United States
Environmental Protection
Agency
EPA-452/R-97-007
December 1997
Air
Mercury Study
Report to Congress
Volume V:
Health Effects of Mercury
and Mercury Compounds
Office of Air Quality Planning & Standards
and
Office of Research and Development
-------
MERCURY STUDY REPORT TO CONGRESS
VOLUME V:
HEALTH EFFECTS OF MERCURY AND MERCURY COMPOUNDS
December 1997
Office of Air Quality Planning and Standards
and
Office of Research and Development
U.S. Environmental Protection Agency
-------
TABLE OF CONTENTS
Page
U.S. EPA AUTHORS iv
SCIENTIFIC PEER REVIEWERS v
WORK GROUP AND U.S. EPA/ORD REVIEWERS viii
LIST OF TABLES ix
LIST OF FIGURES xii
LIST OF SYMBOLS, UNITS AND ACRONYMS xiii
EXECUTIVE SUMMARY ES-1
1. INTRODUCTION 1-1
2. TOXICOKINETICS 2-1
2.1 Absorption 2-1
2.1.1 Elemental Mercury 2-1
2.1.2 Inorganic Mercury 2-2
2.1.3 Methylmercury 2-3
2.2 Distribution 2-4
2.2.1 Elemental Mercury 2-4
2.2.2 Inorganic Mercury 2-4
2.2.3 Methylmercury 2-5
2.3 Metabolism 2-6
2.3.1 Elemental Mercury 2-6
2.3.2 Inorganic Mercury 2-6
2.3.3 Methylmercury 2-6
2.4 Excretion 2-7
2.4.1 Elemental Mercury 2-7
2.4.2 Inorganic Mercury 2-8
2.4.3 Methylmercury 2-8
2.5 Biological Monitoring 2-9
2.5.1 Elemental Mercury 2-9
2.5.2 Inorganic Mercury 2-11
2.5.3 Methylmercury 2-11
2.5.4 Methods of Analysis for Measuring Mercury in Biological Samples 2-12
2.6 Studies on Pharmacokinetic Models 2-12
2.6.1 Introduction 2-12
2.6.2 Inorganic mercury 2-13
2.6.3 Methylmercury 2-13
2.6.4 Discussion 2-15
3. BIOLOGICAL EFFECTS 3-1
3.1 Elemental Mercury 3-1
3.1.1 Critical Noncancer Data 3-1
3.1.2 Cancer Data 3-5
3.1.3 Other Data 3-9
-------
TABLE OF CONTENTS (continued)
Page
3.2 Inorganic Mercury 3-35
3.2.1 Critical Noncancer Data 3-35
3.2.2 Cancer Data 3-39
3.2.3 Other Data 3-43
3.3 Methylmercury 3-59
3.3.1 Critical Noncancer Data 3-59
3.3.2 Cancer Data 3-66
3.3.3 Other Data 3-73
4. SUSCEPTIBLE POPULATIONS 4-1
5. INTERACTIONS 5-1
6. HAZARD IDENTIFICATION AND DOSE-RESPONSE ASSESSMENT 6-1
6.1 Background 6-1
6.1.1 Hazard Identification 6-1
6.1.2 Dose-response Assessment 6-4
6.2 Hazard Identification for Mercury 6-7
6.2.1 Developmental Effects 6-8
6.2.2 Germ Cell Mutagenicity 6-11
6.2.3 Carcinogenic Effects 6-13
6.3 Dose-Response Assessment For Mercury 6-16
6.3.1 Systemic Noncancer Effects 6-16
6.3.2 Developmental Effects 6-54
6.3.3 Germ Cell Mutagenicity 6-55
6.3.4 Carcinogenic Effects 6-55
6.4 Risk Assessments Done By Other Groups 6-56
6.4.1 Food and Drug Administration 6-56
6.4.2 ATSDR 6-57
6.4.3 Department of Energy 6-57
6.4.4 National Institute of Environmental Health Sciences (NIEHS) 6-58
6.4.5 Department of Labor 6-58
6.4.6 Various States 6-58
6.4.7 World Health Organization 6-58
6.4.8 ACGIH 6-59
7. ONGOING RESEARCH AND RESEARCH NEEDS 7-1
7.1 Ongoing Research 7-1
7.2 Research Needs 7-3
8. REFERENCES 8-1
-------
TABLE OF CONTENTS (continued)
Page
APPENDIX A DOSE CONVERSIONS A-l
APPENDIX B SUMMARIES FOR THE INTEGRATED RISK
INFORMATION SYSTEM (IRIS) B-l
APPENDIX C ATTENDEES OF U.S. EPA PEER REVIEW WORKSHOP
ON MERCURY ISSUES C-l
APPENDIX D UNCERTAINTY ANALYSIS OF THE METHYLMERCURY RfD D-l
in
-------
U.S. EPA AUTHORS
Principal Authors:
Beth Hassett-Sipple Rita Schoeny, Ph.D.
Office of Air Quality Office of Water
Planning and Standards Washington, DC
Research Triangle Park, NC
Jeff Swartout
National Center for Environmental
Assessment-Cincinnati
Office of Research and Development
Cincinnati, OH
Contributing Authors:
Kathryn R. Mahaffey, Ph.D.
National Center for Environmental
Assessment-Washington
Office of Research and Development
Washington, DC
Glenn E. Rice
National Center for Environmental
Assessment-Cincinnati
Office of Research and Development
Cincinnati, OH
IV
-------
SCIENTIFIC PEER REVIEWERS
Dr. William J. Adams*
Kennecott Utah Corporation
Dr. Brian J. Alice
Harza Northwest, Incorporated
Dr. Thomas D. Atkeson
Florida Department of Environmental
Protection
Dr. Donald G. Barnes*
U.S. EPA Science Advisory Board
Dr. Steven M. Bartell
SENES Oak Ridge, Inc.
Dr. David Bellinger*
Children's Hospital, Boston
Dr. Nicolas Bloom*
Frontier Geosciences, Inc.
Dr. Mike Bolger
U.S. Food and Drug Administration
Dr. Peter Botros
U.S. Department of Energy
Federal Energy Technology Center
Thomas D. Brown
U.S. Department of Energy
Federal Energy Technology Center
Dr. Dallas Burtraw*
Resources for the Future
Dr. Thomas Burbacher*
University of Washington
Seattle
U.S. Department of Energy
Policy Office, Washington D.C.
Dr. Rick Canady
Agency for Toxic Substances and Disease
Registry
Dr. Rufus Chaney
U.S. Department of Agriculture
Dr. Joan Daisey*
Lawrence Berkeley National Laboratory
Dr. John A. Dellinger*
Medical College of Wisconsin
Dr. Kim N. Dietrich*
University of Cincinnati
Dr. Tim Eder
Great Lakes Natural Resource Center
National Wildlife Federation for the
States of Michigan and Ohio
Dr. Lawrence J. Fischer*
Michigan State University
Dr. William F. Fitzgerald
University of Connecticut
Avery Point
A. Robert Flaak*
U.S. EPA Science Advisory Board
Dr. Bruce A. Fowler*
University of Maryland at Baltimore
Dr. Katherine Flegal
National Center for Health Statistics
Dr. James P. Butler
University of Chicago
Argonne National Laboratory
Elizabeth Campbell
Dr. Steven G. Gilbert*
Biosupport, Inc.
Dr. Cynthia C. Gilmour*
The Academy of Natural Sciences
-------
SCIENTIFIC PEER REVIEWERS (continued)
Dr. Robert Goyer
National Institute of Environmental Health
Sciences
Dr. George Gray
Harvard School of Public Health
Dr. Terry Haines
National Biological Service
Dr. Gary Heinz*
Patuxent Wildlife Research Center
Joann L. Held
New Jersey Department of Environmental
Protection & Energy
Dr. Robert E. Hueter*
Mote Marine Laboratory
Dr. Harold E. B. Humphrey*
Michigan Department of Community Health
Dr. James P. Hurley*
University of Wisconsin
Madison
Dr. Joseph L. Jacobson*
Wayne State University
Dr. Gerald J. Keeler
University of Michigan
Ann Arbor
Dr. Ronald J. Kendall*
Clemson University
Dr. Lynda P. Knobeloch*
Wisconsin Division of Health
Dr. Genevieve M. Matanoski*
The Johns Hopkins University
Dr. Thomas McKone*
University of California
Berkeley
Dr. Malcolm Meaburn
National Oceanic and Atmospheric
Administration
U.S. Department of Commerce
Dr. Michael W. Meyer*
Wisconsin Department of Natural Resources
Dr. Maria Morandi*
University of Texas Science Center at Houston
Dr. Paul Mushak
PB Associates
Harvey Ness
U.S. Department of Energy
Federal Energy Technology Center
Dr. Christopher Newland*
Auburn University
Dr. Jerome O. Nriagu*
The University of Michigan
Ann Arbor
William O'Dowd
U.S. Department of Energy
Federal Energy Technology Center
Dr. W. Steven Otwell*
University of Florida
Gainesville
Dr. Leonard Levin
Electric Power Research Institute
Dr. Steven E. Lindberg*
Oak Ridge National Laboratory
Dr. Jozef M. Pacyna
Norwegian Institute for Air Research
Dr. Ruth Patterson
Cancer Prevention Research Program
Fred Gutchinson Cancer Research Center
VI
-------
SCIENTIFIC PEER REVIEWERS (continued)
Dr. Donald Porcella
Electric Power Research Institute
Dr. Deborah C. Rice*
Toxicology Research Center
Samuel R. Rondberg*
U.S. EPA Science Advisory Board
Charles Schmidt
U.S. Department of Energy
Dr. Pamela Shubat
Minnesota Department of Health
Dr. Ellen K. Silbergeld*
University of Maryland
Baltimore
Dr. Alan H. Stern
New Jersey Department of Environmental
Protection & Energy
Dr. David G. Strimaitis*
Earth Tech
Dr. Edward B. Swain
Minnesota Pollution Control Agency
Dr. Valerie Thomas*
Princeton University
Dr. M. Anthony Verity
University of California
Los Angeles
Dr. Howard A. Simonin*
NYSDEC Aquatic Toxicant Research Unit
Dennis Smith
U.S. Department of Energy
Federal Energy Technology Center
Dr. Ann Spacie*
Purdue University
*With EPA's Science Advisory Board, Mercury Review Subcommitte
vn
-------
WORK GROUP AND U.S. EPA/ORD REVIEWERS
Core Work Group Reviewers:
DanAxelrad, U.S. EPA
Office of Policy, Planning and Evaluation
Angela Bandemehr, U.S. EPA
Region 5
Jim Darr, U.S. EPA
Office of Pollution Prevention and Toxic
Substances
Thomas Gentile, State of New York
Department of Environmental Conservation
Arnie Kuzmack, U.S. EPA
Office of Water
David Layland, U.S. EPA
Office of Solid Waste and Emergency Response
Karen Levy, U.S. EPA
Office of Policy Analysis and Review
Steve Levy, U.S. EPA
Office of Solid Waste and Emergency Response
Lorraine Randecker, U.S. EPA
Office of Pollution Prevention and Toxic
Substances
Joy Taylor, State of Michigan
Department of Natural Resources
U.S. EPA/ORD Reviewers:
Robert Beliles, Ph.D., D.A.B.T.
National Center for Environmental Assessment
Washington, DC
Eletha Brady-Roberts
National Center for Environmental Assessment
Cincinnati, OH
Annie M. Jarabek
National Center for Environmental Assessment
Research Triangle Park, NC
Matthew Lorber
National Center for Environmental Assessment
Washington, DC
Susan Braen Norton
National Center for Environmental Assessment
Washington, DC
Terry Harvey, D.V.M.
National Center for Environmental Assessment
Cincinnati, OH
Vlll
-------
LIST OF TABLES
Page
ES-1 Summary of U.S. EPA Hazard Identification/Dose-response
Assessment for Mercury and Mercury Compounds ES-6
2-1 Reference Values for Total Mercury Concentrations in Biological Media for the General
Population 2-10
2-2 Analytical Methods for the Detection of Mercury in Biological Samples 2-12
3-1 Carcinogenic Effects of Elemental Mercury in Humans: Epidemiological Studies .... 3-7
3-2 Lethality of Elemental Mercury in Humans: Case Studies 3-9
3-3 Lethality of Elemental Mercury in Animals: Inhalation Exposure 3-10
3-4 Neurotoxicity of Elemental Mercury in Humans: Case Studies 3-11
3-5 Neurotoxicity of Elemental Mercury in Humans: Epidemiological Studies 3-13
3-6 Neurotoxicity of Elemental Mercury in Animals: Inhalation Exposure 3-15
3-7 Renal Toxicity of Elemental Mercury in Humans: Case Studies 3-16
3-8 Renal Toxicity of Elemental Mercury in Humans: Epidemiological Studies 3-17
3-9 Renal Toxicity of Elemental Mercury in Animals: Inhalation Exposure 3-18
3-10 Respiratory Toxicity of Elemental Mercury in Humans: Case Studies 3-18
3-11 Respiratory Toxicity of Elemental Mercury in Animals: Inhalation Exposure 3-20
3-12 Cardiovascular Toxicity of Elemental Mercury in Humans: Case Studies 3-21
3-13 Cardiovascular Toxicity of Elemental Mercury in Humans: Epidemiological
Studies 3-21
3-14 Cardiovascular Toxicity of Elemental Mercury in Animals: Inhalation
Exposure 3-22
3-15 Gastrointestinal Toxicity of Elemental Mercury in Humans: Case Studies 3-23
3-16 Gastrointestinal Toxicity of Elemental Mercury in Animals: Inhalation
Exposure 3-24
3-17 Hepatic Toxicity of Elemental Mercury in Humans: Case Study 3-24
3-18 Hepatic Toxicity of Elemental Mercury in Animals: Inhalation Exposure 3-25
3-19 Hematological Toxicity of Elemental Mercury in Humans: Case Studies 3-25
3-20 Hematological Toxicity of Elemental Mercury in Humans: Epidemiological
Studies 3-26
3-21 Immunotoxicity of Elemental Mercury in Humans: Case Study 3-27
3-22 Immunotoxicity of Elemental Mercury in Humans: Epidemiological Studies 3-27
3-23 Immunotoxicity of Elemental Mercury in Animals: Inhalation Exposure 3-27
3-24 Dermal Toxicity of Elemental Mercury in Humans: Case Studies 3-28
3-25 Developmental Toxicity of Elemental Mercury in Humans: Case
Studies 3-29
3-26 Developmental Toxicity of Elemental Mercury in Humans: Epidemiological
Studies 3-30
3-27 Developmental Toxicity of Elemental Mercury in Animals 3-31
3-28 Reproductive Toxicity of Elemental Mercury in Humans: Epidemiological
Studies 3-32
3-29 Reproductive Toxicity of Elemental Mercury in Animals 3-33
3-30 Genotoxicity of Elemental Mercury in Humans 3-34
3-31 Incidence of Selected Lesions in Rats in the NTP (1993) 2-Year Gavage Study 3-41
3-32 Incidence of Renal Tubule Tumors in Male Mice in the NTP (1993)
-------
LIST OF TABLES (continued)
2-Year Gavage Study 3-42
3-33 Carcinogenic Effects of Inorganic Mercury in Animals: Oral Exposure 3-43
3-34 Lethality of Inorganic Mercury in Humans: Case Study 3-43
3-35 Lethality of Inorganic Mercury in Animals: Oral Exposure 3-44
3-36 Neurotoxicity of Inorganic Mercury in Humans: Case Studies 3-45
3-37 Neurotoxicity of Inorganic Mercury in Animals: Oral Exposure 3-45
3-38 Renal Toxicity of Inorganic Mercury in Humans: Case Studies 3-46
3-39 Renal Toxicity of Inorganic Mercury in Animals: Oral Exposure 3-46
3-40 Renal Toxicity of Inorganic Mercury in Animals: Inhalation Exposure 3-48
3-41 Cardiovascular Toxicity of Inorganic Mercury in Animals 3-48
3-42 Gastrointestinal Toxicity of Inorganic Mercury in Humans: Case Studies 3-49
3-43 Gastrointestinal Toxicity of Inorganic Mercury in Animals 3-50
3-44 Hepatic Toxicity of Inorganic Mercury in Animals 3-51
3-45 Immunotoxicity of Inorganic Mercury in Animals 3-52
3-46 Developmental Toxicity of Inorganic Mercury in Animals: Inhalation
Exposure 3-52
3-47 Developmental Toxicity of Inorganic Mercury in Animals: Oral Exposure 3-54
3-48 Reproductive Toxicity of Inorganic Mercury in Humans: Case Study 3-56
3-49 Reproductive Toxicity of Inorganic Mercury in Animals 3-57
3-50 Genotoxicity of Inorganic Mercury in Humans 3-57
3-51 Genotoxicity of Inorganic Mercury in Animals 3-59
3-52 Carcinogenic Effects of Methylmercury in Humans: Epidemiological Studies 3-68
3-53 Carcinogenic Effects of Methylmercury in Animals: Oral Exposure 3-70
3-54 Lethality of Methylmercury in Humans: Case Study of Oral Exposure 3-73
3-55 Lethality of Methylmercury in Humans: Case Studies of Inhalation Exposure 3-74
3-56 Lethality of Methylmercury in Animals 3-74
3-57 Neurotoxicity of Methylmercury in Humans: Case Studies of Oral Exposure 3-77
3-58 Neurotoxicity of Methylmercury in Humans: Case Studies of Inhalation
Exposure 3-78
3-59 Neurotoxicity of Methylmercury in Animals 3-80
3-60 Renal Toxicity of Methylmercury in Animals 3-82
3-61 Cardiovascular Toxicity of Methylmercury in Humans: Case Study 3-82
3-62 Cardiovascular Toxicity of Methylmercury in Animals 3-82
3-63 Gastrointestinal Toxicity of Methylmercury in Animals 3-84
3-64 Immunotoxicity of Methylmercury in Animals 3-84
3-65 Dermal Toxicity of Methylmercury in Humans: Epidemiological Study 3-85
3-66 Developmental Toxicity of Methylmercury in Humans: Case Studies 3-86
3-67 Developmental Toxicity of Methylmercury in Humans: Epidemiologic Studies 3-87
3-68 Developmental Toxicity of Methylmercury in Animals 3-88
3-69 Reproductive Toxicity of Methylmercury in Animals 3-94
3-70 Genotoxicity of Methylmercury in Humans: Case Study 3-95
3-71 Genotoxicity of Methylmercury in Humans: Epidemiology Study 3-95
3-72 Genotoxicity of Methylmercury in Cats 3-96
5-1 Interactions of Mercury with Other Compounds 5-3
6-1 Consensus Decisions of Peer Review Panel 6-18
6-2 Available Data on HairBlood Ration (Total Hg) 6-24
6-3 Incidence of Effects in Iraqi Children by Exposure Group 6-26
-------
LIST OF TABLES (continued)
6-4 Incidence of Effects in Iraqi Adults by Exposure Group 6-26
6-5 Methylmercury Benchmark Dose Estimates (ppm hair) 6-28
6-6 Methylmercury Benchmark Dose Estimates (ppb blood) 6-28
6-7 Density-Based Dose Groupings 6-29
6-8 Uniform Dose Groupings 6-30
6-9 Benchmark Dosed Calculated on Data from Marsh et al. (1987) 6-30
6-10 Summary of Benchmark Doses (BMD) Estimated for Methylmercury 6-32
6-11 Results of Revised Denver Developmental Screening Test (DDST) Administered
to Seychellois Children in Cross-sectional Study 6-33
6-12 Estimates of No Observed Adverse Effect Levels (NOAELs) and Lowest Observed
Adverse Effect Levels (LOAELs) from Human Studies 6-34
6-13 Estimates of NOAELs and LOAELs from Animal Studies 6-39
6-14 NOAELs and LOAELs for Developmental Toxicity of Methylmercury in Animals .... 6-42
6-15 Studies in Non-Human Primates 6-46
7-1 Ongoing Research 7-1
XI
-------
LIST OF FIGURES
Page
6-1 Density of Data Points Relative to Hg Concentration in Hair
for Iraqi Cohort Data 6-29
xn
-------
LIST OF SYMBOLS, UNITS AND ACRONYMS
ATSDR
BML
bw
CAA
CHO
C.I.
CNS
CRAVE
DDST
DHHS
DNA
DWEL
ECG
EEG
EPA
FDA
GABA
Gd
HEC
Hg
Hg-U
IgG
IRIS
LC50
LD50
LOAEL
MF
MMAD
MMC
MMH
MRL
MTD
NAG
NADH
NADPH
NOAEL
NS
NTP
PMA
ppd
RfD
RfDDT
RfC
SCE
SGPT
SH
Agency for Toxic Substances and Disease Registry
Biological monitoring level
Body weight
Clean Air Act as amended in 1990
Chinese hamster ovary
Confidence interval
Central nervous system
Carcinogen Risk Assessment Verification Endeavor
Denver Developmental Screen Test
Department of Health and Human Services
Deoxyribonucleic acid
Drinking water equivalent level
Electrocardiogram
Electroencephalogram
Environmental Protection Agency
Food and Drug Administration
Gamma aminobutyric acid
Gestation day
Human equivalent concentration
Mercury
Urinary mercury
Immunoglobulin G
Integrated Risk Information System
Lethal concentration killing 50 percent of the animals tested (inhalation)
Lethal dose killing 50 percent of the animals tested
Lowest-observed-adverse-effect level
Modifying factor
Mass median aerodynamic diameter
Methylmercuric chloride
Methylmercuric hydroxide
Minimal risk level
Maximum tolerated dose
N-acetyl-b-glycosaminidase
Reduced nicotinamide adenine dinucleotide
Reduced nicotinamide adenine dinucleotide phosphate
No-observed-adverse-effect level
Not specified
National Toxicology Program
Phenyl mercuric acetate
Postpartum day
Reference dose (oral)
Reference dose for developmental toxicity
Reference concentration (inhalation)
Sister chromatid exchange
Serum glutamic-pyruvic transaminase
Sulfhydryl groups
Xlll
-------
LIST OF SYMBOLS, UNITS AND ACRONYMS (continued)
SMR Standard mortality ratio
TOLD Test of Language Development
TWA Time-weighted average
UF Uncertainty factor
UFA Uncertainty factor for interspecies extrapolation
UFH Uncertainty factor for intraspecies extrapolation (animal to human)
UFL Uncertainty factor for use of a LOAEL
UFS Uncertainty factor for use of a subchronic-duration study
WHO World Health Organization
xiv
-------
EXECUTIVE SUMMARY
Section 112(n)(l)(B) of the Clean Air Act (CAA), as amended in 1990, directs the U.S.
Environmental Protection Agency (U.S. EPA) to submit to Congress a comprehensive study on
atmospheric emissions of mercury. This document, which covers the human health effects of mercury
and mercury compounds, is one volume of U.S. EPA's eight-volume Report in response to this directive.
Mercury is a naturally occurring element that is found in air, water and soil. It exists in any of
three valence states: Hg° (elemental mercury), Hg22+ (mercurous mercury), or Hg2+ (mercuric mercury).
Most of the population of the earth have some exposure to mercury as a result of normal daily activities.
The general population may be exposed to mercury through inhalation of ambient air; consumption of
contaminated food, water, or soil; and/or dermal exposure to substances containing mercury. In addition,
some quantity of mercury is released from dental amalgam.
The health effects literature contains many investigations of populations with potentially high
exposure to mercury, including industrial workers, people living near point sources of mercury
emissions, people who consume large amounts offish, and dental professionals. There also are
numerous studies of populations unintentionally exposed to high levels of mercury, such as the Minamata
poisoning episode in Japan. Volume IV (An Assessment Exposure to Mercury in the United States)
presents measured and predicted mercury exposure for various U.S. populations.
The purpose of this volume, Volume V, is to summarize the available health effects information
for mercury and mercury compounds and to present U.S. EPA's analysis for two critical pieces of the risk
assessment paradigm described by the National Academy of Sciences in 1983. Specifically, this volume
contains the hazard identification and dose-response assessments for three forms of mercury: elemental
mercury, mercuric chloride (inorganic mercury),and methylmercury (organic mercury). In order to
characterize risk for any populations, the evaluations presented in this volume must be combined with the
assessment of exposure presented in Volume IV.
Volume V is not intended to be an exhaustive survey of the voluminous health effects literature
available for mercury. Rather, the purpose is to present a brief survey of the studies relevant for
assessing potential human health effects and to present more detailed information on those studies which
form the basis for U.S. EPA's hazard identification and dose-response assessments. The three forms of
mercury which are emphasized in this volume were selected based on data indicating that these are the
predominant forms of mercury to which humans are exposed. In addition, examination of the published
literature indicates that most health data are on these forms. It is acknowledged that certain populations
can be exposed to many types of organic mercurials, such as antiseptics and pesticides. Volume V,
however, deals with methylmercury except in cases where information on another organic is presented
for illustrative purposes.
Toxicokinetics
The toxicokinetics (i.e., absorption, distribution, metabolism, and excretion) of mercury is highly
dependent on the form of mercury to which a receptor has been exposed. Below is a brief summary of
the toxicokinetics information for elemental mercury, mercuric chloride, and methylmercury. Chapter 2
contains a more complete summary of the toxicokinetics information available for mercury.
ES-1
-------
Elemental Mercury
The absorption of elemental mercury vapor occurs rapidly through the lungs, but it is poorly
absorbed from the gastrointestinal tract. Once absorbed, elemental mercury is readily distributed
throughout the body; it crosses both placental and blood-brain barriers. Elemental mercury is oxidized to
inorganic divalent mercury by the hydrogen peroxidase-catalase pathway, which is present in most
tissues. The distribution of absorbed elemental mercury is limited primarily by the oxidation of
elemental mercury to the mercuric ion as the mercuric ion has a limited ability to cross the placental and
blood-brain barriers. Once elemental mercury crosses these barriers and is oxidized to the mercuric ion,
return to the general circulation is impeded, and mercury can be retained in brain tissue. The elimination
of elemental mercury occurs via urine, feces, exhaled air, sweat, and saliva. The pattern of excretion is
dependent on the extent to which elemental mercury has been oxidized to mercuric mercury.
Inorganic Mercury
Absorption of inorganic mercury through the gastrointestinal tract varies with the particular
mercuric salt involved. Absorption decreases with decreasing solubility. Estimates of the percentage of
inorganic mercury that is absorbed vary; as much as 20% may be absorbed. Available data indicate that
absorption of mercuric chloride from the gastrointestinal tract results from an electrostatic interaction
with the brush border membrane and limited passive diffusion. Increases in intestinal pH, high doses of
mercuric chloride causing a corrosive action, a milk diet (e.g., neonates) and increases in pinocytotic
activity in the gastrointestinal tract (e.g., neonates) have all been associated with increased absorption of
inorganic mercury. Inorganic mercury has a limited capacity for penetrating the blood-brain or placental
barriers. There is some evidence indicating that mercuric mercury in the body following oral exposures
can be reduced to elemental mercury and excreted via exhaled air. Because of the relatively poor
absorption of orally administered inorganic mercury, the majority of the ingested dose in humans is
excreted through the feces.
