA. 1982. Air Quality Criteria for Particulate Matter and Sulfur Oxides.
U.S. EPA, Environmental Criteria and Assessment Office, Research Triangle Park,
NC. EPA-600/8-82-029a.
VanGrieken, R.E., C.M. Bresseleers and B.M. Vanderborght. 1977. Chelex-100 ion-
exchange filter membranes for preconcentration in X-ray fluorescence analysis of
water. Anal. Chem. 1*9: 1326-1331.
Veal, J.T., D.Y. Jeter, J.C. Hempel, R.C. Eckberg, W.E. Hatfield and D.J.
Hodgson. 1973. Structural and magnetic characterization of u-hydroxo-bist-
pentaammine chromium(III)] chloride monohydrate, the acid rhodo chloride
complex. Inorg. Chem. 12: 2928-2931.
Veillon, C., W.R. Wolf and B.E. Guthrie. 1979. Determination of chromium in
biological materials by stable isotope dilution. Analytical Chemistry. 51:
1022-1024.
9-60
-------
Veillon, C., K.Y. Patterson and N.A.Bryden. 1982a. Chromium in urine as
measured by atomic absorption spectrometry. Clin. Chem. 28: 2309-2311.
Veillon, C. K.Y. Patterson and N.A. Bryden. 1982b. Direct determination of
chromium in human urine by electrothermal atomic absorption spectrometry. Anal.
Chim. Acta. 136: 233-211.
Venediktova, K.P. and B.V. Gudina. 1976. Clinical-immunological characteris-
tics of allergic dermatitis and eczema in textile workers. Vestn. Dermatol.
Venerol. 10: 32-37.
Venier, P., A. Montaldi, F. Majone, V. Bianchi and A.G. Levis. 1982. Cytotoxic,
Mutagenic and Clastogenic Effects of Industrial Chromium Compounds.
Carcinogenesis. 3(11): 1331-1338.
Venitt, S. and L.S. Levy. 1971. Mutagenlcity of chromates in bacteria and its
relevance to chromate carcinogenesis. Nature. 250: 193-195. (Cited in Langard
and Norseth, 1977; CSD, 1979.)
Versieck, J., J. Hoste, F. Barbier, H. Steyaert, J. De Rudder and H. Michels.
1968. Determination of chromium and cobalt in human serum by neutron activation
analysis. Clin. Chem. 21: 303-308.
Vigliani, B.C. and N. Zurlo. 1955. Erfahrungen der Clinica del Lavoro mit
einigen maximalen Arbeitsplatzkonzentrationen (MAK) von Tndustriegiften. Arch.
Gewerbepath. Gewerbehyg. 13: 528-531. (Cited in NAS, 1971.)
9-61
-------
Visek, W.J., I.E. Whitney, U.S.G. Kuhn, III and C.L. Comer. 1953. Metabolism of
Cr-51 by animals as influenced by chemical state. Proc. Soc. Exp. Biol. Med.
84: 610-615. (Cited in Hertz, 1969; EPA, 1978.)
Wada, 0., S. Manabe, N. Yamaguchi, S. Ishikoawa and H. Yanagisawa. 1983. Low-
molecular-weight, chromium-binding substance in rat lungs and its possible role
in chromium movement. Tndust. Health. 21: 35-41.
Wahlberg, J.E. 1965. Percutaneous absorption of trivalent and hexavalent chro-
mium. Arch. Dematol. 92: 315-318.
Wahlberg, J.E. and G. Wennersten. 1977. Light sensitivity and chromium dermati-
tis. Brit. J. Dermatol. 97(4): 411-416.
Walsh, E.N. 1953. Chromate hazards in industry. J. Amer. Med. Assoc.
153: 1305-1308. (Cited in NIOSH, 1975.)
Watanabe, S. and Y. Fukuchi. 1975. An epidemiological survey on lung cancer in
workers of a chromate-producing industry in Hokkaido, Japan. Presented at Inter-
national Congress on Occupational Health.
Warren, G., P. Schultz, D. Bancroft, K. Bennett, E.H. Abbott and S. Rogers.
1981. Mutagenicity of a series of hexacoordinate Chromium(III) compounds.
Mutat. Res. 90: 111-118.
Watson, G. 1977. Age incidence curves for cancer. Proc. Natl. Acad. Sci.
74: 1341-1342.
9-62
-------
Weast, B.C., Ed. 1980. Handbook of Chemistry and Physics. 61st edition. CRC
Press, Cleveland, Ohio.