Methylmercury
Methylmercury is rapidly and extensively absorbed through the gastrointestinal tract. Absorption
information following inhalation exposures is limited. This form of mercury is distributed throughout the
body and easily penetrates the blood-brain and placental barriers in humans and animals. Methylmercury
transport into tissues appears to be mediated by the formation of a methylmercury-cysteine complex.
This complex is structurally similar to methionine and is transported into cells via a widely distributed
neutral amino acid carrier protein. Methylmercury in the body is considered to be relatively stable and is
only slowly demethylated to form mercuric mercury in rats. It is hypothesized that methylmercury
metabolism may be related to a latent or silent period observed in epidemiological studies observed as a
delay in the onset of specific adverse effects. Methylmercury has a relatively long biological half-life in
humans; estimates range from 44 to 80 days. Excretion occurs via the feces, breast milk, and urine.
Biological Monitoring/Pharmacokinetic Models
Chapter 2 provides information on biological monitoring of mercury as well as a summary of the
development of pharmacokinetic models for mercury. The most common biological samples analyzed
for mercury are blood, urine, and scalp hair. The methods most frequently used to determine the mercury
levels in these sample types include atomic absorption spectrometry, neutron activation analysis, X-ray
fluorescence, and gas chromatography.
ES-2
-------
Both simple and complex multi-compartmental models have been described in the literature. A
recent report (Gearhart et al. 1995) presents an approach based upon data from human, rat, and monkey
data that could be used for characterizing dose-response data both adults and neonates.
Biological Effects
Chapter 3 presents summary information on the toxicity of elemental mercury, mercuric mercury
and methylmercury to various organ systems. The primary targets for toxicity of mercury and mercury
compounds are the nervous system, the kidney, and the developing fetus. Other systems that may be
affected include the respiratory, cardiovascular, gastrointestinal, hematologic, immune, and reproductive
systems. For each form of mercury and each of the endpoints addressed, information from
epidemiological studies, human case studies, and animal toxicity studies is summarized in tabular form.
Critical studies are discussed in the accompanying text.
Elemental Mercury
A number of epidemiological studies have been conducted that examined cancer mortality and/or
morbidity among workers occupationally exposed to elemental mercury. All of these studies, however,
have limitations which compromise the interpretation of their results; these limitations include small
sample sizes, probable exposure to other known lung carcinogens, failure to consider confounding factors
such as smoking, and/or failure to observe correlations between estimated exposure and the cancer
incidence. Only one animal study was identified that examined cancer incidence in animals exposed (by
injection) to elemental mercury. While tumors were found at contact sites, the study was incompletely
reported as to controls and statistics and, thus, considered inadequate for the purpose of risk assessment.
Findings from genotoxicity assays are limited and do not provide supporting evidence for a carcinogenic
effect of elemental mercury.
Effects on the nervous system appear to be the most sensitive toxicological endpoint observed
following exposure to elemental mercury. Symptoms associated with elemental mercury-induced
neurotoxicity include the following: tremors, initially affecting the hands and sometimes spreading to
other parts of the body; emotional lability, often referred to as "erethism" and characterized by
irritability, excessive shyness, confidence loss, and nervousness; insomnia; neuromuscular changes (e.g.,
weakness, muscle atrophy, muscle twitching); headaches; polyneuropathy (e.g., paresthesia, stocking-
glove sensory loss, hyperactive tendon reflexes, slowed sensory and motor nerve conduction velocities);
and memory loss and performance deficits in test of cognitive function. At higher concentrations,
adverse renal effects and pulmonary dysfunction may also be observed.
A few studies have provided suggestive evidence for potential reproductive toxicity associated
with exposure to elemental mercury. Data from two studies in rats demonstrate developmental effects of
elemental mercury exposure. These were behavioral changes associated with both in utero and perinatal
exposure.
Inorganic Mercury
There is no evidence in humans linking exposure to mercuric chloride with carcinogenic effects.
Data in animals are limited. Focal hyperplasia and squamous cell papillomas of the forestomach as well
as thyroid follicular adenomas and carcinomas were observed in male rats gavaged with mercuric
chloride. In the same study, evidence for an increased incidence of squamous cell forestomach
papillomas in female rats and renal adenomas and carcinomas in male mice were considered equivocal.
ES-3
-------
All increased tumor incidences were observed in excess of the maximum tolerated dose (MTD). In this
context, the relevance of the tumors to human health evaluation has been questioned. Results from in
vitro and in vivo tests for genotoxicity have been mixed and do not provide strong supporting data for
carcinogenicity.
There are some data indicating that mercuric chloride may be a germ cell mutagen. Positive
results have been obtained for chromosomal aberrations in multiple systems, and evidence suggests that
mercuric chloride can reach female gonadal tissue.
The most sensitive general systemic adverse effect observed following exposure to inorganic
mercury is the formation of mercuric mercury-induced autoimmune glomerulonephritis. The production
and deposition of IgG antibodies to the glomerular basement membrane can be considered the first step
in the formation of this mercuric-mercury-induced autoimmune glomerulonephritis.
Several studies in animals have evaluated the potential for developmental toxicity to occur
following exposure to various inorganic salts. While the evidence suggests that developmental effects
may occur, all of the studies have significant limitations.
Methylmercury
Three human studies that examined the relationship between methylmercury and cancer
incidence were considered extremely limited because of study design inappropriate for risk assessment or
incomplete data reporting. Evidence from animal studies provides limited evidence of carcinogenicity.
Male ICR and B6C3F1 mice exposed orally to methylmercuric chloride were observed to have an
increased incidence of renal adenomas, adenocarcinomas, and carcinomas. Renal epithelial cell
hyperplasia and tumors, however, were observed only in the presence of profound nephrotoxicity
suggesting that the tumors may be a consequence of reparative changes to the damaged kidneys. Tumors
were observed at a single site, in a single species and sex.
Methylmercury appears to be clastogenic but not a potent mutagen. Studies have also shown
evidence that methylmercury may induce mammalian germ cell chromosome aberrations. There are a
number of studies in both humans and experimental animals that show methylmercury to be a
developmental toxicant. Neurotoxicity in offspring is the most commonly observed effect and the effect
seen at lowest exposures.
A significant body of human studies exists for evaluating the potential systemic toxicity of
methylmercury. This data base is the result of studying two large scale poisoning episodes in Japan and
Iraq as well as several epidemiological studies assessing populations that consume significant quantities
offish. In addition, much research on the toxicity of methylmercury has been conducted in animals
including non-human primates.
The critical target for methylmercury toxicity is the nervous system. The developing fetus may
be at particular risk from methylmercury exposure. Offspring born of women exposed to methylmercury
during pregnancy have exhibited a variety of developmental neurological abnormalities, including the
following: delayed onset of walking, delayed onset of talking, cerebral palsy, altered muscle tone and
deep tendon reflexes, and reduced neurological test scores. Maternal toxicity may or may not have been
present during pregnancy for those offspring exhibiting adverse effects. For the general population, the
critical effects observed following methylmercury exposure are multiple central nervous system effects
including ataxia and paresthesia.
ES-4
-------
A latent or silent period has been observed in some epidemiological and animal studies
indicating a delay in the onset of adverse effects. It is hypothesized this delay may be related to
methylmercury metabolism.
Sensitive Subpopulations
A susceptible population is a group that may experience more severe adverse effects at
comparable exposure levels or adverse effects at lower exposure levels than the general population. The
greater response of these sensitive subpopulations may be a result of a variety of intrinsic or extrinsic
factors. For mercury, the most sensitive subpopulations may be developing organisms. Data are also
available indicating that other factors may be associated with the identification of sensitive
subpopulations including the following: age; gender; dietary insufficiencies of zinc, glutathione, or
antioxidants; predisposition for autoimmune glomerulonephritis; and predisposition for acrodynia. More
information on sensitive subpopulations is presented in Chapter 4.
Interactions
There are data demonstrating that a number of substances affect the pharmacokinetics and/or
toxicity of mercury compounds. Of most interest is the potential interaction of selenium and mercury.
Selenium is known to bioaccumulate in fish, so exposure to methylmercury from fish consumption may
be associated with exposure to increased levels of selenium. There are data indicating that selenium co-
administered with methylmercury can form selenium-methylmercury complexes. The formation of these
complexes may temporarily prevent methylmercury-induced tissue damage but also may delay excretion
of the methylmercury. Thus, formation of selenium-methylmercury complexes may not reduce
methylmercury toxicity but rather may delay onset of symptoms. More information is needed to
understand the possible interaction of selenium with methylmercury.
There is potential for interaction between various forms of mercury and ethanol, thiol
compounds, tellurium, potassium dichromate, zinc, atrazine, and vitamins C and E.
Hazard Identification/Dose-Response Assessment
The available toxicological and epidemiological evidence was evaluated, and U.S. EPA risk
assessment guidelines and methodologies were applied to hazard identification for various endpoints;
namely, carcinogenicity, germ cell mutagenicity, developmental toxicity, and general systemic toxicity.
Data supported quantitative assessments of systemic toxicity. For elemental mercury, an inhalation
reference concentration (RfC1) was calculated; oral reference doses (RfD1) were calculated for inorganic
mercury and methylmercury. Data for carcinogenicity of inorganic and methylmercury were judged to be
inadequate in humans and limited from animal bioassays. The carcinogenicity data for all forms of
mercury evaluated were not sufficient to support a quantitative assessment. No quantitative estimates
were done for developmental toxicity. Table ES-1 summarizes the hazard identification and dose-
response information for elemental mercury, inorganic mercury, and organic mercury. The bases for
these decisions and the methodologies applied are presented in Chapter 6.
1 The oral RfD and the inhalation RfC are estimates (with uncertainty spanning perhaps an order of
magnitude) of a daily exposure to the human population (including sensitive subpopulations) that is
likely to be without an appreciable risk of deleterious health effects during a lifetime.
ES-5
-------
Table ES-1
Summary of U.S. EPA Hazard Identification/Dose-response Assessment
for Mercury and Mercury Compounds
Form
of
Mercury
Elemental
Inorganic
Organic
Oral RiD
(mg/kg-day)
n/a"
0.0003C
(mercuric
chloride)
0.0001'
(methyl-
mercury)
Inhalation
RfC
(mg/m3)
0.0003b
Not"1
verifiable
n/a
Cancer
Weight-of-
evidence
Rating
D, not classifiable
as to human
carcinogenicity
C, possible
human carcinogen
C, possible
human carcinogen
Cancer
Slope
Factor
n/a
n/a
n/a
Germ Cell
Mutagenicity
Low weight of
evidence
Moderate weight
of evidence
High weight of
evidence
Developmental
Toxicity
Data Base
Characterization
Insufficient human
evidence; sufficient
animal evidence
Insufficient
evidence
Sufficient human
and animal data
* Not available; data do not support development of a value at this time.
b Critical effect is neurological toxicity (hand tremor; increases in memory disturbances; slight subjective and objective
evidence of autoimmune dysfunction) in adults.
0 Critical effect is renal toxicity resulting from an autoimmune disease caused by the accumulation of a hapten-mercury
complex in the glomerular region of the kidneys.
d Data were judged insufficient for calculation of RfC.
' Critical effect is neurological toxicity in progeny of exposed women, RfD calculated using a benchmark dose (10%).
Ongoing Research
While much data has been collected on the potential toxicity of mercury and mercury
compounds, much is still unknown. Two ongoing epidemiological studies are now providing critical
information on the developmental toxicity of methylmercury. One study, being conducted in the
Seychelles Islands, is evaluating dose-response relationships in a human population with dietary
exposures (fish) at levels believed to be in the range of the threshold for developmental toxicity. The
second study, conducted in the Faroe Islands, is assessing mercury exposure in a population that
consumes a relatively large quantity of marine fish and marine mammals. Children exposed to
methylmercury in utero and followed through 6 years of age have been assessed for mercury exposure
and neurological developmental. Published data from these studies are summarized in Chapter 3.
Implications of ongoing research is discussed along with uncertainties in risk assessments in Chapter 6.
Research Needs
Specifically, information is needed to reduce the uncertainties associated with the current oral
RfDs and inhalation RfCs. More work with respect to both dose and duration of exposure would also
allow for potentially assessing effects above the RfD/RfC. Limited evidence suggests that
methylmercury and mercuric chloride are possible human carcinogens. Data are not sufficient to classify
the potential carcinogenicity of elemental mercury. Research on mode of action in induction of tumors at
high mercury dose will be of particular use in defining the nature of the dose response relationship for
carcinogenicity. At this time data have been judged insufficient for calculation of quantitative
developmental toxicity estimates for elemental and inorganic mercury; research toward this end should
be encouraged. While some pharmacokinetic models have been developed additional work to ensure the
ES-6
-------
applicability of these to risk assessment should be pursued. In particular work aimed at validation of a
fetal pharmacokinetic model and research in support of toxicokinetics will be useful.
Conclusions
The following conclusions progress from those with greater certainty to those with lesser
certainty.
• The three forms of mercury discussed in this Report can present a human health hazard.
• Neurotoxicity is the most sensitive indicator of adverse effects in humans exposed to
elemental mercury and methylmercury.
• Immune-mediated kidney toxicity is the most sensitive indicator of toxic effects of
exposure to inorganic mercury. This judgement is largely based on results in
experimental animals.
• Methylmercury is a developmental toxicant in humans.
• Methylmercury is likely to be a human germ cell mutagen. This judgement is based on
data from human studies, genetic toxicology studies in animals and a consideration of the
pharmacokinetics of methylmercury.
• An RfD for ingested methylmercury based on neurotoxic effects observed in Iraqi
children exposed in utero is 1 x 10~4 mg/kg-day. The threshold estimate derived using a
benchmark dose approach is not model dependent (polynomial vs. Weibull). The
estimate is not much affected by data grouping, but is dependent on response
classification and on parameters used in determination of ingestion relative to measured
mercury in hair.
• An RfC for inhaled elemental mercury based on neurotoxic effects in exposed workers is
3 x 10~4mg/m3.
• An RfD for ingested inorganic mercury based on immune-mediated kidney effects in
Brown-Norway rats is 3 x 10~4 mg/kg-day.
• Elemental mercury is a developmental toxicant in experimental animals. If the
mechanisms of action producing developmental toxicity in animals occur in humans,
elemental mercury is very likely to produce developmental effects in exposed human
populations. U.S. EPA has made no estimate of dose response for developmental effects
of elemental mercury.
• Methylmercury and inorganic mercury produce tumors in experimental animals at toxic
doses. If the mechanisms of action which induced tumors in the animal models could
occur in humans, it is possible that tumors could be induced in exposed humans by these
forms of mercury. It is likely, however, that cancer would be induced only after
mercury exposures in excess of those producing other types of toxic response.
ES-7
-------
There are many uncertainties associated with this analysis, due to an incomplete understanding
of the toxicity of mercury and mercury compounds. The sources of uncertainty include the
following:
• The data serving as the basis for the methylmercury RfD were from a population
ingesting contaminated seed grain. The nutritional status of this group may not be
similar to that of U.S. populations. The exposure was for a short albeit critical period of
time. It is likely that there is a range of response among individuals to methylmercury
exposure. The selenium status of the exposed Iraqi population is not certain, nor is it
established the extent to which selenium has an effect on mercury toxicity.
• There was no NOAEL (no-observable-adverse-effect level) for estimation of a threshold
for all developmental endpoints. A benchmark was estimated using a Weibull model on
grouped data. Use of an estimate other than the 95% lower limit on 10% response
provides alternate estimates. Other modeling approaches using data which have not been
grouped provide similar estimates. Benchmark doses, NOAELs, LOAELs, from other
human studies provide support for the benchmark used in the RfD.
• Ingestion levels of methylmercury associated with measured mercury in hair were
estimated based on pharmacokinetic parameters derived from evaluation of the extant
literature. Use of other plausible values for these parameters results in (relatively small)
changes in the exposure estimate.
• While there are data to show that the developing fetus is more susceptible to
methylmercury toxicity than adults, there are not sufficient data to support calculation of
a separate RfD for children (vs. adults).
• The RfD for inorganic mercury is based on data in experimental animals; there is
uncertainty in extrapolation to humans. It is thought that these animals constitute a good
surrogate for a sensitive human subpopulation. The data were from less than lifetime
exposures; there is uncertainty in extrapolation to a lifetime RfD. There was no NOAEL
in the studies; there is uncertainty in extrapolation to a NOAEL or in estimation of a
threshold for effects in animals.
• The RfC for elemental mercury was based on studies in exposed workers for which there
is no reported NOAEL; there is uncertainty in estimating the no effect level in these
populations. There is uncertainty as to whether reproductive effects could be occurring
at lower exposure levels than those which produced the observed neurotoxicity.
• There are insufficient data to determine whether elemental mercury induces carcinogenic
effects in experimental animals.
• Data are not sufficient to judge if elemental and inorganic mercury are germ cell
mutagens.
• U.S. EPA did not formally evaluate data on mercury for reproductive effects.
ES-8
-------
To improve the risk assessment for mercury and mercury compounds. U.S. EPA would need the
following:
• Results from ongoing studies in human populations with measurable exposure to
methylmercury.
• Results for immune-mediated kidney effects from lifetime studies of sensitive animals
exposed to inorganic mercury. Definitive data from human studies on effects of
exposure to inorganic mercury.
• Data on inhalation effects of inorganic mercury exposure.
• Dose response data for developmental effects of elemental and inorganic mercury.
• Reproductive studies and analysis for all forms of mercury.
• Data on mode of action of inorganic and methylmercury tumor induction.
• Validated physiologically-based pharmacokinetic models for mercury which include a
fetal component.
Based on the extant data and knowledge of developing studies, the following outcomes can be
expected:
• Human populations exposed to sufficiently high levels of elemental mercury will have
increased incidence of neurotoxic effects.
• Human populations exposed to sufficiently high levels of methylmercury either in utero
or post partum will have increased incidence of neurotoxic effects.
• Human populations exposed to sufficiently high levels of inorganic mercury will have
increased incidence of systemic effects including immune-mediated kidney effects.
• The RfDs and RfC calculated by U.S. EPA for systemic toxic effects of mercury are
expected to be amounts of exposure that can be incurred on a daily basis for a lifetime
without anticipation of adverse effects. This expectation is for populations including
susceptible subpopulations.
• The RfDs are protective against carcinogenic effects; tumor induction in animals was
observed only at doses likely to produce systemic toxic effects.
ES-9
-------
1. INTRODUCTION
Section 112(n)(l)(B) of the Clean Air Act (CAA), as amended in 1990, requires the U.S.
Environmental Protection Agency (U.S. EPA) to submit a study on atmospheric mercury emissions to
Congress. The sources of emissions that must be studied include electric utility steam generating units,
municipal waste combustion units and other sources, including area sources. Congress directed that the
Mercury Study evaluate many aspects of mercury emissions, including the rate and mass of emissions,
their health and environmental effects, technologies to control such emissions and the costs of such
controls.
In response to this mandate, U.S. EPA has prepared an eight-volume Mercury Study Report to
Congress. The eight volumes are as follows:
I. Executive Summary
II. An Inventory of Anthropogenic Mercury Emissions in the United States
III. Fate and Transport of Mercury
IV. An Assessment of Exposure to Mercury in the United States
V. Health Effects of Mercury and Mercury Compounds
VI. An Ecological Assessment for Anthropogenic Mercury Emissions in the United States
VII. Characterization of Human Health and Wildlife Risks from Mercury Exposure in the
United States
VIII. An Evaluation of Mercury Control Technologies and Costs
This volume (Volume V) addresses the potential human health effects associated with exposure
to mercury. It summarizes the available human and animal studies and other supporting information
relevant to the toxicity of mercury and mercury compounds in humans. It also summarizes U.S. EPA's
current overall assessments of hazard and quantitative dose-response for various categories of toxic
effects. This volume presents data relevant to assessment of potential effects on human health for
elemental mercury, inorganic mercury and methylmercury. Organic mercury compounds other than
methylmercury are generally not considered in this volume. Chapter 2 discusses the toxicokinetics of
mercury, including information on absorption, distribution, metabolism and excretion. Chapter 3 is a
summary of the toxicity literature for mercury. It is organized into three main subsections, corresponding
to elemental mercury, inorganic mercury and methylmercury. Within each of these subsections, the
study data are presented according to the effect type (e.g., death, renal toxicity, developmental toxicity,
cancer). For each effect type, separate summary tables in similar formats are used to present the
available data from human epidemiological studies, human case studies, and animal studies.
Chapter 6, Hazard Identification and Dose-Response Assessment, presents U.S. EPA's
assessments of the hazard presented by various forms of mercury and, where possible, the quantitative
dose-response information that is used in risk assessments of mercury. Chapters 4 and 5 briefly discuss
populations with increased susceptibility to mercury and interactions between exposure to mercury and
other substances. Ongoing research and research needs are described in Chapter 7, and Chapter 8 lists
the references cited. Appendix A documents the dose conversion equations and factors used. Appendix
B consists of RfD, RfC and cancer risk summaries for U.S. EPA's Integrated Risk Information System
(IRIS). Appendix C lists the participants of a U.S. EPA-sponsored workshop on mercury issues held in
1987. Appendix D presents an analysis of uncertainty and variability in the methylmercury human
effects threshold estimate.
1-1
-------
2. TOXICOKINETICS
This chapter describes the toxicokinetics (i.e., absorption, distribution, metabolism and
excretion) of mercury and mercury compounds in the body. Biomarkers of exposure and methods of
analysis for measuring mercury levels in biological samples are discussed. The biotransformation of
mercury in the environment is discussed in Volume III.
The absorption of elemental mercury vapor occurs rapidly through the lungs, but it is poorly
absorbed from the gastrointestinal tract. Oral absorption of inorganic mercury involves absorption
through the gastrointestinal tract; absorption information for the inhalation route is limited.
Methylmercury is rapidly and extensively absorbed through the gastrointestinal tract.
Once absorbed, elemental mercury is readily distributed throughout the body; it crosses both
placental and blood-brain barriers. Elemental mercury is oxidized to inorganic divalent mercury by the
hydrogen peroxidase-catalase pathway, which is present in most tissues. The oxidation of elemental
mercury to the inorganic mercuric cation in the brain can result in retention in the brain. Inorganic
mercury has poor lipophilicity and a reduced capacity for penetrating the blood-brain or placental
barriers. Once elemental mercury crosses the placental or blood-brain barriers and is oxidized to the
mercuric ion, return to the general circulation is impeded, and mercury can retained in brain tissue.
Recent studies indicate that transport and distribution of methylmercury is carrier-mediated.
Methylmercury penetrates the blood-brain and placental barriers, can be converted to mercuric ion, and
may accumulate in the brain and fetus.
The elimination of elemental mercury occurs via the urine, feces and expired air. Exposure to
mercuric mercury results in the elimination of mercury in the urine and feces. Methylmercury is excreted
primarily in the feces (mostly in the inorganic form) by humans.
2.1 Absorption
2.1.1 Elemental Mercury
2.1.1.1 Inhalation
Elemental mercury vapors are readily absorbed through the lungs. Studies in human volunteers
have shown that approximately 75-85% of an inhaled dose of elemental mercury vapor was absorbed by
the body (Nielsen-Kudsk 1965; Oikawa et al. 1982; Teisinger and Fiserova-Bergerova 1965; Hursh 1985;
Hursh et al. 1985). The high lipid solubility of elemental mercury vapor relative to its vapor pressure
favors its rapid diffusion across alveolar membranes and dissolution in blood lipids (Berlin et al. 1969b).
2.1.1.2 Oral
Liquid metallic mercury is very poorly absorbed from the gastrointestinal tract. In rats, less than
0.01% of an ingested dose of metallic mercury was absorbed (Bornmann et al. 1970). The release of
mercury vapor from liquid elemental mercury in the gastrointestinal tract and the subsequent absorption
of the released vapor is limited by reaction of the mercury with sulfur to form mercuric sulfide. The
mercuric sulfide coats ingested metallic mercury, preventing release of elemental vapor (Berlin 1986).
2-1
-------
2.1.1.3 Dermal
Elemental mercury vapor is absorbed through the skin of humans at an average rate of 0.024 ng
Hg/cm2 (skin) for every one mg/m3 in the air (Hursh et al. 1989). This rate of dermal absorption is
sufficient to account for less than 3% of the total amount absorbed during exposure to mercury vapor
(greater than 97% of the absorption occurs through the lungs). Dermal absorption of liquid metallic
mercury has been demonstrated in experimental animals (Schamberg et al. 1918); however, the extent of
absorption was not quantified. Koizumi et al. (1994) measured mercury absorption through the skin of
F344 rats exposed to solutions of industrially generated dust containing mercury. After 3 days mean
blood concentration in dust-exposed rats was 15.5 //g/L compared to 3 //g/L for saline controls.
2.1.2 Inorganic Mercury
2.1.2.1 Inhalation
There is limited information suggesting that absorption occurs after inhalation of aerosols of
mercuric chloride. Clarkson (1989) reported absorption to be 40% in dogs via inhalation. Inhalation
exposure of rats to an aerosol of a 1% mercuric chloride solution for 1 hour/day, 4 days/week for 2
months resulted in retention of 5-6 //g HgCl/HR aerosol/100 g body weight or approximately 37-44 (ig
Hg/kg-day (Bernaudin et al. 1981). The authors prepared the aerosol "with reference to the maximum
allowable air concentrations (0.10 mg Hg/m3) for a man". Retention was defined at the end of exposure
as total mercury in the rat carcass minus skin and hair. It is unknown to what extent the amount retained
represented absorption through the lungs or absorption of material cleared from the respiratory tract by
mucociliary activity and ultimately swallowed.
2.1.2.2 Oral
The absorption of mercuric mercury from the gastrointestinal tract has been estimated at
approximately 7-15% in human volunteers following oral administration of radiolabeled inorganic
mercury (Miettinen 1973; Rahola et al. 1973). Recent data from studies in mice, however, suggest that
"true" absorption may be closer to 20% but appears lower due to intestinal pH, compound dissociation,
age, diet, rapid biliary secretion and excretion in the feces (Kostial et al. 1978; Nielsen 1992). Because
the excretion of absorbed mercury is rapid, mercury levels detected in the gastrointestinal tract most
likely represent both unabsorbed and excreted mercury in the studies by Miettinen (1973) and Rahola et
al. (1973). The absorption of mercuric chloride from the gastrointestinal tract is not believed to depend
on any specific transport mechanism, reactive sulfhydryl groups, or oxidative metabolism (Foulkes and
Bergman 1993). Rather, uptake appears to result from an electrostatic interaction with the brush border
membrane and limited passive diffusion. Several factors have been identified that modulate absorption
of mercuric mercury from the gastrointestinal tract. At high doses, the corrosive action of mercuric
chloride may increase its uptake by breaking down membrane barriers between the ions and the blood.
Increases in intestinal pH also increase absorption (Endo et al. 1990). Increased uptake also occurs in
neonates (Kostial et al. 1978). The increased absorption in neonates is believed to be due in part to the
milk diet of neonates (increased absorption was observed in adults given a milk diet) and in part to the
increased pinocytotic activity in the gastrointestinal tract that occurs in the very young (Kostial et al.
1978). Diffusion through aqueous channels present in the immature brush border of neonates has also
been suggested to account for the greater absorption in the very young (Foulkes and Bergman 1993).
Absorption of mercuric salts from the gastrointestinal tract varies with the particular salt
involved. Absorption decreases with decreasing solubility (Endo et al. 1990). For example, the poorly
2-2
-------
soluble salt mercuric sulfide is not absorbed from the gastrointestinal tract as well as the more soluble
mercuric chloride salt (Sin et al. 1983).
Mercurous salts in the form of calomel (long in use as a therapeutic agent) are insoluble in water
and are poorly absorbed from the gastrointestinal tract (Clarkson 1993a). Long term use of calomel,
however, has resulted in toxicity in humans (Davis et al. 1974).
2.1.2.3 Dermal
Dermal absorption of mercuric chloride has been observed in treated guinea pigs (Skog and
Wahlberg 1964). Approximately 2-3% of an applied dose was absorbed during a 5-hour period.
Absorption was measured both by disappearance of the applied compound and by appearance in kidney,
liver, urine and blood.
2.1.3 Methylmercury
2.1.3.1 Inhalation
Inhaled methylmercury vapors are absorbed through the lungs. Fang (1980) did not measure
percent absorbed but showed a correlation between tissue mercury levels and both exposure level and
duration in rats exposed to radioactively labelled methylmercury vapor.
2.1.3.2 Oral
Methylmercury is efficiently absorbed from the gastrointestinal tract. Approximately 95% of
methylmercury in fish ingested by volunteers was absorbed from the gastrointestinal tract (Aberg et al.
1969; Miettinen 1973). Similarly, when radiolabeled methylmercuric nitrate was administered in water
to volunteers, uptake was greater than 95% (Aberg et al. 1969).
Reports of the percentage of absorbed methylmercury distributed to the blood range from 1% to
10%. Following the ingestion of a single meal of methylmercury-contaminated fish, Kershaw et al.
(1980) found that blood accounted for 5.9% of absorbed methylmercury, while Miettinen et al. (1971)
found an initial value of 10%, decreasing to about 5% over the first 100 days. In a population that
chronically ingested fish with high methylmercury levels, approximately 1% of the absorbed dose was
distributed to the blood (Sherlock et al. 1982).
2.1.3.3 Dermal
Dermal absorption of the methylmercuric cation (CH3Hg)+ (as the dicyandiamide salt) has also
been observed in treated guinea pigs (Skog and Wahlberg 1964). Approximately 3-5% of the applied
dose was absorbed during a 5-hour period. Absorption was measured both by disappearance of the
applied compound and by appearance in kidney, liver, urine and blood.
2-3
-------
2.2 Distribution
2.2.1 Elemental Mercury
Because of its lipophilicity, absorbed elemental mercury vapor readily distributes throughout the
body, crossing the blood-brain barrier in humans (Hursh et al., 1976; Nordberg and Serenius, 1969) and
the placenta in rats and mice (Clarkson et al., 1972). The distribution of absorbed elemental mercury is
limited primarily by the oxidation of elemental mercury to the mercuric ion and reduced ability of the
mercuric ion to cross membrane barriers. The oxidation is sufficiently slow, however, to allow
distribution to all tissues and organs. Once it is oxidized to the mercuric ion, it is indistinguishable from
Hg2+ from inorganic sources (i.e., the highest levels of mercury accumulate in the kidneys) (Hursh et al.
1980; Rothstein and Hayes 1964). Based on an in vitro study by Hursh et al. (1988), oxidation of
mercury in the blood is slow and, therefore, inhaled mercury reaches the brain primarily unoxidized (i.e.,
as dissolved vapor) and is available for rapid penetration into brain cells. Once in the brain, oxidation of
elemental mercury to mercuric mercury in the brain enhances for the accumulation of mercury in these
tissues (Hursh et al. 1988; Takahata et al. 1970). For example, ten years after termination of exposure,
miners exposed to elemental mercury vapor had high concentrations of mercury (> 120 ppm) in the brain
(Takahata et al. 1970). A similar effect occurs when elemental mercury reaches the fetus and (after
oxidation) accumulates in the tissues as inorganic mercury (Dencker et al. 1983).
In the blood, elemental mercury initially distributes predominantly to the red blood cells; at
20 minutes, 98% of the mercury in the blood is found in the red blood cells. Several hours following
parenteral, oral or inhalation exposure, however, a stable ratio of red blood cell mercury to plasma
mercury of approximately 1:1 is established (Gerstner and Huff, 1977; Clarkson, 1972; Cherian et al.,
1978). The rise in plasma mercury levels was suggested to be due to binding to protein sulfhydryl groups
by mercuric mercury formed when the elemental mercury was oxidized.
2.2.2 Inorganic Mercury
In contrast to elemental mercury vapor and methylmercury, mercuric mercury does not penetrate
the blood-brain or placental barriers easily. Levels of mercury observed in the rat brain after injection of
mercuric nitrate were 10-fold lower than after inhalation of an equivalent dose of elemental mercury
vapor (Berlin et al. 1969a). Similarly, mercuric mercury shows only limited ability to penetrate to the
fetus (Garrett et al. 1972). Mercuric mercury does, however, accumulate in the mouse placenta (Berg
and Smith 1983; Mitani et al. 1978; Suzuki et al. 1984). In the blood, the mercuric ion is bound to
sulfhydryl groups present in the plasma and erythrocytes. The ratio of human red blood cell mercuric
mercury to plasma mercuric mercury is approximately 1:1 (0.53:1.20) (Hall etal. 1994). The half-life in
blood for humans was reported to range from 19.7 to 65.6 days in a study of five subjects treated with i.v.
mercuric nitrate (Hall et al. 1994). From the blood, mercuric mercury initially distributes to liver, but the
highest levels are generally observed in the kidneys (Newton and Fry 1978). With time after exposure,
accumulation in the kidneys may account for up to 90% of the total body burden (Rothstein and Hayes
1960). The mercury levels in the kidney are dose dependent, with increasing amounts occurring with
higher administered dose levels (Cember, 1962). The highest concentration of mercuric mercury in the
kidneys is found in the proximal tubules. Mercuric mercury induces metallothionein production in the
kidneys (Piotrowski et al. 1974). The high metallothionein levels in the kidneys may contribute to the
kidney's accumulation of mercuric mercury (Piotrowski et al. 1973).
In neonates, lower proportions of mercuric mercury distribute to the kidneys than in adult
animals (Jugo 1976). This results in higher distribution to other tissues. The protective blood-brain
2-4
-------
barrier is incomplete in fetal and neonatal animals, which may also contribute to the increased mercury
levels in immature brain. For example, the higher levels in the neonatal brain of rats and guinea pigs are
believed to be associated with the decrease in renal sequestration of the mercuric ion (Jugo 1976;
Yoshida et al. 1989). The higher levels observed in the livers of rat neonates may be attributable to
increased distribution to organs other than the kidney as well as to higher levels of neonatal hepatic
metallothionein (Daston et al. 1986).
2.2.3 Methylmercury
Methylmercury is distributed throughout the body, easily penetrating the blood-brain and
placental barriers in humans and animals (Clarkson 1972; Hansen 1988; Hansen et al. 1989; Nielsen and
Andersen 1992; Soria et al. 1992; Suzuki et al. 1984). By contrast with elemental mercury, studies in rats
indicate that methylmercury transport into tissues is mediated by the formation of a methylmercury-
cysteine complex (Aschner and Aschner 1990; Tanaka et al. 1991, 1992; Kerper et al. 1992). The
complex is structurally similar to methionine and is transported into cells via a widely distributed neutral
amino acid carrier protein. Methylmercury associates with water-soluble molecules (e.g., proteins) or
thiol-containing amino acids because of the high affinity of the methylmercuric cation (CHgHg^for the
sulfhydryl groups (SH)~. Complexes of methylmercury with cysteine have been identified in blood, liver
and bile of rats (Aschner and Aschner 1990).
Al-Shahristani and Shihab (1974) calculated a "biological half-life" of methylmercury in a study
of 48 male and female subjects who had ingested seed grain contaminated by organic mercurials. The
half-life ranged from 35 to 189 days with a mean of 72 days; it was determined from distribution of
mercury along head hair.
The blood half-life is 49-164 days in humans (Aberg et al. 1969; Miettinen et al. 1971) and
10-15 days in monkeys (Rice et al. 1989). Smith et al. (1994) determined a blood half-life of 32-60 days
in a study of seven adult males given i.v. methylmercury. In the blood, methylmercury is found
predominantly in the red blood cells (Kershaw et al. 1980; Thomas et al. 1986). In humans, the ratio of
red blood cell methylmercury to plasma methylmercury is approximately 20:1. This ratio varies in
animal species; the ratio is approximately 20:1 in primates and guinea pigs, 7:1 in mice, greater than
100:1 in rats and 42:1 in cats (Rollins et al. 1975; Magos 1987).
The clinical significance of the differences in the distribution of various forms of mercury in the
blood is that it permits diagnosis of the type of mercury to which an individual has been exposed. Short-
chain alkyl mercury compounds such as methylmercury or ethyl mercury are very stable in the body,
whereas long-chain compounds may be metabolized over time to the mercuric ion. The mercury
distribution in the blood, therefore, may shift from a distribution characteristic of methylmercury to one
more suggestive of inorganic mercury (Berlin 1986; Gerstner and Huff 1977).
Mercury has been found in the umbilical cord of human newborns at levels comparable to
maternal blood levels (Grandjean et al. 1992a). For lactating mothers, the clearance of mercury from the
blood appears to be faster than for non-lactating women. Lactating individuals have a blood half-life of
42 days compared to 75 days for non-lactating females among a group of people who had consumed
contaminated seed grain (Greenwood et al. 1978). This finding may be due to excretion of mercury via
the milk, increased food intake by mothers (which enhances biliary excretion) and/or altered hormonal
patterns in lactating mothers (which affect the excretion pattern).
2-5
-------
Methylmercury transport across the blood-brain barrier in rats may involve an amino acid carrier
(Kerper et al. 1992). Following acute exposure to methylmercury, most of the mercury in the brain is in
the organic form; however, with chronic exposures, a greater amount of the mercury in the brain is in the
inorganic form, suggesting that the rate of demethylation increases with long-term exposure (Aschner
and Aschner 1990). Rice (1989a, 1989b) demonstrated that tissue half-life in the brain may be
significantly longer than the blood half-life for methylmercury.
The bioaccumulation of methylmercury can be affected by age and sex (Thomas et al. 1982,
1986, 1988). After administration of methylmercury to rats, the females had higher peak levels of
mercury in the kidneys, primarily as methylmercury, compared to the males; inorganic mercury levels did
not differ significantly between the sexes (Thomas et al. 1986). Accumulation of mercury in the body is
also found to be higher in neonatal rats (Thomas et al. 1988) than in adult rats (Thomas et al. 1982). Ten
days after administration of methylmercury, 94% of the dose was still detected in neonates while =60%
was retained in adults (Thomas et al. 1988). The longer retention of mercury in the neonates may be
attributed to various factors including the high amount of mercury accumulated in the pelt of the
neonates due to lack of clearance (Thomas et al. 1988) and the lack of a fully developed biliary transport
system in the neonates (Ballatori and Clarkson 1982).
2.3 Metabolism
2.3.1 Elemental Mercury
Elemental mercury dissolved in the blood is rapidly oxidized in red blood cells to mercuric
mercury by catalase in the presence of hydrogen peroxide (Halbach and Clarkson 1978). Catalase is
found in many tissues, and oxidation by this pathway probably occurs throughout the body (Nielsen-
Kudsk 1973). The pathway is saturable, however, and hydrogen peroxide production is the rate-limiting
step (Magos et al. 1989). Blood and tissue levels of mercuric mercury following exposure to high
concentrations of elemental mercury are, therefore, lower than would be expected based on levels
observed following exposure to low levels.
2.3.2 Inorganic Mercury
Several investigators have observed exhalation of elemental mercury vapor after oral
administration of mercuric mercury to rats and mice, indicating that mercuric mercury in the body can be
reduced to elemental mercury (Clarkson and Rothstein 1964; Dunn et al. 198 la, 198 Ib; Sugata and
Clarkson 1979). The reduction of mercuric ion to elemental mercury may occur via cytochrome c,
NADPH and NADH, or a superoxide anion produced by the xanthine-xanthine oxidase system (Ogata et
al. 1987). There is no evidence that mercuric mercury is methylated to form methylmercury in
mammalian cells. The studies of Rowland et al. on the intestinal flora of the Wistar rat show that
microbes are responsible for at least a portion of mercuric chloride methylation in the gut.
Mercurous mercury is unstable in biological fluids and rapidly disassociates to one molecule of
elemental mercury and one ion of mercuric mercury (Clarkson 1972).
2.3.3 Methylmercury
Methylmercury in the body is relatively stable and is only slowly demethylated to form mercuric
mercury in rats (Norseth and Clarkson 1970). The demethylation appears to occur in tissue macrophages
(Suda and Takahashi 1986), intestinal microflora (Nakamura et al. 1977; Rowland et al. 1980) and fetal
2-6
-------
liver (Suzuki et al. 1984). In vitro demethylation has been reported to involve hydroxyl radicals
produced by cytochrome P-450 reductase (Suda and Hirayama 1992) or hypochlorous acid scavengers
(Suda and Takahashi 1992). Organic mercury compounds with longer alkyl chains are more readily
metabolized over time to the mercuric ion (Berlin, 1986).
Methylmercury metabolism may be related to the latent or silent period observed in
epidemiological studies from two methylmercury poisonings. During the latent period, both during and
after the cessation of exposure, the patient feels no untoward effects. It is possible that a number of
biochemical changes may take place in parallel during this period, and some may not be causatively
related to the clinical outcome. Ganther (1978) has hypothesized that the carbon-mercury bond in
methylmercury undergoes homolytic cleavage to release methyl free radicals. The free radicals are
expected to initiate a chain of events involving peroxidation of lipid constituents of the neuronal cells.
The onset of symptoms is delayed for the period of time that cellular systems are able to prevent or repair
effects of lipid peroxidation. When the cellular defense mechanisms are overwhelmed, rapid and
progressive degeneration of the tissue results. In the Iraqi poisoning incident, the latent period before
toxic signs were noted varied from a matter of weeks to months. By contrast, in the Japanese poisoning
incident, the latency was as long as a year or more. The difference in duration of the latent period may in
part be due to the presence of selenium in the fish ingested by the Japanese population. The role of
selenium in mercury toxicity is discussed further in Chapter 5.
2.4 Excretion
2.4.1 Elemental Mercury
Excretion of mercury after exposure to elemental mercury vapor may occur via exhaled air,
urine, feces, sweat and saliva. The pattern of excretion of elemental mercury changes as elemental
mercury is oxidized to mercuric mercury. During and immediately after an acute exposure, when
dissolved elemental mercury is still present in the blood, glomerular filtration of dissolved mercury vapor
occurs, and small amounts of mercury vapor can be found in the urine (Stopford et al. 1978). Mercury
vapor present in the blood may also be exhaled; human volunteers exhaled approximately 7% of the
retained dose within the first few days after exposure (Hursh et al. 1976). The half-life for excretion via
the lungs is approximately 18 hours. Approximately 80% of the mercury accumulated in the body is
eventually excreted as mercuric mercury. As the body burden of mercury is oxidized from elemental
mercury to mercuric mercury, the pattern of excretion becomes more similar to mercuric mercury
excretion. The majority of the excretion of mercuric mercury occurs in the feces and urine (Cherian et al.
1978). During the first few days after exposure of humans to mercury vapor, approximately four times
more mercury was excreted in the feces than in the urine (Cherian et al. 1978). With time, as the relative
mercury content of the kidneys increases, excretion by the urinary route also increases (Rothstein and
Hayes 1964). Tissue levels of mercury decrease at different rates, but the half-life for excretion of
whole-body mercury in humans (58 days) is estimated to be approximately equal to the half-life of
elimination from the kidneys (64 days), where most of the body burden is located (Hursh et al. 1976).
Excretion via the urine may be increased if mercury-induced damage of the renal tubular epithelium has
happened and exfoliation of damaged mercury-containing cells occurs (Magos 1973).
Excretion via sweat and saliva are thought to contribute only minimally to total excretion under
normal circumstances. In workers who have perspired profusely, however, the total amount of mercury
excreted in the sweat during 90 minutes ranged from 50% to 200% of that found in a 16-hour composite
sample of urine (Lovejoy et al. 1974).
2-7
-------
2.4.2 Inorganic Mercury
Because of the poor absorption of orally administered mercuric mercury, the majority (=85%) of
an ingested dose in humans is excreted in the feces within a few days after administration (Miettinen
1973). Hall et al. (1994) showed that for five adult male volunteers given i.v. mercuric nitrate and
evaluated for 70 days, 6.3-35% of the dose was excreted in urine and 17.9-38.1% in feces. For absorbed
inorganic mercury, the half-life for excretion has been estimated to be =40 days (Rahola et al. 1973) and
67 days with a range of 49-96 days (Hall et al. 1994). Information on the routes of excretion for
absorbed inorganic mercury are limited, but excretion would be expected to be similar to that of
inorganic mercury formed in rats by the oxidation of elemental mercury (Rothstein and Hayes 1964).
The majority of absorbed inorganic mercury is excreted in the urine (Berlin 1986).
Glomerular filtration is not thought to contribute substantially to urinary excretion of mercuric
mercury (Cherian et al. 1978). Rather, mercuric mercury is excreted in the urine primarily as sulfhydryl
conjugates (with cysteine or N-acetylcysteine) actively transported into the tubular lumen. Urinary levels
correlate with renal mercury concentrations rather than blood mercury levels.
Fecal excretion of mercury occurs as the result of excretion in the saliva, secretion through the
epithelium of the small intestines and colon and secretion in the bile (Berlin 1986). Secretion of
mercuric mercury in the bile is believed to result from active transport of a mercury-glutathione complex
across the canalicular membrane via the glutathione carrier (Ballatori and Clarkson 1982).
Mercuric mercury may also be excreted in breast milk during lactation (Yoshida et al. 1992).
The levels in breast milk are proportional to the plasma content. In maternal guinea pigs, milk levels
were approximately half of that found in plasma. After termination of exposure, however, mercury levels
in milk decreased at a slower rate than plasma mercury levels.
2.4.3 Methylmercury
Like inorganic mercury, methylmercury has a relatively long half-life of approximately 70-80
days in the human body (Aberg et al. 1969; Bernard and Purdue 1984; Miettinen 1973). Recently a
shorter half-life of 44 days was estimated by Smith et al. (1994) in their study of seven adult males
treated i.v. with methylmercury. In this study methylmercury and inorganic mercury concentrations in
blood and excreta were determined separately based on differential extractability into benzene. The
predominant species in the blood was methylmercury; there was no detectable methylmercury in the
urine.
The long half-life of methylmercury in the body is due, in part, to reabsorption of methylmercury
secreted into the bile (hepato-biliary cycling) (Norseth and Clarkson, 1971). In this cycle,
methylmercury forms a complex with glutathione in the hepatocyte, and the complex is secreted into the
bile via a glutathione carrier protein (Clarkson, 1993b). The methylmercury-glutathione complex in the
bile may be reabsorbed from the gallbladder and intestines into the blood. When microorganisms found
in the intestines demethylate methylmercury to form mercuric mercury, this cycle is broken, and fecal
excretion of mercury from methylmercury occurs (Rowland et al. 1980). Mercuric mercury is poorly
absorbed from the intestines, and that which is not reabsorbed is excreted in the feces. In humans,
approximately 90% of the absorbed dose of methylmercury is excreted in the feces as mercuric mercury.
Excretion via the urine is minor but slowly increases with time; at 100 days after dosing, urinary
excretion of mercury accounted for 20% of the daily amount excreted. The urinary excretion of mercury
may reflect the deposition of demethylated mercury in the kidneys and its subsequent excretion.
2-8
-------
In animals, the predominant route of methylmercury elimination also is the feces (Farris et al.
1993; Rollins et al. 1975; Thomas et al. 1987). As in humans, biliary excretion of methylmercury and its
demethylation in gastrointestinal flora have been reported in rats (Farris et al., 1993). After a single oral
dose of methylmercury, the major elimination route was the feces (65% of the administered dose as
inorganic mercury and 15% of the administered dose as methylmercury) and the minor route was urine
(1% of the administered dose as inorganic mercury and 4% of the administered dose as methylmercury)
(Farris etal. 1993).
In rat and monkey neonates, excretion of methylmercury is severely limited (Lok 1983; Thomas
et al. 1982). In rats dosed prior to 17 days of age, essentially no mercury was excreted (Thomas et al.
1982). By the time of weaning, the rate of excretion had increased to adult levels. The failure of
neonates to excrete methylmercury may be associated with the inability of suckling infants to secrete bile
(Ballatori and Clarkson 1982) and the decreased ability of intestinal microflorato demethylate
methylmercury during suckling (Rowland et al. 1977).
Methylmercury is also excreted in breast milk (Bakir et al. 1973; Sundberg and Oskarsson 1992).
The ratio of mercury in breast milk to mercury in whole blood was approximately 1:20 in women
exposed to methylmercury via contaminated grain in Iraq between 1971 and 1972 (Bakir et al. 1973).
Evidence from the Iraqi poisoning incident also showed that lactation decreased blood mercury clearance
half-times from 75 days in males and nonlactating females to 42 days in lactating females; the faster
clearance due to lactation was confirmed in mice (Greenwood et al. 1978). In mice, of the total mercury
in the breast milk, approximately 60% was estimated to be methylmercury. Skerfving (1988) has found
that 16% of mercury in human breast milk is methylmercury. Studies in animals indicate that the
mercury content of breast milk is proportional to the mercury content of plasma (Sundberg and
Oskarsson, 1992; Skerfving, 1988).
2.5 Biological Monitoring
This section describes the various biological media most frequently used when assessing mercury
exposure. In addition, this section describes the available analytical methods for measuring mercury in
biological samples. Reference values for mercury in standard biological media from the general
population are shown in Table 2-1. These values represent total mercury, not individual mercury species.
For hair and blood, these have been indexed to fish consumption as the most common route of exposure
in humans.
2.5.1 Elemental Mercury
Blood and urinary mercury are common to assess occupational mercury exposure.
2.5.1.1 Blood
In workers chronically exposed to mercury vapor, a good correlation was observed between
intensity of mercury vapor exposure and levels of mercury in the blood at the end of a workshift (Roels et
al. 1987). The usefulness of blood as a biomarker for exposure to elemental mercury depends on the time
elapsed since exposure and the level of exposure. For recent, high-level exposures, whole blood analysis
may be used to assess exposure (Clarkson et al. 1988). Mercury in the blood peaks rapidly, however, and
decreases with an initial half-life of approximately two to four days (Cherian et al. 1978). Thus,
evaluation of blood mercury is of limited value if a substantial amount of time has elapsed since
exposure. Also, dietary methylmercury contributes to the amount of mercury measured in blood. At low
2-9
-------
levels of elemental mercury exposure, the contribution of dietary methylmercury to the total blood
mercury may be high relative to that of the inhaled mercury, limiting the sensitivity of this biomarker.
Several studies have
Table 2-1
Reference Values for Total Mercury Concentrations in Biological Media for
the General Population
Medium
Whole blood
Fish consumption:
No fish meals
2 meals/week
2-4 meals/week
more than 4 meals/week
Urine
Scalp hair
Fish consumption:
once/mo
once/2 wk
once/wk
once/day
Mercury Concentration
1-8 ug/L
2 ug/L
2.0 ug/L
4.8 ug/L
8.4 ug/L
44.4 ug/L
4-5 ug/L
2ug/g
1 .4 ug/g
1.9 ug/g
2.5 ug/g
11.6 ug/g
Reference
WHO (1990)
Nordbergetal. (1992)
Brune (1991)
WHO (1990)
WHO (1990)
Airey(1983)
separated whole blood into its plasma and erythrocyte fractions in order to evaluate potential
confounding factors due to the presence of methylmercury (95% of methylmercury is found in the red
blood cell). Some published values indexed to fish consumption are in Table 2-1.
2.5.1.2 Urine
Urinary mercury is thought to indicate most closely the mercury levels present in the kidneys
(Clarkson et al. 1988). For most occupational exposures, urinary mercury has been used to estimate
exposure. In contrast to blood mercury levels, urinary mercury peaks approximately 2-3 weeks after
exposure and decreases at a much slower rate with a half-life of 40-60 days for short-term exposures and
90 days for long-term exposures (Barregard et al. 1992; Roels et al. 1991). The urine remains, therefore,
a more appropriate indicator for longer exposures than blood samples. As little dietary methylmercury is
excreted in the urine, the contribution of ingested methylmercury to the measured levels is not expected
to be high. Good correlations have been observed between urinary mercury levels and air levels of
mercury vapor; however, these correlations were obtained after correcting urinary mercury content for
variations in the urinary excretion rate (using urinary creatinine content or specific gravity) and after
standardizing the amount of time elapsed after exposure (Roels et al. 1987). Such steps are necessary
because considerable intra- and interindividual variability has been observed in the urinary excretion rate
(Barber and Wallis 1986; Piotrowski et al. 1975). Even when such precautions are taken, intraindividual
variability remains at =18% (Barregard et al. 1992; Roels et al. 1987).
2.5.1.3 Exhaled Air
Exhaled air has been suggested as a possible biomarker of exposure to elemental mercury vapor
because a portion of absorbed mercury vapor is excreted via the lungs. Excretion by this route has a half-
2-10
-------
life of approximately 18 hours (Hursh et al. 1976). At low levels of exposure, however, mercury vapor
released from dental amalgam may contribute substantially to the measured amount of mercury.
2.5.2 Inorganic Mercury
No information was identified in the literature that specifically assessed biological indicators for
inorganic mercury exposure. The information presented above for detection of mercury in blood and
urine after occupational exposure to elemental mercury vapor should also apply to inorganic mercury
exposures because elemental mercury vapor is rapidly converted to mercuric mercury after absorption.
2.5.3 Methylmercury
Blood and scalp hair are the primary indicators used to assess methylmercury exposure.
2.5.3.1 Blood
Because methylmercury freely distributes throughout the body, blood is a good indicator medium
for estimating methylmercury exposure. Because an individual's intake may fluctuate, blood levels may
not necessarily reflect mercury intake over time (Sherlock et al. 1982; Sherlock and Quinn, 1988). At
steady state, blood levels have been related to dose by the following equation (Kershaw et al. 1980):
, C x b x V
a =
Axf
Where:
C = concentration in blood (expressed in //g/L)
V = volume of blood (expressed as L)
b = the kinetic rate constant (day"1)
A = absorption rate (unitless)
F = fraction of dose that is present in blood
d = intake (//g/day)
It is useful to measure blood hematocrit and mercury concentrations in both whole blood and
plasma. From these data, the red blood cell to plasma mercury ratio may be determined, and interference
from exposure to high levels of elemental or inorganic mercury may be estimated (Clarkson et al. 1988).