Weisenberger, R. 1976. Cutting oils and coolants are chief culprits in worker
skin problems. Int. J. Occup. Health Safety. 1976: 16-21.
White, P.P. 1934. The Dermatergoses or Occupational Affections of the Skin.
4th ed. London, H.K. Lewis and Co., Ltd. 716 pp. (Cited in NAS, 1974.)
Whitnack, G.C. 1975. Single-sweep polarographic technique use in micropollu-
tion studies of ground and surface waters. Anal. Chem. 47: 618-621.
Whittemore, A. 1978. Quantitative theories of carcinogenesis. In: Advance in
cancer research. New York, NY: Academic Press, Inc. pp. 55-58.
Whittemore, A. and B. Keller. 1978. Quantitative theory of carcinogenesis.
Society for Industrial and Applied Mathematics Review. 20: 1-30.
Wild, D. 1978. Cytogenetic effects in the mouse of 17 chemical mutagens and
carcinogens evaluated by the micronucleus test. Mutat. Res. 56: 319-327.
Windholz, M., Ed. 1976. The Merck Index. Ninth edition. Merck and Co., Inc.,
Rahway, NJ. p. 287.
Winchester, J.W., L. Weixiu, R. Lixin, W. Mingxing and W. Maenhaut. 1981. Fine
and coarse aerosol composition from a rural area in North China. Atmos. Environ.
15: 933-937.
9-63
-------
Winston, J.R. and E.N. Walsh. 1951. Chromate dermatitis in railroad employees
working with diesel locomotives. J. Amer. Med. Assoc. 147: 1133-1131!. (Cited
in NIOSH, 1975.)
Withey, J. 1983. Memorandum addressed to Si Duk Lee, Bnvironmental Criteria and
Assessment Office, U.S. Environmental Protection Agency, Research Triangle Park,
North Carolina, dated March H, 1983.
Witmer, V., K. Cooper and J. Kelly. 1982. Effects of plating efficiency and
lowered concentration of salts on mutagenicity assays with Ames Salmonella
strains. Adv. Exp. Med. Biol. 136B: 1271-1281*.
Wolf, W.R. 1976. Coupled gas chromatography-atomic absorption spectrometry for
the nanogram determination of chromium. Anal. Chem. M8: 1717-1720.
Wolf, W.R., M.L. Taylor, B.M. Hughes, T.O. Tiernan and R.E. Sievers. 1972.
Determination of chromium and beryllium at the program level by gas. Chromato-
graphy-mass Spectrometry. Anal. Chem. M: 616-618.
Yamaguchi, S., K. Sano, and N. Shimojo. 1983. On the biological half-time of
hexavalent chromium in rats. Ind. Health. 21: 25-3^.
Yamamoto et al. 1982. Toxicol. Appl. Pharmacol. 59: 513-523.
Yen, H..C., R.F. Phalen and O.G. Raabe. 1976. Factors influencing the deposition
of inhaled particles. Environ. Health Perspect. 15: 117. (Cited in U.S. EPA,
1982).
9-64
-------
Ziet and Mitchell. 1976. Contamination Control in Trace Element Analysis. John
Wiley and Sons, Inc., New York.
Zoller, W.H., E.S. Gladney and R.A. Duce. 1971. Atmospheric concentrations and
sources of trace metals at the South Pole. Science. 183; 198-200.
Zook>f E.G., J.J. Powell, B.M. Hackley, J.A. Emerson, J.F. Brooker and G.M. Knobl.
1976. National Marine Fisheries Service preliminary survey of selected seafoods
for mercury, lead, cadmium, chromium, and arsenic content. J. Agricul. Food
Chem. 24: 47-53.
Zvaifler, N.I. 1944. Chromic acid poisoning resulting from inhalation of mist
developed from five per cent chromic acid solution. I. Medical aspects of chromic
acid poisoning. J. Ind. Hyg. Toxicol. 26: 124-126. (Cited in NAS, 1974.)
9-65
-------
-------
r' a
.H O
*•' -- o
O ',0 «
O ,-(
o
O, 'j.1
IX j o)
— I' -sjJ
+' O
-^ CQ .--.)
: o
"
so
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Please make all necessary changes on the above label,
detach or copy and return to the address in the upper
left-hand corner
If you oo not wish to receive these reports CHECK HERE a,
detach, or copy this cover, and return to the address in the
upper left-hand corner
------- |