2.5.3.2 Scalp Hair
Scalp hair can also be a good indicator medium for estimating methylmercury exposure (Phelps
et al. 1980). Methylmercury is incorporated into scalp hair at the hair follicle in proportion to its content
in blood. The hair-to-blood ratio in humans has been estimated as approximately 250:1 expressed as //g
Hg/g hair to mg Hg/L blood, but some difficulties in measurements, interindividual variation in body
burden, differences in hair growth rates, and variations in fresh and saltwater fish intake have led to
varying estimates (Birke et al. 1972; Skerfving 1974). Once incorporated into the hair, the
methylmercury is stable, and, therefore, gives a longitudinal history of blood methylmercury levels
2-11
-------
(Phelps et al. 1980; WHO, 1990). Analysis of hair mercury levels may be confounded by adsorption of
mercury vapor onto the hair strands (Francis et al. 1982).
2.5.4 Methods of Analysis for Measuring Mercury in Biological Samples
The most common methods used to determine mercury levels in blood, urine and hair of humans
and animals include atomic absorption spectrometry (AAS), neutron activation analysis (NAA), X-ray
fluorescence (XRF) and gas chromatography (GC). Table 2-2 identifies the major characteristics of these
methods.
Table 2-2
Analytical Methods for the Detection of Mercury in Biological Samples
Method
NAA
AAS
GC — Electron capture
XRF
Able to Distinguish
Methylmercury?
No
No
No"
Yes
No
Detection Limit (ppm)
0.1
9
PPB range
1.0
"low ppm"
References
Byrne and Kosta( 1974)
WHO (1976)
Hatch and Ott (1968)
Magos and Clarkson (1972)
Von Burg etal. (1974)
Cappon and Smith (1978)
Marsh etal. (1987)
" The Magos and Clarkson method estimates methylmercury by subtracting the inorganic mercury content from
the total mercury content.
2.6
Studies on Pharmacokinetic Models
2.6.1 Introduction
Pharmacokinetic modeling is a process by which administered dose, such as the amount of a
compound instilled into the body via inhalation, ingestion or parenteral route is used to estimate
measures of tissue dose which may not always be accessible to measurement by direct experimentation.
A pharmacokinetic model is employed to predict relevant measures of tissue dose under a wide range of
exposure conditions. In practice, the pharmacokinetic models used may incorporate features such as
compartmental analysis and physiologically-based models.
Reports available on the in vivo distribution of several types of mercury compounds provide
different physiokinetic relationships between the structure of mercury compounds and their behavior in
living organisms because the studies reported have been carried out under different experimental
conditions. Takeda et al. (1968) reported that in the rat, alkyl mercury compounds such as ethylmercuric
chloride and butylmercuric chloride were excreted more slowly and were retained in higher concentration
for a longer time in the body than mercuric chloride and phenylmercuric chloride. The distribution of
mercury in the brain was found to depend on the structure of the mercury compounds; relatively high
accumulation was observed for ethyl and n-butyl mercury compounds. Sebe and Itsuno (1962) reported
that after oral administration methyl-, ethyl-, and n-propylmercury compounds were neurotoxic to rats;
2-12
-------
n-butylmercury was not neurotoxic and thus presumably did not cross the blood-brain barrier. By
contrast, Suzuki et al. (1963, 1964) reported that ethylmercuric acetate and n-butylmercuric acetate had
similar patterns of distribution when subcutaneously administered to mice.
2.6.2 Inorganic mercury
Few controlled laboratory studies of pharmacokinetics of mercury in humans have been
published (Hursh et al. 1976, Rahola et al. 1973). Rahola et al. (1973) examined mercury absorption and
elimination after oral administration of mercuric nitrate to five male and five female volunteers, and
reported very low and variable rate of gastrointestinal absorption (8 to 25% dose). They reported a half-
time for inorganic mercury in human red blood cells of 16 days and whole body of 46 (32-60) days in
males and somewhat lower values in females. Hursh et al. (1976) found half-times for mercury clearance
from the body of 58 (35-90) days after exposure to mercury vapor. Whole body clearance from the
Rahola et al. (1973) study appeared biphasic with half-times of 2.3 days for the fast compartment and 42
(39-45) days for the slow compartment.
Low and variable rates of absorption of orally administered inorganic mercury in the Rahola et
al. (1973) study prompted Hall et al. (1994) to examine distribution of intravenously administered
inorganic mercury in human volunteers. In order to describe retention of mercury after transient
distributional effects, a one-compartment model was fit to the blood and body burden data after day 10,
assuming first order kinetics. The half-lives observed in the single compartment model for blood and
body burden were 30 (19.7-65.6 days) and 67 (48.6-95.5 days) days, respectively. The authors
attempted closer agreement between observed and predicted values by structuring a multicompartment
model. Measured mercury concentrations in blood, urine, feces, and whole body radioactive levels of
mercury were used in an a posteriori fashion to develop a model comprising six blood compartments,
one compartment each for feces and urine and a delayed compartment for feces. Inter-subject variability
(temporal pattern of blood mercury) and the existence of a kinetically distinct plasma pool (three distinct
compartments) for mercury resulted in equivocal predictions for blood, urine and feces; whether these
findings point to uncertainties of measurement of body burden or incomplete collection of excreta or
suggest other pathways of excretion, such as exhalation or sweating, is unknown. The authors concluded
that this type of complex pattern of blood kinetics, although unusual, is not without precedent. Four
kinetically distinct plasma pools of selenium has been reported after oral dosing with a stable isotopic
tracer (Patterson and Zech 1992). Hall et al. (1994) noted that the apparently linear kinetics observed for
the small tracer doses of i.v. inorganic mercury would likely change with toxicity associated with larger
or more frequent doses.
2.6.3 Methylmercury
Methylmercury is structurally the simplest of the organic mercurials; it bioaccumulates in certain
species offish, some of which are important human and wildlife foods. In order to elucidate the
mechanisms that influence the pharmacokinetics of both methylmercury and mercuric mercury and to
extrapolate further both intra- and inter-species extrapolation of experimental data for these toxins, Farris
and associates (1993) developed a physiological pharmacokinetic model for methylmercury and its
metabolite, mercuric mercury. This was done in growing rats dosed orally with labeled methylmercury
over a period of 98 days. Mercuric mercury accounted for less than 0.5% of total activity. Extensive sets
of metabolism and distribution data were collected to understand the processes that influence the
pharmacokinetics of both methylmercury and mercuric mercury. The model consisted of nine lumped
compartments, each of which represented a major site of mercury accumulation, distribution or
elimination. The carcass served as a residual compartment, which included all tissues and organs not
2-13
-------
separately incorporated into the model. Model simulations in this study were made with experimentally
determined concentrations of both inorganic and methylmercury in blood, brain, kidney and liver. The
data showed bidirectional and symmetric transport of both chemical species between blood and tissues
with relatively slow movement into and out of the brain. Some key parameters remained uncertain; for
example, the rate constant for demethylation is one of the most critical in adopting the model to other
species. This model, however, established a foundation for more complete understanding of
methylmercury pharmacokinetics. With further refinements, it could be applied to other species
including humans. To characterize health hazard from dietary methylmercury better, one needs to
understand the distribution of methylmercury in the body, the extent to which it accumulates and the rate
at which it is eliminated. Farris et al. (1993) noted that following methylmercury dosing there was a
buildup of inorganic mercury in tissues and that excreted mercury was predominantly mercuric;
methylmercury behaved as a single body pool, while mercuric mercury was handled differently in
different tissues.
Smith and associates (1994) made further refinements to the Farris et al. (1993) model. They
reported a multicompartment pharmacokinetic model for methylmercury and mercuric mercury in seven
human volunteers. This model simulated the long-term disposition of methylmercury and inorganic
mercury in humans following a single i.v. dose of radio-labeled methylmercury. This was a tracer
amount to avoid toxic or saturation effects. The behavior of both methylmercury and inorganic mercury
in the body was modeled with the simplest compartmental model which fitted the data; blood, urine and
feces data were used to fit the model. In this model the tracer dose was delivered to the first blood
compartment and subsequently distributed to two extra-vascular methylmercury compartments; two
distinct compartments (urine and feces) for inorganic mercury were added features. This five-
compartment model showed that inorganic mercury accumulated in the body and at longer times was the
predominant form of mercury present. The biological half-life of methylmercury in the body was
calculated to be 44 days, and 1.6% of the body burden was lost each day by both metabolism and
excretion.
To characterize neurological impairments of prenatal methylmercury exposure in children,
Gearhart and associates (1995) applied a more sophisticated multispecies pharmacokinetic model and
statistical dose-response analysis to an epidemiological study of a large population in New Zealand
(Kjellstrom et al. 1989) which featured relatively constant chronic exposure to methylmercury in fish.
The model for methylmercury in this study consisted of an adult with 11 compartments representing both
organ-specific and lumped tissues; eight compartments represented transport of methylmercury as flow-
limited, and three other compartments represented transport as diffusion-limited. The flow-limited
compartments were plasma, kidney, richly perfused, slowly perfused, brain-blood, placenta, liver and gut
compartments; RBC, brain and fetus were the diffusion-limited compartments. There were also four
other compartments in the model which were involved in methylmercury uptake and elimination:
methylmercury in the urine; and methylmercury and inorganic mercury in the hair, feces and the
intestinal lumen. The fetal sub-model for methylmercury consisted of four compartments: fetal plasma,
RBCs, brain and the remaining fetal body. This modeling effort was designed to create a multispecies
model that would be amenable to simulation of the kinetics of methylmercury by simply changing the
species-specific parameters. Unlike Farris et al. (1993), separate red blood cell and plasma
compartments were used to predict changes in kinetics of methylmercury across species due to
differences in the red blood cell/plasma ratio. Different pharmacokinetic parameters, such as
tissue/blood partition coefficients and volume distributions for humans, rats and monkeys, were taken
from different studies published in the current literature. The authors provided a benchmark dose on
results of a battery of neurobehavioral tests in 6-year-old children prenatally exposed to methylmercury
in seafood. Their calculations suggested a NOAEL of 17 ppm Hg in maternal hair for the most sensitive
2-14
-------
neurological event in children. The analysis of the pharmacokinetic model indicated that the fetal brain
concentrations of methylmercury at this NOAEL were on the order of 50 ppb and were associated with
maternal dietary intakes of methylmercury ranging from 0.8 to 2.5 //g/kg-day. These analyses provided
support to the Iraqi data used in the development of the RfD for methylmercury, presented in the risk
assessment chapter (Chapter 6) of this volume.
2.6.4 Discussion
Both simple and complex multi-compartment models have been reported by Hall et al. (1994),
Farris et al. (1993), Smith et al. (1994) and Gearhart et al. (1995). The Hall et al. (1994) paper discussed
a model which employed inorganic mercury data obtained from human studies; however, temporal
patterns of blood mercury and the existence of kinetically distinct plasma pools for mercury present
uncertainties which limit the use of this model in risk assessment. Farris et al. (1993) reported a
multicompartment model using data obtained from rats exposed to methylmercury in diets over a period
of 98 days. They observed a buildup of inorganic mercury in tissues and the conversion of
methylmercury to inorganic mercury could not be accurately predicted by whole-body counting, which
was also subjected to errors from low sensitivity and the inability to compensate for geometric changes
due to redistribution of methylmercury or translocation of inorganic mercury to its target tissues. Smith
and associates (1994) refined this model and presented a multicompartment model using data obtained
from humans given a single i.v. dose of methylmercury. Uncertainties, however, persist in prediction of
methylmercury exposures in food. Since methylmercury causes subtle neurotoxicity in children, this
model may not be predictive of exposure in children. This potential neurotoxicity observed in prenatally
exposed children prompted Gearhart et al. (1995) to develop multicompartment adult and fetal model
using data from rat, monkey and humans. This model was applied to an epidemiology study on which
benchmark dose analysis was used to better characterize the dose-response information rather than the
traditional NOAEL approach. In the risk assessment chapter of this volume, U.S. EPA utilizes a
benchmark dose approach for setting the RfD for methylmercury. A multispecies compartment model
discussed in the Gearhart et al. (1995) report may provide a viable approach because it can use data from
both adults and neonates. This approach can use adult and neonatal effects data from several animal and
human studies to account for evidences of non-linearities in dose-responses. Research is needed to
reduce uncertainties in racial, ethnic, and cultural differences which exist in epidemiological studies.
2-15
-------
3. BIOLOGICAL EFFECTS
This chapter summarizes the available toxicity data on mercury compounds; the information is
tabulated specifically for each form of mercury (i.e., elemental, inorganic and methylmercury) and each
toxicity endpoint. Case studies in humans are distinguished from epidemiological studies and animal
studies. In addition, critical studies for a given endpoint are briefly summarized in the narrative
preceding the corresponding table.
The tables provide information on study design, observed effects, study limitations and any
reported biological monitoring levels (BMLs) of mercury. To the extent possible, BML values have been
reported in consistent units throughout this chapter (|ig/L in body fluids, (ig/g in tissue, (ig/g creatinine in
urine). It was not possible, however, to use completely consistent units because investigators measured
mercury in different media (e.g., blood, urine, or tissue) or used different time frames (e.g., (ig/L urine,
(ig/24 hour urine). In addition, some investigators normalized urine concentrations to the amount of
creatinine, while most did not. An explanation is provided in Appendix A for any dose conversions
required during review and evaluation of the toxicity and carcinogenicity studies reported in the
discussions presented below.
3.1 Elemental Mercury
3.1.1 Critical Noncancer Data
This section describes studies evaluated by U.S. EPA for use in assessing general systemic health
risks, primarily toxicity in exposed workers. Chapter 6 describes the derivation of an inhalation
Reference Concentration (RfC) for elemental mercury based on neurotoxicity observed in several human
occupational studies. For completeness, some of these studies are also presented in tabular form in
succeeding sections.
Fawer et al. (1983) used a sensitive measure of intention tremor (tremors that occur at the
initiation of voluntary movements) in workers occupationally exposed for an average of 26 years to
metallic mercury vapor. A statistically significant difference was seen in the frequency of these tremors
in mercury-exposed workers compared with unexposed workers. The concentration of metallic mercury
in the air was measured, and a time-weighted-average (TWA) of 0.026 mg/m3 over an average of 15.3
years was derived. This was based on the assumption that the workers were exposed to the same
concentration of mercury for the duration of their employment. It should be noted that very little detail
was presented with regard to the measurement of the exposure levels, and that it is likely that there were
variations in the mercury air levels during the period of exposure. Furthermore, the tremors may have
resulted from intermittent exposure to concentrations higher than the TWA.
Piikivi and Tolonen (1989) studied the effects of long-term exposure to mercury vapor on the
electroencephalograms (EEG) of 41 chloralkali workers exposed for a mean of 15.6 ± 8.9 years by
comparison to matched referent controls. They found that the exposed workers, who had mean blood
mercury levels of 12 //g/L and mean urine mercury levels of 20 //g/L, tended to have an increased
number of EEG abnormalities when analyzed by visual inspection only. When the EEGs were analyzed
by computer, the exposed workers had significantly slower and attenuated EEGs as compared to the
referents. These changes were observed in 15% of the exposed workers. The frequency of these changes
correlated with cortical mercury content (measured in other studies); the changes were most prominent in
the occipital cortex, less prominent in the parietal cortex and almost absent in the frontal cortex. The
3-1
-------
authors extrapolated an exposure level associated with these EEG changes of 0.025 mg/m3 from blood
levels based on a conversion factor calculated by Roels et al. (1987).
Piikivi and Hanninen (1989) studied the subjective symptoms and psychological performances
on a computer-administered test battery in 60 chloralkali workers exposed to mercury vapor for a mean
of 13.7 ± 5.5 years as compared to matched referent controls. The exposed workers had mean blood
mercury levels of 10 //g/L and mean urine mercury levels of 17 //g/L. A statistically significant increase
in subjective measures of memory disturbance and sleep disorders was found in the exposed workers.
The exposed workers also reported more anger, fatigue and confusion. No objective disturbances in
perceptual motor, memory or learning abilities were found in the exposed workers. The authors
extrapolated an exposure level associated with these subjective measures of memory disturbance of 0.025
mg/m3 from blood levels based on a conversion factor calculated by Roels et al. (1987).
Both subjective and objective symptoms of autonomic dysfunction were investigated in
41 chloralkali workers exposed to mercury vapor for a mean of 15.6 ± 8.9 years as compared to matched
referent controls (Piikivi 1989). The quantitative non-invasive test battery consisted of measurements of
pulse rate variation in normal and deep breathing in the Valsalva maneuver and in vertical tilt, as well as
blood pressure responses during standing and isometric work. The exposed workers had mean blood
levels of 11.6 //g/L and mean urinary levels of 19.3 //g/L. The exposed workers complained of more
subjective symptoms of autonomic dysfunction than the controls, but the only statistically significant
difference was an increased reporting of palpitations in the exposed workers. The quantitative tests
revealed a slight decrease in pulse rate variations, indicative of autonomic reflex dysfunction, in the
exposed workers. The authors extrapolated an exposure level associated with these subjective and
objective measures of autonomic dysfunction of 0.03 mg/m3 from blood levels based on the conversion
factor calculated by Roels et al. (1987).
Sensory and motor nerve conduction velocities were studied in 18 workers from a mercury cell
chlorine plant (Levine 1982). Time-integrated urine mercury levels were used as an indicator of mercury
exposure. Using linearized regression analysis, the authors found that motor and sensory nerve
conduction velocity changes, (i.e., prolonged distal latencies) were correlated with the time-integrated
urinary mercury levels in asymptomatic exposed workers and occurred when urinary mercury levels
exceeded 25 //g/L. This study demonstrated that elemental mercury exposure can be associated with
preclinical evidence of peripheral neurotoxicity.
Miller et al. (1975) investigated several subclinical parameters of neurological dysfunction in
142 workers exposed to inorganic mercury in either the chloralkali industry or a factory for the
manufacture of magnetic materials. They found that there was a significant increase in average forearm
tremor frequency in workers whose urinary mercury concentration exceeded 50 //g/L as compared to
unexposed controls. Also observed were eyelid fasciculation, hyperactive deep tendon reflexes and
dermatographia, but there was no correlation between the incidence of these findings and urinary
mercury levels.
Roels et al. (1985) examined 131 male and 54 female workers occupationally exposed to
mercury vapor for an average duration of 4.8 years. Urinary mercury (52 and 37 //g/g creatinine for
males and females, respectively) and blood mercury levels (14 and 9 //g/L for males and females,
respectively) were recorded, but atmospheric mercury concentration was not provided. Symptoms
indicative of central nervous system (CNS) disorders were reported but were not related to mercury
exposure. Minor renal tubular effects were detected in mercury-exposed males and females and were
attributed to current exposure intensity rather than exposure duration. Male subjects with urinary
3-2
-------
mercury levels of >50 //g/g creatinine exhibited preclinical signs of hand tremor. It was noted that
females did not exhibit this effect, and that their urinary mercury never reached the level of 50 //g/g
creatinine. A companion study (Roels et al. 1987) related air mercury levels to blood mercury
(Hg-blood) and urinary mercury (Hg-U) values in 10 workers in a chloralkali battery plant. Duration of
exposure was not specified. A high correlation was reported for Hg-air and Hg-U for pre-shift exposure
(r=0.70, p<0.001) and post-shift (r=0.81, p<0.001) measurements. Based on these data and the results of
their earlier (1985) study, the investigators suggested that some mercury-induced effects may occur when
Hg-U levels exceed 50 //g/g creatinine and that this value corresponds to a mercury TWA of =0.04 mg
Hg/m3.
A survey of 567 workers at 21 chloralkali plants was conducted to ascertain the effects of
mercury vapor inhalation (Smith et al. 1970). Mercury levels ranged from <0.01 to 0.27 mg/m3, and
chlorine concentrations ranged from 0.1 to 0.3 ppm in most of the working stations of these plants.
Worker exposure to mercury levels (TWA) varied with 10.2% of the workers exposed to <0.01 mg
Hg/m3, 48.7% exposed to 0.01-0.05 mg Hg/m3, 25.6% exposed to 0.06-0.10 mg Hg/m3, and 4.8%
exposed to 0.24-0.27 mg Hg/m3 (approximately 85% were exposed to mercury levels <0.1 mg/m3). The
duration of employment for the examined workers ranged from one year (13.3%) to >10 years (31%)
with 55.7% of the workers employed for 2 or 9 years. A group of 600 workers not exposed to chlorine
served as a control group. A strong positive correlation (p<0.001) was found between the mercury
TWAs and the reporting of subjective neuropsychiatric symptoms (nervousness, insomnia), occurrence
of objective tremors and weight and appetite loss. A positive correlation (p<0.001) was also found
between mercury exposure levels and urinary and blood mercury levels of test subjects. No adverse
alterations in cardiorespiratory, gastrointestinal, renal, or hepatic functions were attributed to the mercury
vapor exposure. Additionally, biochemical (hematologic data, enzyme activities) and clinical
measurements (electrocardiogram, chest X-rays) were not different between the mercury-exposed and
non-exposed workers. No significant signs or symptoms were noted for individuals exposed to mercury
vapor concentrations <0.1 mg Hg/m3. This study provides data indicative of a no-observed-adverse-
effect level (NOAEL) of 0.1 mg Hg/m3 and a lowest-observed-adverse-effect level (LOAEL) of 0.18 mg
Hg/m3. In a follow-up study conducted by Bunn et al. (1986), however, no significant differences in the
frequency of objective or subjective findings such as weight loss and appetite loss were seen in workers
exposed to mercury at levels that ranged between 50 and 100 mg/m3. The study by Bunn et al. (1986)
was limited, in that little information was provided regarding several methodological questions such as
quality assurance measures and control of possible confounding variables.
Neurological signs and symptoms (i.e., tremors) were observed in 79 workers exposed to metallic
mercury vapor; urinary mercury levels in affected subjects exceeded 500 //g/L. Short-term memory
deficits were seen in workers whose urine levels were less than 500 //g/L (Langolf et al. 1978). Impaired
performance in mechanical and visual memory tasks and psychomotor ability tests was reported by Forzi
et al. (1978) in exposed workers whose urinary mercury levels exceeded 100 //g/L.
Decreased strength, decreased coordination, increased tremor, decreased sensation and increased
prevalence of Babinski and snout reflexes were exhibited by 247 exposed workers whose urinary
mercury levels exceeded 600 //g/L. Evidence of clinical neuropathy was observed at urinary mercury
levels that exceeded 850 //g/L (Albers et al. 1988). Preclinical psychomotor dysfunction was reported to
occur at a higher incidence in 43 exposed workers (mean exposure duration of 5 years) whose mean
urinary excretion of mercury was 50 //g/L. In the same study, workers whose mean urinary mercury
excretion was 71 //g/L had a higher incidence of total proteinuria and albuminuria (Roels et al. 1982).
Postural and intention tremor was observed in 54 exposed workers (mean exposure duration of 1.1 years)
whose mean urinary excretion of mercury was 63 //g/L (Roels et al. 1989). Verbeck et al. (1986)
3-3
-------
observed an increase in tremor parameters with increasing urinary excretion of mercury in 21 workers
exposed to mercury vapor for 0.5-19 years. The LOAEL for this effect was a mean urinary excretion of
35 //g/g creatinine.
The elemental mercury levels reported to be associated with preclinical and symptomatic
neurological dysfunction are generally lower than those found to affect kidney function, as discussed
below.
Piikivi and Ruokonen (1989) found no evidence of glomerular or tubular damage in
60 chloralkali workers exposed to mercury vapor for an average of 13.7 ± 5.5 years as compared to their
matched referent controls. Renal function was assessed by measuring urinary albumin and N-acetyl-p-
glucosaminidase (NAG) activity. The mean blood mercury level in the exposed workers was 14 /ug/L,
and the mean urinary mercury level was 17 /ug/L. The authors extrapolated a NOAEL for kidney effects
based on these results of 0.025 mg/m3 from blood levels based on the conversion factor calculated by
Roelsetal. (1987).
Stewart et al. (1977) studied urinary protein excretion in 21 laboratory workers exposed to
0.01-0.05 mg/m3 of mercury. Their urinary level of mercury was ~35 /ug/L. Increased proteinuria was
found in the exposed workers by comparison to unexposed controls. When preventive measures were
instituted to limit exposure to mercury, proteinuria was no longer observed in the exposed technicians.
Lauwerys et al. (1983) found no change in several indices of renal function (e.g., proteinuria,
albuminuria, urinary excretion of retinol-binding protein, aminoaciduria, creatinine in serum,
P2-microglobulin in serum) in 62 workers exposed to mercury vapor for an average of 5.5 years. The
mean urinary mercury excretion in the exposed workers was 56 /ug/g creatinine, which corresponds to an
exposure level of -0.046 mg/m3 according to a conversion factor of 1:1.22 (airurine [//g/g creatinine])
(Roels et al. 1987). Despite the lack of renal effects observed, 8 workers were found to have an increase
in serum anti-laminin antibodies, which can be indicative of immunological effects. In a follow-up study
conducted by Bernard et al. (1987), however, there was no evidence of increased serum anti-laminin
antibodies in 58 workers exposed to mercury vapor for an average of 7.9 years. These workers had a
mean urinary mercury excretion of 72 /ug/g creatinine, which corresponds to an exposure level of -0.059
mg/m3.
Renal function in 100 chloralkali workers exposed to inorganic mercury vapor for an average of
8 years was studied (Stonard et al. 1983). No changes in the following urinary parameters of renal
function were observed at mean urinary mercury excretion rates of 67 /ug/g creatinine: total protein,
albumin, aracid glycoprotein, p2-microglobulin, NAG and y-glutamyl transferase. When urinary
mercury excretion exceeded 100 /ug/g creatinine, a small increase in the prevalence of higher activities of
NAG and y-glutamyl transferase were observed.
Rosenman et al. (1986) evaluated routine clinical parameters (physical exams, blood chemistry,
urinalysis), neuropsychological disorders, urinary NAG, motor nerve conduction velocities and
occurrence of lenticular opacities in 42 workers of a chemical plant producing mercury compounds. A
positive correlation (p<0.05 to p<0.001) was noted between urinary mercury (levels ranged from
100-250 /ug/L) and the number of neuropsychological symptoms, NAG excretions and decrease in motor
nerve conduction velocities. Evidence of renal dysfunction was seen in 63 chloralkali workers. This
included increased plasma and urinary concentrations of p-galactosidase, increased urinary excretion of
high-molecular weight proteins and a slightly increased plasma p2-microglobulin concentration. The
3-4
-------
incidence of these effects increased in workers whose urinary mercury excretion exceeded 50 //g/g
creatinine (Buchet et al. 1980).
Increased urinary NAG levels were found in workers whose urinary mercury levels exceeded
50 //g/L (Langworth et al. 1987). An increase in the concentration of urinary brush border proteins (BB-
50) was observed in 20 workers whose mean urinary mercury excretion exceeded 50 /ug/g creatinine
(Mutti et al. 1985). Foa et al. (1976) found that 15 out of 81 chloralkali workers exposed to 0.06-0.30
mg/m3 mercury exhibited proteinuria. An increased excretion of p-glutamyl transpeptidase, indicative of
renal dysfunction, was found in 509 infants dermally exposed to phenylmercury via contaminated diapers
(Gotelli et al. 1985).
The elemental mercury levels reported to be associated with preclinical and symptomatic
neurological dysfunction and kidney effects are lower than those found to affect pulmonary function, as
discussed below.
McFarland and Reigel (1978) described the cases of six workers who were acutely exposed (4-8
hours) to calculated metallic mercury vapor levels of 1.1-44 mg/m3. These men exhibited a combination
of chest pains, dyspnea, cough, hemoptysis, impairment of pulmonary function (reduced vital capacity),
diffuse pulmonary infiltrates and evidence of interstitial pneumonitis. Although the respiratory
symptoms resolved, all six cases exhibited chronic neurological dysfunction, presumably as a result of
the acute, high-level exposure to mercury vapor.
Lilis et al. (1985) described the case of a 31-year-old male who was acutely exposed to high
levels of mercury vapor in a gold extracting facility. Upon admission to the hospital, the patient
exhibited dyspnea, chest pain with deep inspiration, irregular infiltrates in the lungs and reduced
pulmonary function (forced vital capacity). The level of mercury to which he was exposed is not known,
but a 24-hour urine collection contained 1,900 //g Hg/L. Although the patient improved gradually over
the next several days, he still showed signs of pulmonary function abnormalities (e.g., restriction and
diffusion impairment) 11 months after exposure.
Levin et al. (1988) described four cases of acute high-level mercury exposure during gold ore
purification. The respiratory symptoms observed in these four cases ranged from minimal shortness of
breath and cough to severe hypoxemia. The most severely affected patient exhibited mild interstitial lung
disease both radiographically and on pulmonary function testing. One patient had a urinary mercury
level of 245 /ug/L upon hospital admission. The occurrence of long-term respiratory effects in these
patients could not be evaluated since all but one refused follow-up treatment.
Schweinsberg (1994) evaluated mercury in hair, blood and urine of subjects who had amalgam
fillings, who consumed fish or who had occupational exposure. The first group consisted of 67 males
aged 16-72 yrs, mean age 36. The fish-eating population was 149 males and females (age 22-80, mean =
47) who either did or did not eat fish from the Rhine River. The workers were 105 male and female
employees of a Thuringian thermometer plant (ages 19- 65, mean = 42). Both fish consumption and the
presence of mercury amalgam fillings resulted in measurable mercury in blood or urine. The range of
mercury in the workers was higher by about 100 fold. The authors present most of the data graphically;
they report that persons without amalgam fillings who did not consume Rhine fish or work in the
thermometer factory had blood mercury in the range of 0.2 (ig/1 (detection limit) and 0.4 (ig/1 (n not
given). It appears that from Figure 5 of the paper that blood mercury levels ranged between 100-120
(ig/1 in 5 workers. The authors state that no changes in nerve conduction velocities or N-acetyl-p-D-
glucosaminidase activities were observed (data not shown in paper).
3-5
-------
3.1.2 Cancer Data
3.1.2.1 Human Data
A number of epidemiological studies were conducted that examined mortality among elemental
mercury vapor-exposed workers. Conflicting data regarding a correlation between mercury exposure and
an increased incidence of cancer mortalities have been obtained. All of the studies have limitations
which compromise interpretation of their results. These studies are summarized in Table 3-1.
A retrospective cohort study examined mortality among 5,663 white males who worked between
1953 and 1958 at a plant in Oak Ridge, Tennessee, where elemental mercury was used for lithium isotope
separation (Cragle et al. 1984). The workers were divided into three cohorts: exposed workers who had
been monitored on a quarterly basis for mercury levels in urine (n=2,133); workers exposed in the
mercury process section for whom urinalysis monitoring data were not collected (n=270); and unexposed
workers from other sections of the nuclear weapons production facility (n=3,260). The study subjects
worked at least 4 months during 1953-1958 (a period when mercury exposures were likely to be high);
mortality data from death certificates were followed through the end of 1978. The mean age of the men
at first employment at the facility was 33 years, and average length of their employment was greater than
16 years with a mean 3.73 years of estimated mercury exposure. Air mercury levels were monitored
beginning in 1955, and during 1955 through the third quarter of 1956, air mercury levels were above 100
(ig/m3 in 30-80% of the samples. Thereafter, air mercury levels decreased to concentrations below 100
(ig/m3. The mortality experience [standard mortality ratio (SMR)] of each group was compared with the
age-adjusted mortality experience of the U.S. white male population. Among exposed and monitored
workers, there were no significant increases in mortality from cancer at any site, even after the level or
length of exposure was considered. A significantly lower mortality from all causes was observed. There
was an excess of deaths due to lung cancer in the exposed, monitored workers (42 observed, 31.36
expected) but also in the unexposed workers (71 observed and 52.93 expected). The SMR for each group
was 1.34; the elevated incidence of lung cancer deaths was, therefore, attributed to some other factor at
the plant and/or to lifestyle factors (e.g., smoking) common to both the exposed and unexposed groups.
Study limitations include small cell sizes for cancer mortality, which limited the statistical stability of
many comparisons.
Barregard et al. (1990) studied mortality and cancer morbidity between 1958 and 1984 in 1,190
workers from eight Swedish chloralkali plants that used the mercury cell process in the production of
chlorine. The men included in the study had been monitored for urinary or blood mercury for more than
one year between 1946 and 1984. Vital status and cause of death were ascertained from the National
Population Register and the National Bureau of Statistics. The cancer incidence of the cohort was
obtained from the Swedish Cancer Register. The observed total mortality and cancer incidences were
compared with those of the general Swedish male population. Comparisons were not made between
exposed and unexposed workers. Mean urinary mercury levels indicated a decrease in exposure between
the 1950s and 1970s; the mean urinary mercury level was 200 (ig/L during the 1950s, 150 (ig/L during
the 1960s and 50 (ig/L in the 1970s. Mortality from all causes was not significantly increased in exposed
workers. A significant increase in deaths from lung tumors with greater than 10 years of latency was
seen in exposed workers (rate ratio, 2.0; 95% C.I. 1.0-3.8), but 9 of the 10 observed cases of lung cancer
occurred among workers (457 of the 1,190) possibly exposed to asbestos as well as to mercury. No dose
response was observed with respect to mercury exposure and lung tumors. This study is limited because
no quantitation was provided on smoking status, and results were confounded by exposure to asbestos.
3-6
-------
Table 3-1
Carcinogenic Effects of Elemental Mercury in Humans: Epidemiological Studies
Species/
No. per Sex
Human/
2,133 M
Human/376
cases of lung
cancer (6 were
hatmakers), 892
controls
Human/
cohort/3454 M,
5787 F
Human/9,912 M
(369 silicotics,
9,543
nonsilicotics)
Human/
cohort/799 M
Human/1 190 M
Exposure
Duration
>4 months
(occup)
NS
(occup)
NS
(occup)
NS
(occup)
> 1 year
(occup)
at least one
year
Dose
(mg/m3)
NS, but up
to 80% of
air samples
in early
years were
>0. 10; this
declined to
1-10% in
later years
NS
NS
NS
NS
NS
grouped by
years x U-
Hg;
>1000
ug/L,
1000-2000
ug/L,
>2000 ug/L
Effects/Limitations/BML
No biologically significant increase in cancer mortality in workers
at an isotope enrichment plant, compared with unexposed workers
at the same plant, or with age-adjusted mortality of U.S. males.
Lung cancer mortality was increased in exposed workers, but the
increase was not statistically significant and a significant increase
in lung cancer mortality was observed in unexposed workers.
Limitations: small cell size for cancer mortality, limiting
statistical power of comparisons
BML not reported
Increased lung cancer incidence among female hat makers
(p = 0.01). Controls were matched by age, sex and smoking
history.
Limitations: Hat makers were also exposed to arsenic
BML not reported
Increased incidence of glioblastomas among dental professionals.
95% C.I. = 1.3-3.4. Expected incidence was based on all
employed people (Sweden), stratified by age, sex and county.
Limitations: No information was provided on the duration or level
of exposure; subjects were also exposed to chloroform and X-rays.
BML not reported
Increased lung cancer mortality among metal miners (95%
C.I. = 0.94-2.90 for silicotics, 0.98-1.42 for nonsilicotics).
Limitations: Miners were exposed to a variety of metals. Only
274 worked in mercury mines, and data were not reported
separately for this group. Workers were also exposed to radon,
increase may have been related to silicosis
BML not reported
Increased incidence of lung cancer among chloralkali workers, but
there was no association with cumulative mercury dose, years of
employment or latency (95% C.I. = 1.0-2.6). The increase could
be partly explained by an assumed higher smoking incidence and
asbestos exposure. Cancer mortality and incidence compared with
age-adjusted Norwegian population.
Limitations: Subjects were also exposed to chlorine and low levels
of asbestos dust. Limited data available, since reported as an
abstract.
BML not reported
Cancer and mortality rates compared with general population; no
increase in mortality; excess of lung cancers (rate ratio = 2.0; 95%
CI 1.0-3. 8)
Limitations: co-exposure to asbestos
Reference
Cragleetal. 1984
Buiattietal. 1985
Ahlbom et al. 1986
Amandus and
Costello 1991
Ellingsen et al. 1992
Barregard et al. 1990
3-7
-------
Ahlbom et al. (1986) examined the cancer mortality during 1961 to 1979 of cohorts of Swedish
dentists and dental nurses aged 20-64 and employed in 1960 (3,454 male dentists, 1,125 female dentists,
4,662 female dental nurses). Observed incidences were compared with those expected based on cancer
incidence during 1961-1979 among all Swedes employed during 1960 and the proportion of all Swedes
employed as dentists and dental nurses. Data were stratified by sex, age (5-year age groups), and county.
The incidence of glioblastomas among the dentists and dental nurses combined was significantly
increased (SMR, 2.1; 95% C.I. 1.3-3.4); the individual groups had elevated SMRs (2.0-2.5), but the 95%
confidence intervals of these groups included unity. By contrast, physicians and nurses had SMRs of
only 1.3 and 1.2, respectively. Exposure to mercury could not be established as the causative factor
because exposure to other chemicals and X-rays was not ruled out.
Amandus and Costello (1991) examined the association between silicosis and lung cancer
mortality between 1959 and 1975 in white male metal miners (n=9,912) employed in the United States
between 1959 and 1961. Mercury exposures were not monitored. Exposures to specific metals among
the silicotic and nonsilicotic groups were analyzed separately. Lung cancer mortality in both silicotic
and nonsilicotic groups was compared with rates in white males in the U.S. population. Both silicotic
(n=l 1) and nonsilicotic mercury miners (n=263) had significantly increased lung cancer mortality (SMR,
14.03, 95% C.I., 2.89-40.99 for silicotics; SMR, 2.66, 95% C.I. 1.15-5.24 for nonsilicotics). The
analysis did not focus on mercury miners, and confounders such as smoking and radon exposure were not
analyzed with respect to mercury exposure. This study is also limited by the small sample size for
mercury miners.
A case-control study of persons admitted to a hospital in Florence, Italy with lung cancer
between 1981-1983 was performed to evaluate occupational risk factors (Buiatti et al. 1985). Cases were
matched with one or two controls (persons admitted to the hospital with diagnoses other than lung cancer
or suicide) with respect to sex, age, date of admission, and smoking status. Women who had "ever
worked" as hat makers had a significantly increased risk of lung cancer (p=0.01; determined using the
Mantel-Haenszel Chi-square test). The duration of employment as a hat maker averaged 22.2 years, and
latency averaged 47.8 years. Workers in the Italian hat industry were known to be occupationally
exposed to mercury; however, the design of this study did not allow evaluation of the relationship
between cumulative exposure and cancer incidence. In addition, interpretation of the results of this study
is limited by the small sample size (only 6/376 cases reported this occupation) and by exposure of hat
makers to other pollutants including arsenic, a known lung carcinogen.
Ellingsen et al. (1992) examined the total mortality and cancer incidence among 799 workers
employed for more than 1 year in two Norwegian chloralkali plants. Mortality incidence between 1953
and 1988 and cancer incidence between 1953 and 1989 were examined. Mortality and cancer incidence
were compared with that of the age-adjusted general male Norwegian population. No increase in total
cancer incidence was reported, but lung cancer was significantly elevated in the workers (ratio, 1.66;
95% C.I. 1.0-2.6). No causal relationship can be drawn between mercury exposure and lung cancer
because no correlation existed between cumulative mercury dose, years of employment, or latency time.
Also, the prevalence of smoking was 10-20% higher in the exposed workers and many workers were also
exposed to asbestos.
3.1.2.2 Animal Data
Druckrey et al. (1957) injected 0.1 mL of metallic mercury intraperitoneally into 39 rats (males
and females; numbers of each not specified) of the BD III and BD IV strains. Among the rats surviving
longer than 22 months, 5 out of 12 developed peritoneal sarcomas (three females and two males). All
3-8
-------
sarcomas were observed to have droplets of mercury present. Although severe kidney damage was
reported in all treated animals, there were no renal tumors or tumors at any site other than the peritoneal
cavity.
3.1.3 Other Data
3.1.3.1 Death
Accidental exposure to high concentrations of elemental mercury vapor for short amounts of time
has led to deaths in humans (Table 3-2). The cause of death in all available reports was respiratory
failure. The onset of death occurred six hours to 23 days after exposure to mercury vapors (Campbell
1948; Kanluen and Gottlieb 1991; Rowens et al. 1991; Soni et al. 1992; Taueg et al. 1992). Urinary
mercury concentrations indicated that body levels were up to 10 times higher than controls. Only acute-
duration studies were found that directly linked elemental mercury vapor exposure to death.
Table 3-2
Lethality of Elemental Mercury in Humans: Case Studies
Species/
No. per Sex
Human/ 1 F
(4-month old)
Human/2 M,
2 F (adult,
2 elderly)
Human/ 1 F
(1-yrold)
Human/2 M,
2 F (adults)
Human/2
F (children)
Exposure
Duration
5hr
=24hr
<6hr
NS
(Acute)
Several
months
Dose
(mg/m3)
NS
NS
NS
<0.91 at 11-
lS days post-
exposure
0.01-0.04
several
months after
initial spill
Effects/Limitations/BML
Increased creatinine excretion; necrotic stomach mucosa;
degeneration of convoluted tubules; death due to pulmonary
edema
Limitation: Limited exposure data
BML not reported
Respiratory distress; CNS alterations; nausea; tubular
necrosis of proximal tubules in kidneys
Limitation: Limited exposure data
BML Range: 4.6-219 ug/L in urine
Breathing difficulty; distended abdomen
Limitation: Limited exposure data
BML not reported
Dyspnea; respiratory failure; death at 1 1-24 days
postexposure
BML Range: 94-423 ug/L in urine
Numbness in fingers and toes; absence of deep tendon
reflexes; visual field defects; weakness
BML not reported
Reference
Campbell 1948
Kanluen and
Gottlieb 1991;
Rowens etal. 1991
Sonietal. 1992
Taueg etal. 1992
Taueg etal. 1992
Animal studies reveal that pulmonary edema and asphyxiation result from acute high-
concentration exposure to elemental mercury vapors (Table 3-3). Exposure to elemental mercury vapors
for two hours at a concentration of 27 mg Hg/m3 resulted in death of 20 of 32 rats (Livardjani et al.
1991). Rabbits exposed for 1 to 30 hours to 28.8 mg Hg/m3 of elemental mercury vapors appeared to be
3-9
-------
less affected. Death occurred in only one of two rabbits exposed for 30 hours (Ashe et al. 1953).
Exposure to the same concentration for a shorter duration resulted in no deaths.
Table 3-3
Lethality of Elemental Mercury in Animals: Inhalation Exposure
Species/
Strain/
No. per Sex
per Group
Rabbit/strain
NS/14 (sex
NS)
Rat/Wistar/64
M/duration
Exposure
Duration
1-30 hr
1 or 2 hr
Dose
(mg/m3)
28.8
0,30
Effects/Limitations/BML
LD50 for 30 hours; all rabbits exposed for shorter periods
survived.
Limitations: There was no control group, and details on effects
were lacking.
BML: 5,320 ug/L in blood
Death was due to asphyxiation; pulmonary edema and fibrosis
were observed. No animals exposed for 1 hour died by 15 days,
and all animals exposed for 2 hours died within 5 days.
Limitations: No control group; limited reporting of histology
BML Range: 391-4,558 ug/L in blood at 1-15 days
postexposure
Reference
Asheetal. 1953
Livardjani etal.
1991
3.1.3.2 Neurological
Case reports from accidental exposures to high concentrations of mercury vapors (Adams et al.
1983; Aronow et al. 1990; Barber 1978; Bluhm et al. 1992a; Fagala and Wigg 1992; Foulds et al. 1987;
Friberg et al. 1953; Hallee 1969; Jaffe et al. 1983; Karpathios et al. 1991; Lilis et al. 1985; McFarland
and Reigel 1978; Sexton et al. 1976; Snodgrass et al. 1981; Taueg et al. 1992) as well as studies of
populations chronically exposed to potentially high concentrations (Ehrenberg et al. 1991; Friberg et al.
1953; Roels et al. 1982; Sexton et al. 1978) have provided considerable information about the
neurotoxicity of elemental mercury vapor. These studies have shown effects on a wide variety of
cognitive, sensory, personality and motor functions.
Occasionally, hearing loss, visual disturbances (visual field constriction), and/or hallucinations
have also occurred. In general, symptoms have been observed to subside after removal from exposure.
However, persistent effects (tremor, cognitive deficits) have been observed in occupationally exposed
subjects 10 to 20 years after cessation of exposure (Albers et al. 1988; Ellingsen et al. 1993; Kishi et al.
1993).
3-10
-------
Symptoms of Mercury Vapor-induced Neurotoxicity
The most prominent symptoms associated with mercury vapor-induced neurotoxicity include the following:
• tremors ~ initially affecting the hands and sometimes spreading to other parts of the body
• emotional lability ~ often referred to as "erethism" and characterized by irritability, excessive
shyness, confidence loss and nervousness
• insomnia
• neuromuscular changes ~ weakness, muscle atrophy, muscle twitching
• headaches
• polyneuropathy ~ paresthesias, stocking-glove sensory loss, hyperactive tendon reflexes, slowed
sensory and motor nerve conduction velocities
• memory loss and performance deficits in tests of cognitive function
Studies of workers exposed to elemental mercury vapor have reported frank neurotoxicity at
exposure levels greater than 0.1 mg/m3 (Smith et al. 1970) or at levels resulting in urinary mercury of
greater than 300 (ig in a 24-hour urine sample (Bidstrup et al. 1951). Several studies, however, have
shown evidence of neurotoxicity at approximately 2- to 4-fold lower concentrations. Self-reported
memory disturbances, sleep disorders, anger, fatigue, confusion and/or hand tremors were increased in
workers chronically exposed to an estimated 0.025 mg/m3 (blood levels of approximately 10 (ig/L)
(Langworth et al. 1992a; Piikivi and Hanninen 1989). Also, objective measures of cognitive and/or
motor function in exposed populations have shown significant differences from unexposed controls
(Ehrenberg et al. 1991; Fawer et al. 1983; Liang et al. 1993; Ngim et al. 1992; Piikivi and Tolonen 1989;
Piikivi et al. 1984; Roels et al. 1982, 1989).
Table 3-4
Neurotoxicity of Elemental Mercury in Humans: Case Studies
Species/
No. per Sex
Human/ 1 M
(adult)
Human/6 M
Human/5 M, 6
F (adults and
children)/
12 controls
(sexNS)
Human/2 M, 2
F (adults)
Exposure
Duration
8-9 mo
(occup)
<8hr
51-176 d
3d
Dose
(mg/m3)
0.02-0.45
44.3
(est.)
0.1-1.0
NS
Effects/Limitations/BML
Fatigue, irritability in an electrochemical industry worker
Limitations: small sample size; concomitant exposure to
chlorine; limited data reporting
BML: 680-900 ug/L in urine
Tremor; irritability; visual and hearing abnormalities
Limitations: small sample size; limited data reporting
BML Range: 1,060-3,280 ug/24 hr urine
Nervousness, insomnia and inattentiveness were more
common than in controls; altered EEGs and personality
changes also noted
Limitations: small sample size
BML: 1 83-620 ug/L in blood at first measure
Headache, slowed speech
Limitation: Small sample size; limited exposure data
BML: 82-5700 ua/24hr urine
Reference
Fribergetal. 1953
McFarland and
Reigel 1978
Sexton etal. 1978
Snodgrass et al. 1981
3-11
-------
Table 3-4 (continued)
Neurotoxicity of Elemental Mercury in Humans: Case Studies
Species/
No. per Sex
Human/ 1 M
(adult)
Human/1 F (8-
month old)
Human/ 1 M
Human/ 1 F
(child)
Human/ 1 M
(child)
Human/17-26
M
Human/ 1 F
(child)
Human/2 F
(children)
Exposure
Duration
2d
= ld
-2hr
2 mo
2wk
<16hr
6 mo
Several months
Dose
(mg/m3)
NS
NS
NS
NS
NS
NS
NS
0.01-0.04
several
months after
initial spill
Effects/Limitations/BML
Delayed neurotoxicity: paresthesias; muscle
fasciculations; hyperactive deep tendon reflexes
Limitation: small sample size; exposure data limited
BML: 98.75 ug/L in urine 3.5 months after exposure
Seizures; weakness; short-term hearing deficit; cortical
atrophy
Limitations: Exposure data limited
BML Range: 16-43 ug/24hr urine
Dizziness, weakness
Limitation: small sample size; limited exposure data
BML: 1,900 ug/L urine on first day
Lethargy; irritability
Limitations: small sample size; limited reporting of
symptoms; limited exposure data
BML: 214 ug/L in 24 hr urine
Tremor; sleep disturbance; anxiety; cold hands and feet
Limitation: small sample size; limited exposure data
BML: 130 ug/24hr urine
Fatigue, headaches, irritability, depression, anxiety,
tremor, impaired performance on visual-motor tests
(p<0.05) reported in welders following accidental
exposure.
Limitation: Chronic exposure to other metals; exposure
data limited
BML: -60 ug/L in blood at 20 d postexposure
Peripheral neuropathy; erethism; dizziness; depression;
irritability
Limitation: small sample size; exposure data limited
BML: 686 ug/24 hr urine
Numbness in fingers and toes; absence of deep tendon
reflexes; visual field defects; weakness
BML not reported
Reference
Adams etal. 1983
Jaffeetal. 1983
Lilisetal. 1985
Fouldsetal. 1987
Karpathios et al.
1991
Bluhmetal. 1992a
Fagala and Wigg
1992
Taueg et al.
1992
3-12
-------
Table 3-5
Neurotoxicity of Elemental Mercury in Humans: Epidemiological Studies
Species/
No. per Sex
Human/27
cases (sex NS)
Human/3 M, 6
F exposed/10
M, 30 F
controls
Human/43
exposed/47
controls (sex
NS)
Human/23 M
exposed/22 M
control
Human/ 12
exposed/12
controls (sex
NS)
Human/26
exposed M/25
control M
Human/36 M
exposed/36
controls
Exposure
Duration
3 mo-39 yr
(occup)
NS
(occup)
>6 months
Mean: 5.3yr
(occup)
NS (occup)
3 mo-8 yr
(occup)
Avg: 15.3yr
(occup)
Avg: 16.9yr
(10-37 yr)
(occup)
Dose
(mg/m3)
0-1.67
(est.)
NS
NS
NS
NS
0.026
(TWA)
personal
monitoring
0.022-
0.028 (est.)'
Effects/Limitations/BML
161 electric meter repair workers were examined, and 22
were found to be symptomatic; there were 5 index cases.
Tremor; irritability; visual impairment were observed
Limitation: Concomitant exposure to other chemicals is
likely.
BML Range: 1,495-7,950 ug/24 hr urine
Neuropsychological tests showed irritability, tremor, memory
loss, poor coordination, visual impairment; altered
electrophysiology (p<0.05) in thermometer manufacturing
employees
Limitation: Exposure data limited
BML Range: 4-1,101 ug/24 hr urine
Objectively assessed tremor and eye-hand coordination
tended to be higher in the exposed group, with a significant
(p<0.05) difference on one test. There was a tendency toward
a dose-response, but it was not statistically significant.
Exposed group worked in amalgam or chloralkali plants;
control workers were matched from the same plants, but
unexposed.
Limitation: Exposure data limited
BML: 29.2 ug/L in blood (range: 5.3-135); 95.5 ug/g
creatinine in urine (range 9.9-286)
Decreased nerve conduction velocity (p<0.05); visual
impairment (p<0.01); higher distress levels
Of a sample of 298 dentists, the exposed group had "tissue"
mercury levels in the top 20%; the controls were age-
matched, with no detectable tissue mercury. Tissue mercury
in the head and wrist was measured using x-ray fluorescence.
Limitation: Exposure data limited
BML: >20 ug/g in tissue
In a battery of objective tests, the following findings were
significant: tremor (p<0.025); decreased verbal intelligence;
short- and long-term memory impairment (p<0.01); fatigue
(p<0.01).
The exposed group worked with amalgam (n=4) or were
exposed to mercuric chloride, methoxyethyl mercuric
chloride, methoxyethylmercuric acetate (n=8). Controls were
matched by age, sex, education, ethnic background.
Limitation: Exposure data limited; small sample size
BML Range: <10-670 ug/L urine
Objectively assessed tremor was significantly (p = 0.001)
elevated in the exposed group and correlated with exposure
duration. Exposed group worked in fluorescent tube factories
(n=7), chloralkali plants (n=12), or in acetaldehyde
production. The control subjects worked at the same
factories but had not been exposed to mercury.
Mean BML: 8,280 ug/L in blood; 20 ug/g creatinine in urine
By comparison to age-matched controls, chloralkali workers
had memory impairment, decreased verbal intelligence
(p<0.01).
BML: >15ug/L in blood; >56 ug/L in urine
Reference
Bidstrup et al. 1951
Vroom and Greer
1972
Roelsetal. 1982
Shapiro etal. 1982
Williamson et al.
1982
Faweretal. 1983
Piikivietal. 1984
3-13
-------
Table 3-5 (continued)
Neurotoxicity of Elemental Mercury in Humans: Epidemiological Studies
Species/
No. per Sex
Human/60 M
exposed/60 M
controls
Human/41 M
exposed/41 M
controls
Human/54 M
exposed/48
controls
Human/10 M,
62 F exposed/9
M, 60 F control
Human/89
exposed/75
controls
Human/60 M,
38 F
exposed/27 M,
27 F controls
Human/77 M
exposed/53 M
controls
Human/1 17 M
exposed/76
controls
Exposure
Duration
Avg: 13.7yr
(5-28 yr)
(occup)
5-27 yr
(occup)
Avg: 7.7 yr
(1-20 y
(occup)
Avg: 5 yr
(occup)
>lyr
(occup)
lOhr/d
6d/wk
0.7-24 yr
(occup)
>lyr
Avg: 7.9 yr
(occup)
389 min/d
duration NS
(occup)
Dose
(mg/m3)
0.025 (est.)
0.025 (est.)
NS
0.076 (avg)
0.003-0.27
(range)
0.025
0.014
(TWA)
0.059
1.5-3.3
Effects/Limitations/BML
In a psychological and psychomotor test battery, there were
statistically significant differences in subjective tests
(memory disturbance, mood; p<0.01) and an objective test
(hand-eye coordination, p<0.001). Subjects were chlorine-
alkali workers and controls were age-matched.
BML: 10.4 ug/L avg. in blood; 17.9 ug/g creatinine avg. in
urine
Attenuation of power density spectrum of EEG in chloralkali
workers (p<0.01); controls were age-matched;slight increase
in subjective symptoms of autonomic (cardiovascular)
dysfunction and a slight decrease in pulse rate variations
(cardiovascular reflex response).
BML: 67.8 ug/L in blood; 20.6 ug/g creatinine in urine
Chloralkali and amalgam workers had impaired eye-hand
coordination (p<0.001) and hand steadiness (p<0.02) in
objective tests, by comparison to unexposed matched controls
from the same plants.
Limitation: Exposure data limited
Geometric mean BML: 24 ug/L blood; 63 ug/g creatinine
Neurological exam found difficulty with heel-to-toe gait
(p<0.05) in thermometer manufacturers; control population
worked at a nearby electronics manufacturer.
Avg. BML: 73.2 ug/g creatinine in urine
Increased tiredness, memory disturbance, based on interviews
(p<0.001), but no effect on psychometric tests or tremor in
chloralkali workers.
BML: 1 1 ug/L in blood; 25.4 ug/g creatinine in urine
Impaired performance on neurobehavioral tests in dentists
(p<0.05); severity of effect correlated with exposure.
Limitations: Concomitant exposure to physical and vibration
load (affecting dexterity tests); confounding exposure to folk
medicine, BML: Mean 9.8 ug/L in blood; range: 0.63-57.3
ug/L in blood
In a study of ex-chloralkali workers (avg. 12.3 yr since last
exposure) compared with age-matched controls, sensory
nerve conduction velocity and visual evoked response
correlated with mercury exposure (p<0.05)
BML: 3,190 ug/g creatinine in urine current, 106 ug/L in
urine during exposure
Ex-mercury miners tested 18 years after the closure of the
mine had lower scores on objective neuropsychological tests
(motor coordination, reaction time, short-term memory) than
age- and education-matched controls (p<0.01). 76 of the
miners had a history of mercury poisoning, but subjective
symptoms had generally decreased since exposure
BML not reported
Reference
Piikivi and
Hanninen 1989
Piikivi and Tolonen
1989
Roelsetal. 1989
Ehrenbergetal.
1991
Lang worth et al.
1992a
Ngim et al. 1992
Ellingson et al. 1993
Kishietal. 1993
3-14
-------
Table 3-5 (continued)
Neurotoxicity of Elemental Mercury in Humans: Epidemiological Studies
Species/
No. per Sex
Human/19 M,
69 F
exposed/97
controls
Exposure
Duration
>2yr
Avg. 10.4 yr
(occup)
Dose
(mg/m3)
0.033 (avg)
0.008-0.085
(range)
Effects/Limitations/BML
Increased fatigue and confusion; impaired performance on
neurobehavioral tests in fluorescent lamp factory workers
(p<0.01)
Avg. BML: 25 ug/L in urine
Reference
Liang etal. 1993
* Estimate by extrapolating from urinary mercury levels (Roels et al. 1987).
In animals, as in humans, adverse neurological effects are observed after exposure to elemental
mercury vapor. Effects observed in rabbits and mice after subchronic exposures included tremors, ataxia,
paralysis, failure to respond to light and decreased conditioned avoidance responding (Fukuda 1971;
Ganser and Kirschner 1985; Kishi et al. 1978). Pathologic changes (unspecified) were observed in the
brains of rabbits at 0.86 mg Hg/m3 (Ashe et al. 1953).
Table 3-6
Neurotoxicity of Elemental Mercury in Animals: Inhalation Exposure
Species/
Strain/
No. per Sex
per Group
Rat/Albino/7 M
exposed/6 M
control
Mouse/
C57BL6J/
No. and sex NS
Rabbit/774
strain NS/
31(sexNS)
Rabbit/ strain
NS/6M
Exposure
Duration
12-42 wk
5d/wk
3hr/d
3.5 wk
5d/wk
20-40 min/d
l-12wk
5d/wk
7hr/d
13 wk
4d/wk
6hr/d
Dose
(mg/m3)
0,3
NS
(saturated)
0.86
0,4
Effects/Limitations/BML
Tremor; decline in conditioned avoidance and conditioned
escape responses. First significant effect at 20 weeks
(p<0.05)
Limitation: Only one level tested
BML Range: 11.18-17.83 ug/g in cerebrum (wet weight)
Ataxia; motor dysfunction
Limitations: Poorly defined exposure conditions; limited
data reporting on effects
BML not reported
Mild to moderate pathological changes in brains
Limitations: One exposure level; limited data reporting
BML: Brain level 1.2 ug/g
Tremor
Limitation: No control
BML: 0.8-3.9 ug/g wet weight (brain)
Reference
Kishi etal. 1978
Ganser and
Kirschner 1985
Ashe et al. 1953
Fukuda 1971
3-15
-------
3.1.3.3 Renal
The kidney is a sensitive target organ following inhalation exposure to elemental mercury. Acute
accidental exposure in private homes or as a result of industrial accidents resulted in symptoms ranging
from slight changes in urinary acid excretion to transient renal failure with proteinuria, nephrosis and
necrosis of the proximal convoluted tubules (Bluhm et al. 1992b; Jaffe et al. 1983; Rowens et al. 1991;
Tubbs et al. 1982). Proteinuria, proximal tubule damage and glomerulosclerotic changes were also
reported in a workers occupationally exposed for up to 2.5 years; in two cases the exposure levels were
measured at 0.02 to 0.45 mg/m3 (Friberg et al. 1953; Kazantzis et al. 1962). Comparisons of exposed
workers to unexposed controls found increased urinary N-acetyl-p-D-glucosaminidase in workers
exposed to 0.025 mg/m3 and increased incidence of proteinuria (Roels et al. 1982).
Table 3-7
Renal Toxicity of Elemental Mercury in Humans: Case Studies
Species/
No. per
Sex
Human/2 M
(adult)
Human/3 M
Human/2 M/
41 M controls
Human/ 1 F
(8-month old)
Human/
2M, 2F
Human/ 1 1 M
Exposure
Duration
8-9 mo
(occup)
4 mo-2.5 yr
(occup)
NS (occup)
= 1 day
Once
<16hr
Dose
(mg/m3)
0.02-0.45
NS
NS
NS
NS
NS
Effects/Limitations/BML
Proteinuria and nephrosis in electrochemical industry workers
Limitations: small sample size; concomitant exposure to
chlorine
BML: 160-900 ug/L in urine
Heavy albuminuria; transient renal failure; proximal tubule
damage; glomeruloscierotic changes
Limitations: small sample size; concomitant exposure to other
mercurials and other compounds; limited exposure data
BML Range: 1,100-1,440 ug/L in urine
Proteinuria; glomerulonephritis in chemical plant workers
Limitation: small sample size; concomitant exposure to other
metals; limited exposure data
BML Range: 174-548 ug/24 hr urine
Acute renal failure (proteinuria, glucosuria, granular casts)
BML Range: 16 ug/24 hr urine
Necrosis of proximal tubule; increased serum urea nitrogen and
creatinine in 2 subjects
Limitations: small sample size; limited exposure data
BML Range: 94-220 ug/L Hg in urine
Hyperchloremia, low normal bicarbonate in urine in welders
following accidental exposure
Limitations: No information on pre-exposure range
BML: —60 ug/L in blood at 20 d postexposure
Reference
Friberg et al. 1953
Kazantzis et al. 1962
Tubbs etal. 1982
Jaffe etal. 1983
Rowens et al.
1991
Bluhm etal. 1992b
3-16
-------
Table 3-8
Renal Toxicity of Elemental Mercury in Humans: Epidemiological Studies
Species/
No. per Sex
Human/21 NS
Human/43
exposed/47
controls (sex
NS)
Human/ 62 M
exposed, 60 M
controls
Human/100 M
Human/58 M
exposed
Human/41 M
exposed, 41 M
controls
Human/ 60 M
exposed, 60 M
controls
Exposure
Duration
NS
(occup)
>6 months
Mean: 5.3 yr
(occup)
1-25 yr
(avg5.5yr)
(occup)
8 yr (avg)
(occup)
7.9 yr (avg.)
1-20 yr
(occup)
13. T- 5. Syr
(occup)
Dose
(mg/m3)
0.01-0.05
NS
0.046 (est.)
NS
0.059 (est)
0, 0.025
NS
Effects/Limitations/BML
Increased proteinuria in exposed pathology laboratory
workers compared to unexposed controls. Proteinuria cleared
when mercury exposure was limited.
BML: =35 ug/L in urine
Proteinuria significantly elevated (p<0.05). Exposed group
worked in amalgam or chloralkali plants; control workers
were matched from the same plants but unexposed.
Limitations: Exposure data limited
BML: 29.2 ug/L in blood (range: 53-135); 95.5 ug/g
creatinine in urine (range 9.9-286)
Among exposed chloralkali plant or zinc-amalgam factory
workers, renal function parameters were not different from
unexposed controls. Circulating anti-laminin antibodies
found in eight exposed, 0 controls. No dose-effect
relationship between blood or urine levels and occurrence of
anti-laminin antibodies. Exposure level estimated using
Roels et al (1987) conversion factor.
BML: 16 ug/L (range 2.5-75.6) in blood; 56 ug/g creatinine
(range 3-272) in urine
Small increase in prevalence of higher activities of NAG and
gamma-glutamyl transferase in chloralkali workers w/urinary
mercury excretion >100 ug/g creatinine. No renal function
changes in workers w/mean urine 67 ug/g creatinine
Limitation: Exposure data limited
BML: 67->100 ug/g creatinine in urine
Follow-up to Lauwerys et al. (1983). In contrast to the earlier
study, there was no evidence of anti-laminin antibodies in
exposed workers
BML: 72 ug/g creatinine in urine
Increased urinary N-acetyl-p-D-glucosaminidase in a group of
chloralkali workers. Controls were age-matched.
BML: 15.6 ug/L in blood
No evidence of glomerular or tubular damage (effect on
urinary albumin or N-acetyl-p-glucosaminidase activity) in
chloralkali workers compared to controls. NOAEL of 25
mg/m3 based on Roels et al. (1987) conversion factor.
Limitation: Exposure data limited
BML: 14 ug/L in blood; 17 ug/L in urine
Reference
Stewartetal. 1977
Roels etal. 1982
Lauwerys et al 1983
Stonardetal. 1983
Bernard etal. 1987
Barregard et al.
1988
Pilkivi and
Ruokonen 1989
Only one study was found of kidney effects in animals from exposure to elemental mercury
vapor (Ashe et al. 1953). The observed effects supported the human data, with kidney effects ranging
from moderate unspecified pathological changes at shorter durations to necrosis and cellular
degeneration at longer durations. Limited quantitative data were reported.
3-17
-------
Table 3-9
Renal Toxicity of Elemental Mercury in Animals: Inhalation Exposure
Species/
Strain/
No. per Sex
per Group
Rabbit/strain
NS/14 (sex
NS)
Exposure
Duration
1-30 hr
Dose
(mg/m3)
28.8
Effects/Limitations/BML
Kidney pathology correlated with exposure duration,
ranging from moderate changes at 1 hour to widespread
necrosis at 30 hours.
Limitations: No control group, limited data reporting
BML Range: 20-5,320 ug/L in blood
Reference
Ashe et al. 1953
3.1.3.4 Respiratory
Respiratory toxicity in humans following exposure to elemental mercury vapors has been
characterized by pulmonary edema and congestion, coughing, interstitial pneumonitis, respiratory failure
and absence of air in lungs at time of histopathological examination (Bluhm et al. 1992a; Hallee 1969;
McFarland and Reigel 1978; Milne et al. 1970; Snodgrass et al. 1981; Taueg et al. 1992). One case of
occupational exposure to elemental mercury vapor occurred due to a faulty thermostat that heated to
450°F and vaporized the mercury it contained. Signs included cough, chest pains, reduced vital capacity
and pneumonitis, which began within hours of the onset of exposure (McFarland and Reigel 1978).
Accidental exposure to elemental mercury vapors in private homes has led to interstitial pneumonia,
dyspnea, lung disease and respiratory failure (Hallee 1969; Snodgrass et al. 1981; Taueg et al. 1992). In
each case, signs of toxicity persisted for days to months following acute exposure. No studies were
identified regarding respiratory effects in humans following intermediate or chronic exposures to
elemental mercury vapor.
Table 3-10
Respiratory Toxicity of Elemental Mercury in Humans: Case Studies
Species/
No. per
Sex
Human/ 1 M,
1 F (adults)
Human/4 M
Exposure
Duration
<12hr
2.5-5 hr
(occup)
Dose
(mg/m3)
NS
1.1-1.7
(est.)
Effects/Limitations/BML
Dyspnea; interstitial pneumonia; fibrosis; moderate restrictive
lung disease
Limitation: Case study
BML Range: 191-557 ug/24hr urine
Cough; chest tightness occurred following an accidental
exposure of electrochemical industry workers
Limitation: Case study
BML Range: 100-130 ug/L in urine 10-14 days postexposure
Reference
Hallee 1969
Milne etal. 1970
3-18
-------
Table 3-10 (continued)
Respiratory Toxicity of Elemental Mercury in Humans: Case Studies
Species/
No. per
Sex
Human/6 M
Human/2 M,
2 F (adults)
Human/ 1 M
Human/17 M
Human/2 M,
2 F (adults)
Exposure
Duration
<8hr
3d
~2hr
<16hr
-24 hr
Dose
(mg/m3)
44.3
(est.)
NS
NS
NS
NS
Effects/Limitations/BML
Pneumonitis; cough; chest pain
Limitations: Case study; limited data reporting
BML Range: 1,060-3,280 ug/24 hr urine
Cough; dyspnea
Limitation: Case study
BML: 82-5700 ug/24 hr urine
Reduced vital capacity and dynamic lung volumes, shortness of
breath
Limitation: Case study
BML: 1,900 ug/L urine on first day
Congestion; dyspnea; lung infiltrates in up to 15/17 welders
interviewed following accidental exposure
Limitation: Limited data reporting of effects or exposure
BML: -60 ug/L in blood 20 d postexposure
Adult respiratory distress syndrome; respiratory failure
BML Range: 4.6-219 ug/L in urine
Reference
McFarland and
Reigel 1978
Snodgrass et al.
1981
Lilisetal. 1985
Bluhmetal. 1992a
Tauegetal. 1992
Rats exposed to 27 mg Hg/m3 as elemental mercury vapor for one hour exhibited dyspnea, and
exposure for two hours resulted in death by asphyxiation (Livardjani et al. 1991). Histopathological
analyses revealed necrosis of the alveolar membrane, presence of hyaline membranes and evidence of
pulmonary edema. Acute-duration studies with rabbits revealed degeneration and necrosis of the lungs
(Ashe et al. 1953). Gage (1961) reported congestion and necrosis of the lungs following intermediate-
duration exposure to elemental mercury vapor at a concentration of 1 mg Hg/m3.
3-19
-------
Table 3-11
Respiratory Toxicity of Elemental Mercury in Animals: Inhalation Exposure
Species/
Strain/
No. per Sex
per Group
Rat/Wistar/
6F
Rat/Wistar/64
M/ duration
Rabbit/ strain
NS/14(sexNS)
Exposure
Duration
7wk
100 hr/wk
5d/wk
1 or 2 hr
1-30 hr
Dose
(mg/m3)
l
0,27
28.8
Effects/Limitations/BML
Congestion; necrosis of lung
Limitation: Limited data reporting
BML: 10 ug/rat in lungs
Death by asphyxiation; lung edema; hyaline membranes;
necrosis of alveolar epithelium
BML Range: 391-4,558 ug/L in blood
Pathology correlated with exposure duration and ranged
from mild changes at 1 hour to marked cellular degeneration
and necrosis at 30 hours. In another study, Ashe reported
no respiratory damage in rats exposed to 0.1 mg/m3 for 72
weeks.
BML Range: 20-5,320 ug/L in blood
Reference
Gage 1961
Livardjani et al.
1991
Ashe et al. 1953
3.1.3.5 Cardiovascular
Signs of cardiovascular toxicity in humans after acute exposure to elemental mercury include
tachycardia, elevated blood pressure and heart palpitations (Bluhm et al. 1992a; Snodgrass et al. 1981;
Soni et al. 1992). Intermediate-duration exposure to elemental mercury vapors produced similar effects
(i.e., tachycardia and elevated blood pressure) (Fagala and Wigg 1992; Foulds et al. 1987). Barregard et
al. (1990) performed a study on chloralkali workers and showed that they had an increased risk of
ischemic heart disease and cerebrovascular disease. These workers, however, were exposed to other
chemicals and to magnetic fields which may have affected the results. Piikivi (1989) demonstrated a
positive correlation between heart palpitations and urinary mercury concentrations in workers from a
chloralkali plant. It is unclear from the available scientific literature, however, whether the effects on
cardiovascular function (e.g., tachycardia, elevated blood pressure) are due to direct cardiac toxicity or to
indirect toxicity (e.g., due to effects on neural control of cardiac function) of elemental mercury.
3-20
-------
Table 3-12
Cardiovascular Toxicity of Elemental Mercury in Humans: Case Studies
Species/
No. per Sex
Human/2 M, 2
F (adults)
Human/ 1 F
(child)
Human/17 M
Human/ 1 F
(child)
Human/ 1 M (3-
yr old)
Exposure
Duration
3d
2 mo
<16hr
6 mo
<6hr
Dose
(mg/m3)
NS
NS
NS
NS
NS
Effects/Limitations/BML
Elevated blood pressure; tachycardia
Limitations: Case study; limited exposure data
BML Range: 82-5,700 ug/24 hr urine
Elevated blood pressure; tachycardia
Limitations: Case study; limited exposure data
BML Range: 214-296 ug/L in 24 hr urine
Palpitations in 5/17 welders interviewed following accidental
exposure
Limitation: Exposure data limited
BML: —60 ug/L in blood 20 d postexposure
Elevated blood pressure; tachycardia
Limitations: Case study; limited exposure data
BML: 686 ug/24 hr urine
Tachycardia
Limitations: Case study; limited exposure data
BML not reported
Reference
Snodgrass etal.
1981
Foulds et al. 1987
Bluhmetal. 1992a
Fagala and Wigg
1992
Sonietal. 1992
Table 3-13
Cardiovascular Toxicity of Elemental Mercury in Humans: Epidemiological Studies
Species/
No. per
Sex
Human/41 M
exposed/41
M controls
Human/
26 M
Exposure
Duration
16 yr (avg)
5-27 yr
(occup)
10 yr (avg)
(occup)
Dose
(mg/m3)
0.03
(est.)
Avg
samples:
0.025-
0.050
Effects/Limitations/BML
Palpitations in chloralkali workers (p<0.05); no significant effect
on cardiovascular reflex responses compared to matched controls.
BML Range: 3.5-52.5 ug/L in urine; avg 19.3 ug/L in urine
Increased mortality due to ischemic heart and cerebrovascular
disease in chloralkali workers, compared to matched controls.
Limitation: Possible confounding due to shift work
BML: Decrease from 200 ug/L in urine in 1950's to <50 ug/L in
1990
Reference
Piikivi 1989
Barregard et al.
1990
Few animal studies were located regarding cardiovascular effects after exposure to elemental
mercury vapor. Studies in rabbits report unspecified cellular degeneration and necrosis of the
cardiovascular system following both acute and intermediate exposure (Ashe et al. 1953). Ashe et al.
(1953), however, concluded that the concentration of mercury is a better indicator of cardiovascular
toxicity than the duration of exposure, especially at lower exposure levels.
3-21
-------
Table 3-14
Cardiovascular Toxicity of Elemental Mercury in Animals: Inhalation Exposure
Species/
Strain/
No. per
Sex per
Group
Rabbit/strain
NS/14 (sex
NS)
Rabbit/ strain
NS/16 (sex
NS)
Rabbit/strain
NS/31(sex
NS)
Exposure
Duration
1-30 hr
1-11 wk
5d/wk
7hr/d
12 wk
5d/wk
7hr/d
Dose
(mg/m3)
28.8
6
0.86
Effects/Limitations/BML
Pathology correlated with exposure duration and ranged from
mild changes in the heart at 1 hour to marked cellular
degeneration and necrosis at > 12 hours.
Limitations: No controls; limited data reporting;
only one dose level tested
BML Range: 20-5,320 ug/L in blood
Mild to moderate pathological changes of the heart. Pathological
changes observed in subchronic studies were correlated with the
exposure concentration but not exposure duration.
Limitations: No controls; limited data reporting;
only one dose level tested
BML Range: 70-3,000 ug/L in blood
Most animals had mild heart pathology, but 2 animals each at 6
and 7 weeks had marked cellular degeneration and necrosis, with
focal fibrosis.
Limitations: Limited data reporting; only one dose level
BML Range: 50-620 ug/L blood
Reference
Asheetal. 1953
Asheetal. 1953
Ashe et al. 1953
3.1.3.6 Gastrointe stinal
Gastrointestinal effects have been reported by persons exposed to elemental mercury vapor. The
most common sign of mercury poisoning is stomatitis (inflammation of the oral mucosa), which is
usually reported following acute, high concentration exposure to elemental mercury vapors (Bluhm et al.
1992a; Snodgrass et al. 1981). Sexton et al. (1978), however, reported signs of bleeding gingiva in 12
people exposed to mercury vapors for two months after metallic mercury was spilled in two homes, and
Schwartz et al. (1992) reported bleeding gums in a child exposed to mercury vapors for two to four
weeks. In addition, Vroom and Greer (1972) documented mercury intoxication in nine workers at a
thermometer manufacturing plant; the workers complained of sore gums and lesions on the oral mucosa
after long-term exposure. Other commonly reported gastrointestinal effects include nausea, vomiting,
diarrhea and abdominal cramps (Bluhm et al. 1992a; Campbell 1948; Lilis et al. 1985; Sexton et al. 1978;
Snodgrass et al. 1981; Vroom and Greer 1972).
3-22
-------
Table 3-15
Gastrointestinal Toxicity of Elemental Mercury in Humans: Case Studies
Species/
No. per Sex
Human/ 1 F
(4-month old)
Human/3 M,
6F
Human/5 M,
6 F (adults and
children)/12
controls (sex
NS)
Human/2 M,
2 F (adults)
Human/ 1 M
Human/ 17 M
Exposure
Duration
5hr
NS
(occup)
51-176 d
3 days
~2hr
<16hr
Dose
(mg/m3)
NS
NS
0.1-1.0
NS
NS
NS
Effects/Limitations/BML
Difficulty swallowing; abdominal pain; necrosis of stomach
mucosa and duodenum
Limitations: Case study; exposure data limited
BML not reported
Sore gums; diarrhea in thermometer manufacturing employees
Limitation: Exposure data limited
BML Range: 4-1,101 ug/24hr urine
Nausea, vomiting, abdominal pain, anorexia, diarrhea, bleeding
gingiva more common than in controls
Limitations: Small sample size; no statistical analysis
BMLavg: 3 .7 ug/L in urine;
BML Range: 183-620 ug/L in blood
Nausea; vomiting; swelling of gums
Limitation: Case study
BML Range: 13-5,700 ug/24hr urine
Nausea; vomiting
Limitations: Case study; limited reporting of symptoms;
exposure data limited
BML Range: 900-1,900 ug/L in urine (over 3 days)
Diarrhea; cramps in up to 1 1/17 welders accidentally exposed
Limitations: Case study; exposure data limited
BML not reported
Reference
Campbell 1948
Vroom and Greer
1972
Sexton etal. 1978
Snodgrass etal.
1981
Lilis et al. 1985
Bluhmetal. 1992a
Very little information is available concerning gastrointestinal toxicity after exposure to
elemental mercury vapors. Ashe et al. (1953) exposed rabbits to mercury vapors for 1-30 hours at a
concentration of 28.8 mg Hg/m3 and found unspecified cellular degeneration and necrosis. When rabbits
were exposed to 6 mg Hg/m3 for 1-11 weeks, changes in the colon were seen during histopathological
analysis (Ashe et al. 1953).
3-23
-------
Table 3-16
Gastrointestinal Toxicity of Elemental Mercury in Animals: Inhalation Exposure
Species/
Strain/
No. per
Sex per
Group
Rabbit/ strain
NS/14 (sex
NS)
Rabbit/ strain
NS/16 (sex
NS)
Exposure
Duration
1-30 hr
1-11 wk
5d/wk
7hr/d
Dose
(mg/m3)
28.8
6.0
Effects/Limitations/BML
Earliest effect on colon (mild pathological changes) occurred at 2
hr; marked cellular degeneration and necrosis observed at 30 hr
Limitations: No controls; limited data reporting
BML Range: 20-5,320 ug/L in blood
No effects or mild histopathological changes in colon
Limitations: No controls; limited data reporting
BML Range: 70-3,600 ug/L blood
Reference
Asheetal. 1953
Asheetal. 1953
3.1.3.7 Hepatic
Biochemical changes in hepatic enzymes were noted in a child who was exposed for
approximately one day to an unspecified concentration of elemental mercury vapors (Jaffe et al. 1983).
Serum glutamic-pyruvic transaminase (SGPT) and bilirubin levels were elevated, and synthesis of
hepatic coagulation factors was reduced. No human studies were identified regarding the hepatic toxicity
of mercury following intermediate or chronic exposures to elemental mercury vapors.
Table 3-17
Hepatic Toxicity of Elemental Mercury in Humans: Case Study
Species/
No. per
Sex
Human/ 1 F
(8-month old)
Exposure
Duration
-Id
Dose
(mg/m3)
NS
Effects/Limitations/BML
Elevated serum alanine amino-transferase and bilirubin
Limitations: Case study; exposure data limited
BML: 16 ug/24 hr urine
Reference
Jaffe etal. 1983
Ashe et al. (1953) performed histopathological analyses on rabbits after exposing them for one to
30 hours or for one to 11 weeks to elemental mercury vapors. The analyses revealed necrosis and
cellular degeneration of the liver. No other animal studies were identified regarding the hepatic toxicity
of mercury vapors following inhalation exposure.
3-24
-------
Table 3-18
Hepatic Toxicity of Elemental Mercury in Animals: Inhalation Exposure
Species/
Strain/
No. per
Sex per
Group
Rabbit/strain
NS/14 (sex
NS)
Rabbi!/ strain
NS/16 (sex
NS)
Exposure
Duration
1-30 hr
1-11 wk
5d/wk
7hr/d
Dose
(mg/m3)
28.8
6.0
Effects/Limitations/BML
Pathology correlated with exposure duration. Moderate changes
first occurred at 2 hr and widespread necrosis at 30 hr
Limitations: No controls; limited data reporting
BML Range: 20-5,320 ug/L in blood
Pathology was somewhat correlated with exposure duration and
ranged from mild to marked cellular degeneration with necrosis.
Limitations: No controls; limited data reporting
BML Range: 70-3,600 ug/L blood
Reference
Asheetal. 1953
Ashe et al. 1953
3.1.3.8 Hematological
After acute-duration exposure to high concentrations of elemental mercury vapor, onset of "metal
fume fever" may occur; this syndrome is characterized by leukocytosis with fever, chills and fatigue
(Campbell 1948; Haddad and Stenberg 1963; Jaffe et al. 1983). Intermediate-duration exposure to
mercury vapors led to an elevated white blood cell count in a 12-year-old female after exposure for six
months (Fagala and Wigg 1992). Volunteers with dental amalgams had significantly decreased
hemoglobin and hematocrit compared to controls without dental amalgams (Siblerud 1990).
Table 3-19
Hematological Toxicity of Elemental Mercury in Humans: Case Studies
Species/
No. per
Sex
Human/ 1 F
(child)
Human/ 1 M
(3. Syr old)
Exposure
Duration
6 mo
2-4 wk
Dose
(mg/m3)
NS
NS
Effects/Limitations/BML
Elevated white cell count
Limitations: Case study; exposure data limited; skin lesions
could have led to elevated count
BML: 686 ug/24 hr urine
Thrombocytopenia
Limitation: Exposure data limited
BML: 151 ug/L in blood
Reference
Fagala and Wigg
1992
Schwartz et al. 1992
3-25
-------
Table 3-20
Hematological Toxicity of Elemental Mercury in Humans: Epidemiological Studies
Species/
No. per Sex
Human/47 (sex
NS)
Human/41 M
exposed/55
controls
Human/20 M,
30 F
exposed/21 M,
30 F control
Exposure
Duration
NS
(occup)
NS
(occup)
NS
Dose
(mg/m3)
<0.1
Range:
0.106-
0.783
NS
Effects/Limitations/BML
Decreased y-aminolevulinic acid dehydratase and
cholinesterase activity in erythrocytes, effects were
significantly (p<0.01) correlated to urinary mercury
BML Range: 2-472 ug/g of creatinine in urine.
Increased a2-macroglobulin and ceruloplasmin in mercury
plant workers compared to unexposed controls (p<0.001)
BML Range: 29-545 ug/L in urine
Subjects with amalgams had decreased (p<0.02) mean
hemoglobin (14.66±1.09 g/dL in subjects vs. 14.88±1.14 g/dL
in controls) and mean hematocrit (43.15±3.66% in subjects vs.
43.91±3.61% in controls). These reductions were significantly
(p<0.01) correlated with increasing urine mercury in the
subjects with amalgam.
Limitations: Subjects identified through newspaper ads may
have introduced self-selection bias; only mean data reported.
BMLavg: 3 .7 ug/L in urine
Reference
Wada et al. 1969
Benckoetal. 1990
Siblerud 1990
No animal studies were identified regarding the hematological toxicity of mercury vapors
following inhalation exposure.
3.1.3.9 Immunological
The available evidence suggests that the immune reaction to elemental mercury exposure is
idiosyncratic, with either increases or decreases in immune activity depending on genetic predisposition.
Although there is evidence for an overall suppression of the humoral immune response among exposed
workers (Moszczynski et al. 1990), this effect has not been consistently observed (Bencko et al. 1990;
Langworth et al. 1992b). The failure to observe consistent decreases in antibody content of the serum
may be due to small numbers of workers in each group who develop an autoimmune reaction upon
exposure to mercury. For example, small numbers of workers exposed to elemental mercury vapors have
had elevated levels of antiglomerular basement membrane and anti-DNA antibodies (Cardenas et al.
1993; Langworth et al. 1992b) or granular deposition of IgG and complement C3 in the renal glomeruli
(Tubbsetal. 1982).
3-26
-------
Table 3-21
Immunotoxicity of Elemental Mercury in Humans: Case Study
Species/
No. per
Sex
Human/2 M
Exposure
Duration
NS
(occup)
Dose
(mg/m3)
<0.1
Effects/Limitations/BML
Deposition of IgG and C3 in glomeruli of chemical plant workers
Limitation: Case study
BML Range: 174-548 ug/24 hr urine
Reference
Tubbsetal. 1982
Table 3-22
Immunotoxicity of Elemental Mercury in Humans: Epidemiological Studies
Species/
No. per Sex
Human/41 M
exposed/55
controls
Human/50
exposed/50
controls
Exposure
Duration
NS
(occup)
1.5-25yr
Avg: 11 yr
(occup)
Dose
(mg/m3)
Range:
0.106-
0.783
NS
Effects/Limitations/BML
Decreased IgG; increased IgA and IgM in mercury plant
workers (p<0.05)
BML Range: 29-545 ug/L in urine
Abnormally high anti-DNA antibody litre (p<0.01)
Mean BML: 31.9 ug/L in urine
Reference
Bencko et al. 1990
Cardenas et al. 1993
An autoimmune response to mercury has been produced in a susceptible strain of rats (Brown
Norway) exposed to mercury vapor (Hua et al. 1993). In these rats, increased levels of serum IgE and
antilaminin autoantibodies, deposition of IgG deposits in the renal glomeruli and proteinuria were
observed.
Table 3-23
Immunotoxicity of Elemental Mercury in Animals: Inhalation Exposure
Species/
Strain/
No. per
Sex per
Group
Rat/BN/3-4
M, 3-4 F
Exposure
Duration
5 wk
6 or 24 hr/d
Dose
(mg/m3)
0, 1
Effects/Limitations/BML
Increased serum IgE; anti-laminin autoantibody litre, IgG
deposits along glomerular capillary walls (p<0.001)
Mean BML: 90.3 ug/L in blood
Reference
Huaelal. 1993
3-27
-------
3.1.3.10 Dermal
Exposure to elemental mercury vapors for acute or intermediate durations may elicit a response
known as acrodynia or "pink disease", which is characterized by peeling palms of hands and soles of feet,
excessive perspiration, itching, rash, joint pain and weakness, elevated blood pressure and tachycardia
(Fagala and Wigg 1992; Karpathios et al 1991; Schwartz et al 1992). Children seem to be the most
susceptible to acrodynia, although adults may be affected to a lesser degree (Warkany and Hubbard
1953). One man experienced a rash and stomatitis after inhalation exposure to mercury when repairing a
cell in a chloralkali plant (Bluhm et al. 1992a); however, dermal exposure may have also occurred.
Table 3-24
Dermal Toxicity of Elemental Mercury in Humans: Case Studies
Species/
No. per Sex
Human/ 1 M
(child)
Human/ 17 M
Human/ 1 F
(child)
Human/ 1 M
(3-yr old)
Exposure
Duration
2wk
<16hr
6 mo
2-4 wk
Dose
(mg/m3)
NS
NS
NS
NS
Effects/Limitations/BML
Red palms and soles; perspiration; rash
Limitations: Case study; concomitant dermal exposure
possible; exposure data limited
BML: 130 ug/24hr urine
Conjunctivitis; dermatitis in 8/17 welders exposed in an
accident
Limitation: Exposure data limited
BML not reported
Peeling skin on palms and soles
Limitations: Case study; exposure data limited
BML: 686 ug/24 hr urine
Maculopapular whole body rash
Limitations: Case study; exposure data limited
BML: 151 ug/L in blood
Reference
Karpathios et al.
1991
Bluhm etal. 1992a
Fagala and Wigg
1992
Schwartz et al. 1992
No animal studies were identified regarding the dermal toxicity of mercury vapors following
inhalation exposure.
3.1.3.11 Developmental
Although few reports have addressed the effects of maternal exposure to elemental mercury
vapor on the developing fetus, the available information suggests that maternal exposure to sufficiently
high concentrations of elemental mercury vapor may adversely affect the developing fetus. A study of
the pregnancies of Polish dental professionals showed a high frequency of malformations of a
nonspecified nature (Sikorski et al. 1987). In contrast, a study of Swedish dental professionals found no
increases in malformations, abortions, or stillbirths (Ericson and Kallen 1989). An increase in low birth
weight infants was noted in the offspring of female dental nurses (Ericson and Kallen 1989); however, in
this same study similar effects were not observed for either dentists or dental technicians, and
socioeconomic factors may have contributed to the effects observed. It is unknown to what extent
discrepancies in the results of the above studies are attributable to differences in mercury exposure levels
(only the study by Sikorski et al. (1987) attempted to assess exposure levels) or to other confounders.
3-28
-------
Table 3-25
Developmental Toxicity of Elemental Mercury in Humans: Case Studies
Species/
No. per Sex
Human/ 1 F
Human/ 1 F
Human/ 1 F
Exposure
Duration
8 mo
(occup)
2yr
(occup)
= 17wk
Dose
(mg/m3)
NS
NS
0.02-0.06
Effects/Limitations/BML
Infant death at birth; fetal hepatomegaly; spontaneous
abortion in 2 successive pregnancies of a thermometer-
manufacturing worker. Maternal toxicity included tremors,
motor incoordination, hyperreflexivity, stomatitis.
Limitations: Case study; exposure data limited; maternal
toxicity also occurred.
BML not reported
Delivery of viable infant at term to thermometer factory
worker with mild peripheral neuropathy attributed to mercury.
Maternal toxicity included slight decrease in sensory reflexes.
Limitations: Case study; exposure data limited; no
neurological assessment of infant; slight maternal toxicity also
reported.
BML: Mother: 875 ug/L in urine; Offspring: 2.5 ug/L in
urine
Delivery of normal child who met all developmental
milestones. No maternal toxicity reported.
Limitation: Case study; no psychodevelopmental testing
BML: Mother: 230 ug/L in 24 hr urine at 17 wk, then
declined; Offspring: 3,000 ug/g in hair
Reference
Derobert et al. 1950
Melkonian and
Baker 1988
Thorp etal. 1992
The few animal studies that were identified indicate that inhalation of elemental mercury vapor
may be toxic to the developing animal. In an abstract, Steffek et al. (1987) reported decreased fetal
weight in offspring of rats exposed to elemental mercury vapor during gestation. Increased fetal and
postnatal deaths were also reported by Baranski and Szymczyk (1973) among rats exposed to elemental
mercury vapor for three weeks prior to mating and then again on gestation days 1-20, and increased
resorptions in rats exposed on gestation days 10-15 or 1-20.
Pregnant Sprague-Dawley rats (12/group) were exposed on gestation days 11-14 and 17-20 to
elemental mercury vapors (1.8 mg/m3) for one or three hours/day (Danielsson et al. 1993). Litters were
culled to 4 males and 4 females. Behavioral testing was done on one male and one female adult from
each litter; the authors state that for behavioral testing 8 were tested for each group. There was no
difference between controls and treatment groups for maternal weight gain. There was no obvious
mercury toxicity in the dams. Offspring exposed in utero were no different from controls in the following
measures: body weight; clinical signs; pinna unfolding; surface righting reflex development; tooth
eruption; and results of a negative geotaxis test at days 7, 8 or 9 post par turn. At 3 months of age,
exposed male but not female rats showed significant decrements in four measures of spontaneous motor
activity: locomotion, rearing, rearing time and total activity. By 14 months, the high-dose animals
showed hyperactivity in the same test. Females were not evaluated in other adult behavioral tests. A test
for habituation to novel environment at 7 months of age showed significant differences between controls
and treated males on four measures. At 4 months, mercury-treated males had significantly higher latency
in a maze learning test; at 15 months, there was no difference between controls and treated rats in a
circular swim maze test.
3-29
-------
Table 3-26
Developmental Toxicity of Elemental Mercury in Humans: Epidemiological Studies
Species/
No. per Sex
Human/ 349 F
exposed, 215 F
controls
Human/57 F
Human/8157 F
exposed
Exposure
Duration
NS
(occup)
0.5-27 yr
(occup)
NS
(occup)
Dose
(mg/m3)
NS
NS
NS
Effects/Limitations/BML
Rates of pregnancy and labor complications were high among
women exposed to elemental mercury. Insufficient detail
provided to evaluate dose-response relationship.
Limitation: Lack of exposure or effect data
BML not reported
In a study of 57 dental professionals (117 pregnancies),
reproductive failure (spontaneous abortion, stillbirth or
congenital malformation-not described further) was higher
than among unexposed controls, and the effect correlated
with exposure level (p=0.004). No maternal toxicity
reported.
Limitations: Small study group; control group not described;
exposure data limited
BMLavg: 0.527 ug/g in scalp hair
Study of infants born to dental workers, compared with the
general population. Based on medical registry, no increase in
malformations, abortions, or stillbirths. Increased incidence
of low birth weight infants among offspring of dental
assistants (risk ratio 1 .2, 95% C.I. 1 .0-1 .3), but the risk ratio
was decreased for dentists, suggesting a socioeconomic
effect. Case-control study of infants with neural tube defects
found none born to dentists, but the expected number was
only 0.5. No maternal toxicity reported.
Limitation: Exposure data limited
BML not reported.
Reference
Mishinova et al.
1980
Sikorskietal. 1987
Ericson and Kallen
1989
Early postnatal exposure (during a period of rapid brain growth) resulted in subtle behavioral
changes when the rats were tested as young adults (Fredriksson et al. 1992). Eight litters/group, culled to
8 individuals, were exposed to 0.05 mg/m3 for either 1 or 4 hr/day. Exposure was on days 11-17 of age.
There were no signs of overt toxicity or changes in body weight. Spontaneous motor activity was
evaluated at 2 and 4 months. The high-dose group showed increased rearing at the early test, but the
repeat test indicated hypoactivity. The low-dose group was no different from controls at two months; at
four months this group showed increased total activity and decreased rearing. In the spatial learning test
administered at 6 months, low- dose rats had increased time to complete the task. High-dose animals
were observed to have increases in time to complete the task and in numbers of errors. No information
was given on the number of males and females tested or on any differences in behavior dependent on
gender.
3-30
-------
Table 3-27
Developmental Toxicity of Elemental Mercury in Animals: Inhalation Exposure
Species/
Strain/
No. per Sex
per Group
Rat/Strain
NS/23-24 F
Rat/Sprague-
Dawley/NS F
Rat/Sprague
Dawley/ 4M, 4F
Rat/Sprague-
Dawley/8 F
Squirrel
monkey/lOM, IF
Exposure
Duration
Group I: 6 hr/d,
6-8 weeks
before
fertilization:
Group II: 3 wk
before mating
and Gd 7-20
6 or 20d
24 hr/d
Gd 10-15 or
Gd 1-20
1 or 3 hr/day on
Gd 11-14 plus
17-20
7d
1 or 4 hr/d
on post-partum
days 11-17
4-7 hr/day, 5
d/week, during
gestation
Dose
(mg/m3)
0,2.5
0,0.1,0.5,
1.0
0, 1.8 mg/m3
0, 0.05
0.5 or 1.0
(1304 to
4305 ,ug
total)
Effects/Limitations/BML
Group I: Decreased number of live pups (p<0.05);
decreased relative kidney (p<0.01) and liver weights
(p<0.05) and increased ovaries (p<0.05) in 2-month-old
pups. Group II: Mean number of live fetuses lower than
in controls
Limitations: Wide range in actual mercury concentration
(0.5-4.8 mg/m3); only one level tested; maternal toxicity
BML not reported
Increased resorptions (LOAEL = 0.5 for Gd 10-15 and
1.0 for Gd 1-20); decreased maternal and fetal weights in
group exposed to 1.0 on Gd 1-20.
Limitations: Reported only as an abstract; limited study
details; maternal toxicity
BML not reported
Hypoactivity at 3 months; hyperactivity at 14 months;
decrement in habitation to novel environment at 7
months; retarded learning in radial arm maze at 4 months
but no difference from controls in circular swim maze at
15 months. BML ranges for control through high dose
group (mg Hg/kg in organs): 0.001-0.012 (brain); 0.004-
0.112 (liver); 0.002-0.068 (kidney).
Limitations: Limited testing of female offspring; no
evaluation of differences between males and females;
small numbers of rats/group.
Impaired spatial learning at 6 months (p<0.01); increased
locomotor activity in objective test (p<0.01)
BML Range: 0.017-0.063 ug/g in brain
Limitations: No information on gender-specific
behavioral effects; small number of animals/group.
Instability in lever-press durations and steady-state
performance under concurrent schedules; aberrant
transitions in treated animals. Five male monkeys born at
same time served as controls; there were 5 treated M and
one treated F. Maternal BML ranges 0.025 to 0.18 ,ug/g.
Reference
Baranski and
Szymczyk 1973
Steffeketal. 1987
Danielssonetal.
1993
Fredrikssonetal.
1992
Newland et al. 1996.
3.1.3.12 Reproductive
Most studies that have examined the effects of occupational exposure to elemental mercury
vapor on reproductive function have failed to find evidence of adverse effects (Alcser et al. 1989;
Brodsky et al. 1985; Erfurth et al. 1990; Ericson and Kallen 1988; Heidam 1984; Lauwerys et al. 1985;
McGregor and Mason 1991). A few studies have shown at least suggestive evidence that elemental
mercury exposure may adversely affect reproductive function. In females exposed occupationally to
metallic mercury vapor, a correlation was observed between scalp hair mercury and reproductive failure
3-31
-------
or menstrual abnormalities (Sikorski et al. 1987). An increased incidence of pregnancy complications
such as toxicosis or prolonged or hemorrhagic parturition was observed in exposed females when
compared to unexposed controls (Mishonova et al. 1980). A slightly increased incidence of menstrual
disorders in exposed females was reported by DeRosis et al. (1985); however, the statistical significance
of this finding was not presented. No evidence for an effect on fertility was observed in exposed males,
but one study of wives of exposed workers found an increased rate of spontaneous abortions (Cordier et
al. 1991). It is possible that the wives were exposed to mercury as the result of handling contaminated
clothing. None of the above studies presented information on exposure levels, and few presented
biomonitoring data. Thus, it is difficult to compare findings in the various studies.
Table 3-28
Reproductive Toxicity of Elemental Mercury in Humans: Epidemiological Studies
Species/
No. per Sex
Human/728 F
exposed/1034 F
controls
Human/29,514
M, 30,272 F
Human/153 F
exposed/193 F
controls
Human/103 M
exposed/101 M
controls
Human/57 F
exposed
Human/8157 F
Human/247 M
exposed/255 M
controls
Exposure
Duration
NS
(occup)
NS
(occup)
<5-17yr
(occup)
Avg: 5.9yr
(1-25 yr)
(occup)
0.5-27 yr
(occup)
NS
(occup)
4 mo-8 yr
(occup)
Dose
(mg/m3)
NS
NS
<0.01
TWA at
study;
>0.05 for 4
yr
NS
NS
NS
NS
Effects/Limitations/BML
No increase in rate of spontaneous abortions in gardeners,
dental assistants, painters, and factory workers.
BML not reported
No correlation between mercury exposure (low and high) and
rate of spontaneous abortions in dentists, dental assistants, or
their wives.
BML not reported
Slightly increased prevalence of menstrual disorders in
mercury lamp manufacturers, compared with workers subject
to similar stresses but not exposed to mercury.
Limitations: Subjective measures; no statistical analysis
BML not reported
No effect of paternal exposure on fertility of chloralkali,
amalgam or electrical equipment workers, compared to
controls with similar workloads.
Avg BML: 52.4 ug/g creatinine in urine (range 5.1-272)
In a study of 57 dental professionals (117 pregnancies),
reproductive failure (spontaneous abortion, stillbirth or
congenital malformation) was higher than among unexposed
controls, and the effect was correlated with exposure level
(extrapolated from hair Hg levels) (p=0.004). Irregular,
painful, or hemorrhagic menses was correlated with exposure
duration (p=0.005).
Limitations: Small study size; control group not described;
exposure data limited
Avg BML: 0.527 ug/g in scalp hair
Based on medical registry, there was no increase in
spontaneous abortions or stillbirths in pregnancies of dental
professionals, compared to the general population.
BML not reported
No association between paternal exposure and rate of
miscarriages in Department of Energy plant workers.
Limitation: Potential recall bias
BML: reported only as value integrated over time
Reference
Heidam 1984
Brodskyetal. 1985
DeRosis etal. 1985
Lauwerys etal. 1985
Sikorski etal. 1987
Ericson and Kallen
1988
Alcseretal. 1989
3-32
-------
Table 3-28 (continued)
Reproductive Toxicity of Elemental Mercury in Humans: Epidemiological Studies
Species/
No. per Sex
Human/20 M
exposed/21 M
controls
Human/1 52 F
exposed/374 F
controls
Human/
40 M exposed/
63 M controls
Exposure
Duration
2-18 yr
(occup)
NS
(occup)
2-20 yr
(occup)
Dose
(mg/m3)
NS
NS
NS
Effects/Limitations/BML
No correlation between blood or urinary mercury, and male
gonadotropic hormones of chloralkali workers, other
industrially-exposed workers, or dentists and matched
controls.
BML: Avg. 46 ug/g creatinine in urine (workers); 2.3 ug/g
creatinine (dentists)
Increased spontaneous abortions in women whose husbands
were exposed to mercury vapors in chloralkali plants (rate
doubled above 50 ug/L in urine; 95% C.I. = 0.99-5.23)
Limitation: Exposure data limited
Avg BML: 61.9 ug/L in urine (range 26.9-75.9 ug/L)
No correlation between blood or urinary Hg and male
gonadotropic hormones in workers from different industries
(not specified)
Avg BML: 103 ug/g creatinine in urine
Reference
Erfuthetal. 1990
Cordieretal. 1991
McGregor and
Mason 1991
In rats exposed to elemental mercury vapor, prolongation of estrous cycles was observed both
when compared to either unexposed controls or preexposure rates of cycling (Baranski and Szymczyk
1973).
Table 3-29
Reproductive Toxicity of Elemental Mercury in Animals: Inhalation Exposure
Species/
Strain/
No. per Sex
per Group
Rat/Strain
NS/24 F
Exposure
Duration
3 wk
5d/wk
6 hr/d and
Gd 7-20
Dose
(mg/m3)
2.5
Effects/Limitations/BML
Longer estrous cycles, but the effect was not statistically
significant
BML not reported
Reference
Baranski and
Szymczyk 1973
3.1.3.13 Genotoxicity
Cytogenetic monitoring studies in populations exposed occupationally to elemental mercury
vapor provide conflicting evidence for a clastogenic effect of elemental mercury. Early studies reported
increased frequencies of chromosomal aberrations among exposed workers (Popescu et al. 1979;
Verschaeve et al. 1976). These studies, however, were not well-controlled, and the results could not be
reproduced in later studies (Mabille et al. 1984; Verschaeve et al. 1979). Popescu et al. (1979) compared
two groups of men exposed to elemental mercury vapor (Group I, n=4; Group II, n=18) with an
3-33
-------
unexposed group often individuals and found a statistically significant increase in incidence of
chromosome aberrations in the exposed groups. Verschaeve et al. (1976) found an increase in
aneuploidy in lymphocytes of 28 subjects exposed to low concentrations of mercury vapor (by
comparison to seven controls), but these results were not repeated in later studies (Verschaeve et al.
1979). Mabille et al. (1984) did not find increases in structural chromosomal aberrations of lymphocytes
of exposed workers.
More recently, Barregard et al. (1991) demonstrated a correlation between cumulative mercury
exposure and induction of micronuclei among a group of chloralkali workers, suggesting a clastogenic
effect. This study did not show significant differences in frequency or size of micronuclei between the
exposed group to unexposed controls who were matched for age and smoking habits. Neither did they
find a correlation between the induction of micronuclei and current mercury exposure as measured by
blood or urine mercury levels. A correlation, however, was observed between cumulative exposure to
mercury and micronuclei induction in T-lymphocytes in exposed workers suggesting a genotoxic effect.
Table 3-30
Genotoxicity of Elemental Mercury in Humans
Species/
No. per Sex
Human/
8 M, 6 F (exposed)/
3 M, 4 F (control)
Human/4 M
exposed/10 controls
(sexNS)
Human/
28 exposed/
20 controls (sex NS)
Human/
22 exposed/
25 controls (sex NS)
Exposure
Duration
NS
(occup)
9.25 yr (avg)
(occup)
1-11 yr
(occup)
4 yr (avg)
(0.3-15.3 yr)
(occup)
Dose
(mg/m3)
NS
Range:
0.15-0.44
<0.05 at
time of
study
NS
Effects/Limitations/BML
Aneuploidy was significantly (p<0.001) increased in
subjects exposed due to an unstated occupation or as a
result of an accident at a university. Structural aberrations
were not increased.
Limitations: Small study size; smoking status not reported
BML Range: 1-114 ug/L in urine
Chromosome breaks (excluding gaps) were significantly
(p<0.001) increased in whole blood cultures taken from
chemical plant workers. There was no effect on numerical
chromosome aberrations.
Limitations: Small sample size; smoking status not
reported.
BML Range: 142-386 ug/L in urine at study
The incidence of structural and numerical chromosome
aberrations in exposed chloralkali plant workers did not
differ from controls. Eight of the controls were unexposed
workers at the same plant and 12 were taken from the
general population.
Limitations: Small sample size; there were 12 smokers in
the exposed group, 4/8 among the internal controls, and an
unknown number of smokers in the external controls
Avg BML: 35.4 ug/L in urine
No increase in structural chromosome aberrations in zinc
amalgam or chloralkali workers, compared to unexposed
control workers at the same plant.
Limitations: Small sample size; there were 15 smokers in
the exposed group and 12 among the controls; limited
exposure data
Avg BML: 30.6 ug/L in blood; range: 7.5-105 ug/L
Reference
Verschaeve etal.
1976
Popescuetal. 1979
Verschaeve et al.
1979
Mabille etal. 1984
3-34
-------
Table 3-30 (continued)
Genotoxicity of Elemental Mercury in Humans
Species/
No. per Sex
Human/26 M
Exposure
Duration
lOyr(avg)
(min. 1 yr)
(occup)
Dose
(mg/m3)
Avg
samples:
0.025-
0.050
Effects/Limitations/BML
The frequency of micronuclei in lymphocytes was
correlated with cumulative exposure in Swedish chloralkali
workers (p = 0.0035). There was no significant difference
between the frequency in the exposed and control
populations.
Controls were matched by age; exposed and control groups
each had 14 smokers
Limitation: Small study size
Avg BML: 9.6 ug/L in blood
Reference
Barregard et al.
1991
No studies were identified that examined the genotoxicity of elemental mercury in animals
following inhalation exposure. Likewise no studies of genotoxic effects of mercury exposure in vitro
were recovered.
3.2 Inorganic Mercury
Inorganic mercury occurs in numerous forms/compounds; the most common include mercuric
chloride (HgCl2), mercurous chloride (Hg2Cl2), mercuric oxide (HgO). The tables in this section include
a notation in the dose column indicating the specific form of inorganic mercury involved in that study.
Oral doses, shown in mg/kg-day, have been converted to mg Hg/kg-day using the method shown in
Appendix A.
3.2.1 Critical Noncancer Data
This section describes studies evaluated by U.S. EPA for use in assessing general systemic health
risks. Chapter 6 describes the derivation of an oral Reference Dose (RfD) for inorganic mercury based
on several studies wherein kidney diseases consequent to immunological effects were observed. For
completeness, some of these studies are also presented in tabular form in succeeding sections.
3.2.1.1 Human Data
Singer et al. (1987) studied nerve conduction velocity of the median motor, median sensor and
sural nerves in 16 workers exposed to various inorganic mercury compounds (e.g., mercuric oxides and
mercurial chlorides) for an average of 7.3 ± 7.1 years and compared to an unexposed control group using
t-tests. They found a slowing of nerve conduction velocity in motor, but not sensory, nerves that
correlated with increased blood and urine mercury levels and an increased number of neurologic
symptoms. The mean mercury levels in the exposed workers were 1.4 //g/L and 10 //g/L for blood and
urine, respectively. These urine levels are 2-fold less than those associated with peripheral neurotoxicity
in other studies (e.g., Levine et al. 1982). There was considerable variability in the data presented by
Singer et al. (1987), however, and the statistical analyses (t-test) were not as rigorous as those employed
by Levine et al. (1982) who used linearized regression analysis. Furthermore, the subsections in the
Levine et al. (1982) study were asymptomatic at higher urinary levels than those reported to be
associated with subjective neurological complaints in the workers studied by Singer et al. (1987). These
results, therefore, are not considered to be as reliable as those reported by Levine et al. (1982).
3-35
-------
Kazantzis et al. (1962) performed renal biopsies in 2 (out of 4) workers with nephrotic syndrome
who had been occupationally exposed to mercuric oxide, mercuric acetate and probably mercury vapors.
The authors felt that the nephrotic syndrome seen in 3 of the 4 workers may have been an idiosyncratic
reaction since many other workers in a factory survey had similarly high levels of urine mercury without
developing proteinuria. This conclusion was strengthened by work in Brown Norway rats indicating a
genetic (strain) susceptibility and that similar mercury-induced immune system responses have been seen
in affected humans and the susceptible Brown Norway rats (U.S. EPA 1987b).
3.2.1.2 Animal Data
Bernaudin et al. (1981) reported that mercurials administered by inhalation or ingestion to Brown
Norway rats resulted in the development of a systemic autoimmune disease. The mercuric chloride
ingestion portion of the study involved the forcible feeding of either 0 or 3 mg/kg/week of mercuric
chloride to male and female Brown Norway rats for up to 60 days. No abnormalities were reported using
standard histological techniques in either experimental or control rats. Immunofluorescence histology
revealed that 80% (4/5) of the mercuric-exposed rats were observed with a linear IgG deposition in the
glomeruli after 15 days of exposure. After 60 days of mercuric chloride exposure, 100% (5/5) of the rats
were seen with a mixed linear and granular pattern of IgG deposition in the glomeruli and granular IgG
deposition in the arteries. Weak proteinuria was observed in 60% (3/5) of the rats fed mercuric chloride
for 60 days. The control rats were observed to have no deposition of IgG in the glomeruli or arteries as
well as normal urine protein concentrations.
Andres (1984) administered mercuric chloride (3 mg per kg of body weight in 1 mL of water) by
gavage to five Brown Norway rats and two Lewis rats twice a week for 60 days. A sixth Brown Norway
rat was given only 1 mL of water by gavage twice a week for 60 days. All rats had free access to tap
water and pellet food. After 2-3 weeks of exposure, the Brown Norway mercuric chloride-treated rats
started to lose weight and hair. Two of the mercuric chloride-treated Brown Norway rats died 30-40
days after beginning the study. No rats were observed to develop detectable proteinuria during the 60-
day study. The kidneys appeared normal in all animals when evaluated using standard histological
techniques, but examination by immuno-fluorescence showed deposits of IgG present in the renal
glomeruli of only the mercuric-treated Brown Norway rats. The Brown Norway treated rats were also
observed with mercury-induced morphological lesions of the ileum and colon with abnormal deposits of
IgA in the basement membranes of the intestinal glands and of IgG in the basement membranes of the
lamina propria. All observations in the Lewis rats and the control Brown Norway rat appeared normal.
The only chronic oral study designed to evaluate the toxicity of mercury salts was reported by
Fitzhugh et al. (1950). In this study, rats of both sexes (20-24/group) were given 0.5, 2.5, 10, 40 or 160
ppm mercury as mercuric acetate in their food for up to 2 years. Assuming food consumption was equal
to 5% body weight per day, the daily intake would have been 0.025, 0.125, 0.50, 2.0 and 8.0 mg Hg/kg
for the five groups, respectively. At the highest dose level, a slight depression of body weight was
detected in male rats only. The statistical significance of this body weight depression was not stated.
Kidney weights were significantly (p<0.05) increased at the 2-and 8-mg Hg/kg-day dose levels.
Pathologic changes originating in the proximal convoluted tubules of the kidneys were also noted with
more severe effects in females than males. The primary weaknesses of this study were the lack of
reporting (which adverse effects were observed with which dosing groups) and that the most sensitive
strain, the Brown Norway rat, was not used for evaluating the mercury-induced adverse health effects.
NTP (1993) conducted subchronic and chronic gavage toxicity studies on Fischer 344 rats and
B6C3F1 mice to evaluate the effects of mercuric chloride, and the kidney appeared to be the major organ
3-36
-------
of toxicity. These studies were also summarized by Dieter et al. (1992). In the 6-month study, Fischer
344 rats (10/sex/group) were administered 0, 0.312, 0.625, 1.25, 2.5, or 5 mg/kg-day of mercuric chloride
(0.23, 0.46, 0.92, 1.9, and 3.7 mg Hg/kg-day), 5 days/week, by gavage. Survival was not affected,
although body weight gains were decreased in males at high dose and in females at or above 0.46 mg
Hg/kg-day. Alkaline phosphatase and gamma-glutamyl transferase levels in the urine were significantly
elevated in the females exposed to 3.7 mg Hg/kg-day at four and six months of exposure. Absolute and
relative kidney weights were significantly increased in both sexes with exposure to at least 0.46 mg
Hg/kg-day. The kidney weight changes were slightly dose-related in the females. Histopathology
revealed corresponding changes in the kidneys. In males, the incidence of nephropathy was 80% in
controls and 100% for all treated groups; however, severity was minimal in the two low-dose groups and
minimal to mild in the 0.92-mg Hg/kg-day group and higher. In females, there was a significant
increased incidence of nephropathy only at the high-dose group (4/10 with minimal severity).
Nephropathy was characterized by foci of tubular regeneration, thickened tubular basement membrane
and scattered dilated tubules containing hyaline casts. No treatment-related effects were observed in the
other organs; however, histopathology on the other organs was performed only on control and high-dose
rats.
B6C3F1 mice (10/sex/group) were administered gavage doses of 0, 1.25, 2.5, 5, 10, or 20 mg/kg-
day mercuric chloride (0, 0.92, 1.9, 3.7, 7.4, or 14.8 mg Hg/kg-day) 5 days/week for 6 months (NTP
1993). There was a decrease in body weight gain in males at the highest dose tested. Significant
increases occurred in absolute kidney weights of male mice dosed with 3.7 mg Hg/kg-day or more, and
relative kidney weights were increased in male mice at the 7.4 and 14.8 mg Hg/kg-day doses. The kidney
weight changes corresponded to an increased incidence of cytoplasmic vacuolation of renal tubule
epithelium in males exposed to at least 3.7 mg Hg/kg-day. The exposed female mice did not exhibit any
histopathologic changes in the kidneys.
In the 2-year NTP study, Fischer 344 rats (60 per sex per group) were administered 0, 2.5, and 5
mg/kg-day mercuric chloride (1.9 and 3.7 mg Hg/kg-day), 5 days a week, by gavage (Dieter et al. 1992;
NTP 1993). After two years, survival was significantly reduced in the treated male rats compared to the
controls. Mean body weights were significantly decreased in both treated males and females (9-10% and
14-15% decrease from control, respectively). At 15 months, relative kidney weights were significantly
elevated (not dose-related) in all treated groups (15-20% increase from control), and relative brain
weights were significantly elevated (slightly dose-related) in treated females (13-18%). The increased
kidney weights were accompanied by an increase in severity of nephropathy. After two years, there was
an increased incidence of nephropathy of moderate-to-marked severity and increased incidence of tubule
hyperplasia in the kidneys of exposed males compared to the controls. The control males exhibited
nephropathy, primarily of mild-to-moderate severity. Hyperparathyroidism, mineralization of the heart
and fibrous osteodystrophy were observed and considered secondary to the renal impairment. There
were no significant differences found in renal effects between exposed and control females. Other
nonneoplastic effects included an increased incidence of forestomach hyperplasia in the exposed males
and high dose females, increased incidence of nasal inflammation at the high-dose animals, slightly
increased incidence of acute hepatic necrosis in the high-dose males and increased incidence of
inflammation of the cecum in exposed males. Statistical analyses, however, were not performed on these
histopathologic changes.
NTP (1993) also administered to B6C3F1 mice (60/sex/group) daily oral gavage doses of 0, 5, or
10 mg/kg mercuric chloride (0, 3.7 and 7.4 mg Hg/kg-day), 5 days a week, by gavage for 2 years.
Survival and body weights of mice were slightly lower in mercuric chloride treated mice compared to
controls. Absolute kidney weights were significantly increased in the treated males while relative kidney
weights were significantly increased in high-dose males and both low- and high-dose females.
3-37
-------
Histopathology revealed an increase in the incidence and severity of nephropathy in exposed males (mild
severity in low dose and moderate-to-marked severity in high dose) and females (minimal severity in low
dose and minimal-to-mild severity in high dose). Nephropathy was defined as foci of proximal
convoluted tubules with thickened basement membrane and basophilic cells with scant cytoplasm. Some
affected convoluted tubules contained hyaline casts. There was also an increase in nasal cavity
inflammation (primarily infiltration of granulocytes in nasal mucosa) in the exposed animals.
In a 4-week oral study (Jonker et al. 1993), Wistar rats (5-10/sex/group) were fed a diet
containing 15 and 120 ppm mercuric chloride (0.56 or 4.4 mg Hg/kg-day). A significant increase in
relative kidney weight was reported for the low-dose females and high-dose males. There was also an
increase in the incidence of high-dose males that had occasional basophilic tubules in the outer cortex of
the kidneys. In the range-finding study by Jonker et al. (1993), rats were administered 75, 150, or 300
ppm mercuric chloride (2.8,5.6, 11.1 mg Hg/kg-day) in the diet for four weeks. A significant increase in
the relative kidney weights was observed in both sexes for all dose groups; the effect was dose related.
Nephrosis and proteinaceous casts in the kidneys were reported in both sexes at the lowest dose. At 5.6
mg/kg-day, the body weight was significantly decreased in males and serum alkaline phosphatase levels
were elevated in females. At 11.1 mg/kg-day, increased serum aspartate aminotransferase (both sexes),
decreased urinary density (males), increased relative adrenal weight (males), increased serum sodium and
phosphate levels (females) and decrease in body weight (females) were reported.
A series of studies (Boscolo et al. 1989; Carmignani et al. 1989, 1992) reported renal and
cardiovascular changes in rats exposed to mercuric chloride in drinking water. These studies were
limited due to the small number of animals and dose levels tested. Boscolo et al. (1989) evaluated the
renal effects of mercuric chloride in two different rat strains. Male Sprague-Dawley rats (8/group) were
administered 0 or 0.05 mg/mL mercury (0 or 7 mg Hg/kg-day), and male Wistar rats (8/group) received
0, 0.05, or 0.2 mg/mL mercury (0, 7, or 28 mg Hg/kg-day) in drinking water for 350 days. Increases in
blood pressure and cardiac inotropism, without changes in heart rate, occurred in exposed rats of both
strains. Hydropic degeneration and desquamation of the proximal tubular cells were exhibited in kidneys
of Sprague-Dawley rats, with alterations and lysis of lysosomes in tubular cells and thickening of the
basal membrane in the glomeruli. Wistar rats displayed tubular degeneration and membranous
glomerulonephritis in 30% of the glomeruli at 7 mg/kg-day and all glomeruli at 28 mg/kg-day.
Thickening of basal membrane and hypercellularity and alteration of the mesangial matrix in the
glomeruli and hydropic degeneration of tubules were seen in Wistar rats. Similar findings of renal
histopathology alterations and cardiovascular changes were reported by Carmignani et al. (1989) who
administered 0 or 0.05 mg/L of mercury (7 mg/kg-day) to male Sprague-Dawley rats (8/group) for 350
days.
In Carmignani et al. (1992), male Sprague-Dawley rats (8/group) received 0 or 0.2 mg/mL of
mercury (28 mg Hg/kg-day) as mercuric chloride in drinking water for a shorter duration (180 days).
Similar renal changes were observed, as well as IgM deposition in the glomeruli (as shown by
immunofluorescence). In addition, the treated group displayed significantly decreased urinary kallikrein
and creatinine, decreased plasma renin and increased plasma angiotensin-converting enzyme. The
cardiovascular effects were slightly different from Boscolo et al. (1989) and Carmignani et al. (1989);
there was an increase in blood pressure but a decrease in cardiac inotropism in the exposed rats. The
increase in blood pressure was suggested to be due to a vasoconstrictor effect, likely related to a greater
release of noradrenaline from adrenergic neurons and to baroreflex hyposensitivity. The decrease in
contractility was attributed to a direct toxic effect of the mercury on the cardiac muscle because of the
high levels of mercury detected in the heart. The differences in the results of cardiovascular changes for
the studies were not explained.
3-38
-------
To evaluate the effect of mercuric chloride on the development of autoimmunity, female SJL/N
mice (7/group) received 0.625, 1.25, 2.5, or 5 ppm mercuric chloride (0.07, 0.14, 0.28, or 0.56 mg Hg/kg-
day) in drinking water ad libitum for 10 weeks (Hultman and Enestrom 1992). An increase in circulating
antinucleolar antibodies was observed at 0.28 mg Hg/kg-day. The high-dose group had elevated granular
IgG deposits in the renal mesangium and in vessel walls of glomerular capillaries, arteries and arterioles
of the spleen and in intramyocardial arteries. Slight glomerular cell hyperplasia and discrete widening of
the centrolobular zone were also exhibited in the 0.56-mg/kg-day group.
Agrawal and Chansouria (1989) administered 0, 2.6, 5.2, and 10.4 mg Hg/kg-day as mercuric
chloride in drinking water to male Charles Foster rats for 60, 120, or 180 days (5/group). The relative
adrenal gland weight was significantly increased for the dose groups at all durations compared to
controls. Significant increases in adrenal and plasma corticosterone levels occurred in all dose groups at
60 and 120 days; however, changes were not seen after 180 days. The authors suggested that mercuric
chloride may have acted as a chemical stressor in a dose- and duration-dependent manner. The study was
limited because histopathology was not performed on the kidneys, and the adrenal gland was the only
tissue evaluated.
Both male and female Brown Norway rats 7-9 weeks of age were divided into groups of 6-20
animals each (Druet et al. 1978). The numbers of each sex were not stated. The animals were injected
subcutaneously with mercuric chloride 3 times weekly, for 8 weeks, with doses of 0, 0.07, 0.2, 0.4, 0.7,
and 1.5 mg Hg/kg. An additional group was injected with a 0.04 mg/kg for 12 weeks. Antibody
formation was measured by the use of kidney cryostat sections stained with a fluoresceinated sheep anti-
rat IgG antiserum; urinary protein was assessed by the biuret method. Tubular lesions were seen at the
higher dose levels. Proteinuria was seen at doses of 0.07 mg/kg and above, but not at 0.04 mg/kg.
Proteinuria was considered a highly deleterious effect in that affected animals developed
hypoalbuminemia and many died. Fixation of IgG antiserum was detected in all groups except controls.
3.2.2 Cancer Data
3.2.2.1 Human data
No data are available on the carcinogenic effects of inorganic mercury in humans.
3.2.2.2 Animal data
The results from a dietary study in rats and mice show equivocal evidence for carcinogenic
activity in male mice and female rats and some evidence for carcinogenic activity in male rats. Two
other dietary studies show negative evidence for carcinogenicity, but these studies are limited by
inadequacies in the data and experimental design.
Mercuric chloride was administered by gavage in water at doses of 0, 2.5, or 5 mg/kg-day (0, 1.9
and 3.7 mg Hg/kg-day) to Fischer 344 rats (60/sex/group), 5 days a week, for over 104 weeks (NTP
1993). An interim sacrifice (10/sex/dose) was conducted after 15 months of exposure. Complete
histopathological examinations were performed on all animals found dead, killed in extremis, or killed by
design. Survival after 24 months was statistically significantly (p<0.01) lower in low- and high-dose
males; survival was 43%, 17% and 8% in control, low-, and high-dose males, respectively, and 58%,
47%, and 50% in control, low-, and high-dose females, respectively. During the second year of the
study, body weight gains of low- and high-dose males were 91% and 85% of controls, respectively, and
body weight gains of low- and high-dose females were 90% and 86% of controls, respectively. At study
termination, nephropathy was evident in almost all male and female rats including controls, but the
3-39
-------
severity was much greater in treated males; the incidence of "marked" nephropathy was 6/50, 29/50, and
29/50 in control, low- and high-dose males, respectively.
Squamous cell papillomas of the forestomach showed a statistically significant (p<0.001)
positive trend with dose by life table adjusted analysis; the incidences were 0/50, 3/50 and 12/50 in
control, low-, and high-dose males, respectively. The incidence in female rats was 0/50, 0/49 and 2/50 in
control, low- and high-dose groups, respectively. These neoplasms are rare neoplasms in rats and
occurred in only 1 out of 264 historical controls. The incidence of papillary hyperplasia of the stratified
squamous epithelium lining of the forestomach was statistically significantly (p<0.01) elevated in all
dosed males (3/49, 16/50 and 35/50 in control, low- and high-dose males, respectively) and in high-dose
females (5/50, 5/49 and 20/50 in control, low- and high-dose females, respectively). The incidence of
thyroid follicular cell carcinomas was marginally significantly (p=0.044 by logistic regression analysis;
tumors not considered to be fatal) increased in high-dose males (1/50, 2/50 and 6/50 in control, low- and
high-dose groups, respectively). The data, adjusted for survival, also showed a significant (p=0.017)
positive trend in males. The combined incidence of thyroid follicular cell neoplasms (adenoma and/or
carcinoma), however, was not significantly increased (2/50, 6/50 and 6/50 in control, low- and high-dose
males, respectively). In female rats a significant decrease in the incidence of mammary gland
fibroadenomas was observed (15/50, 5/48 and 2/50 in control, low- and high-dose females, respectively).
Table 3-31 gives the incidences of lesions which were increased in treated animals.
The high mortality in both groups of treated males indicates that the maximally tolerated dose
(MTD) was exceeded in these groups and limits the interpretation of the study. NTP (1993) considered
the forestomach tumors to be of limited relevance to humans because the tumors did not appear to
progress to malignancy. NTP (1993) also questioned the relevance of the thyroid carcinomas because
these neoplasms are usually seen in conjunction with increased incidences of hyperplasia and adenomas,
but increases in hyperplasia (2/50, 4/50 and 2/50 in control, low- and high-dose males, respectively) or
adenomas (1/50, 4/50 and 0/50 in control, low- and high-dose males, respectively) were not observed.
In the same study, mercuric chloride was administered by gavage in water at doses of 0, 5, or
10 mg/kg-day (0, 3.7 and 7.4 mg Hg/kg-day), 5 days a week, for 104 weeks to B6C3F1 mice
(60/sex/group) (NTP 1993). An interim sacrifice (10/sex/dose) was conducted after 15 months of
exposure. Terminal survival of male mice was not affected by the administration of mercuric chloride;
survival of high-dose females was slightly lower (p=0.051) than controls (41/60, 35/60 and 31/60 in
control, low- and high-dose females, respectively). Body weight gain was not affected. Female mice
exhibited a significant increase in the incidence of nephropathy (21/49, 43/50 and 42/50 in control, low-
and high-dose females, respectively). Nephropathy was observed in 80-90% of the males in all groups.
The severity of nephropathy increased with increasing dose (1.08, 1.74 and 2.51 in control, low- and
high-dose males, respectively; 0.47, 1.02 and 1.24 in control, low- and high-dose females, respectively).
The incidence of renal tubule hyperplasia was 1/50, 0/50 and 2/49 in control, low- and high-dose males.
3-40
-------
Table 3-31
Incidence" of Selected Lesions in Rats in the NTP (1993) 2-Year Gavage Study
Tumor Site and Type
Dose Group (mg Hg/kg-day)
Males
0
1.9
3.7
Females
0
1.9
3.7
Forestomach
Papillary hyperplasia
Squamous cell papilloma
3/49
0/50
16/50b
3/50
35/50b
12/50C
5/50
0/50
5/49
0/49
20/50b
2/50
Thyroid Follicular Celld
Adenoma
Carcinoma
Adenoma or carcinoma
1/50
1/50
2/50
4/50
2/50
6/50
0/50
6/50'
6/50
-
-
-
-
-
-
-
-
-
* Overall rate
bp<0.01
c p <0.001; trend test also p<0.001
d Data on thyroid follicular cell lesions were reported for males only.
' p = 0.044, logistic regression
As shown in Table 3-32, the combined incidence of renal tubule adenomas and adenocarcinomas
was 0/50, 0/50 and 3/49 in control, low- and high-dose males, respectively. Although no tumors were
seen in the low-dose group, a statistically significant (p=0.032) positive trend for increased incidence
with increased dose was observed. These observations were considered important because renal tubule
hyperplasia and tumors in mice are rare. The two-year historical incidence of renal tubule adenomas or
adenocarcinomas in male mice dosed by gavage with water was 0/205, and only four of the nearly 400
completed NTP studies have shown increased renal tubule neoplasms in mice. NTP did not report a
statistical comparison of the study data to historical control data. Analysis of the reported data with
Fisher's Exact test, however, showed that the incidence of renal tubule adenomas or adenomas and
carcinomas (combined) in the high-dose males was significantly elevated when compared to historical
controls (Rice and Knauf 1994).
3-41
-------
Table 3-32
Incidence" of Renal Tubule Tumors in Male Mice in the
NTP (1993) 2-Year Gavage Study
Adenoma
Adenomacarcinoma
Adenoma or adenomacarcinoma
Dose Group (mg/kg-day)
0
0/50
0/50
0/50
5
0/50
0/50
0/50
10
2/49
1/49
3/49b
a Overall rate
b p = 0.107; trend test p = 0.032
A 2-year feeding study in rats (20 or 24/sex/group; strain not specified) was conducted in which
mercuric acetate was administered in the diet at doses of 0, 0.5, 2.5, 10, 40, and 160 ppm (0, 0.02, 0.1,
0.4, 1.7, and 6.9 mg Hg/kg-day) (Fitzhugh et al. 1950). Survival was not adversely affected in the study.
Increases in kidney weight and renal tubular lesions were observed at the two highest doses. No
statement was made in the study regarding carcinogenicity. This study was not intended as a
carcinogenicity assay, and the number of animals/dose was rather small. Histopathological analyses
were conducted on only 50% of the animals (complete histopathological analyses were conducted on
only 31% of the animals examined), and no quantitation of results or statistical analysis was performed.
No increase in tumor incidence was observed in a carcinogenicity study using white Swiss mice
(Schroeder and Mitchener 1975). Groups of mice (54/sex/group) were exposed until death to mercuric
chloride in drinking water at 5 ppm mercury (0.95 mg Hg/kg-day). No effects on survival or body
weights were observed. After dying, mice were weighed, dissected, gross tumors were detected, and
some sections were made of the heart, lung, liver, kidney and spleen for microscopic examination.
Mercuric chloride was nontoxic in the study. No statistically significant differences were observed in
tumor incidences for treated animals and controls. This study is limited because complete histological
examinations were not performed, only a single dose was tested, and the MTD was not achieved.
The increasing trend for renal tubular cell tumors in mice observed in the NTP (1993) study is
supported by similar findings in mice after chronic dietary exposure to methylmercury (Hirano et al.
1986; Mitsumori et al. 1981, 1990). In these studies, dietary exposure to methylmercuric chloride
resulted in increases in renal tubular tumors at doses where substantial nephrotoxicity was observed.
3-42
-------
Table 3-33
Carcinogenic Effects of Inorganic Mercury in Animals: Oral Exposure
Species/
Strain/
No. per Sex
per Group
Rat/strain
NS/20-24 M,
20-24 F
Rat/F344/
60 M, 60 F
Mouse/Swiss/54
M, 54F
Mouse/
B6C3F1/
60 M, 60 F
Exposure
Duration
2yr
ad lib
2yr
5d/wk
Ix/d
(gavage)
Lifetime
ad lib
2yr
5d/wk
Ix/d
(gavage)
Dose
(mg/kg-day)
0,0.02,0.1,
0.4, 1.7, 6.9"
0, 1.9,3.7
(HgCl2)
0.95
(HgCl2)
0,3.7,7.4
(HgCl2)
Effects/Limitations//BML
No carcinogenicity reported
Limitations: small number of animals/dose; complete
histopathological examinations conducted on only 31% of
animals; no statistical analyses
Thyroid follicular cell carcinomas in males at 3.7.
Limitations: MTD exceeded (high mortality in treated
males); limited relevance of lesions to humans
No carcinogenicity reported
Limitations: Complete histopathological examinations not
performed; only single dose level tested; MTD not
achieved
Renal tubule tumors (adenoma or adenomacarcinoma) in
3/49 males at 7.4 (positive trend test, p=0.0032)
Limitations: Severe nephropathy also observed in high-
dose males.
Reference
Fitzhughetal. 1950
NTP 1993
Schroeder and
Mitchener 1975
NTP 1993
' Phenylmercuric acetate and mercuric acetate
3.2.3 Other Data
3.2.3.1 Death
The estimated lethal dose of inorganic mercury for a 70 kg adult is 10-42 mg Hg/kg (Gleason et
al. 1957). Most deaths attributed to inorganic mercury occur soon after a person ingests a single large
amount of mercury. Causes of death include cardiovascular failure, gastrointestinal damage and acute
renal failure (Troen et al. 1951).
Table 3-34
Lethality of Inorganic Mercury in Humans: Case Study
Species/
No. per
Sex
Human/25 M,
29 F
Exposure
Duration
Once
Dose
(mg/kg-day)
21-37 (est.)
(HgCl2)
Effects/Limitations/BML
Case studies of mercuric chloride poisonings in victims age
2-60 yr; 9 resulted in death (all adults).
BML not reported
Reference
Troen etal. 1951
3-43
-------
The estimated LD50 for rats following oral exposure to mercuric chloride is 25.9 mg Hg/kg;
however, LD50 levels as high as 77.7 mg Hg/kg have been observed in rats (Kostial et al. 1978). Male
rats appear to be more sensitive to the effects of mercuric chloride. This was demonstrated in a chronic-
duration oral study with rats, in which 40/50 males and 21/49 females died at the low dose, 45/50 males
and 20/50 females died at the high dose, compared to 24/50 males and 15/50 females in the controls
(Dieter et al. 1992; NTP 1993). The increase in deaths in the male rats was statistically significant and
were considered to be due to renal lesions. Mortality incidence was not significantly increased in
exposed female groups.
Table 3-35
Lethality of Inorganic Mercury in Animals: Oral Exposure
Species/
Strain/
No. per Sex
per Group
Rat/Albino
(NS)/6 NS
Rat/F344/5 M,
5F
Rat/F344/
50 M, 50 F
Mouse/
B6C3F/5 M, 5
F
Exposure
Duration
Once
(gavage)
14 d
5d/wk
Ix/d
(gavage)
2yr
5d/wk
Ix/d
(gavage)
14 d
5d/wk
Ix/d
(gavage)
Dose
(mg/kg-day)
NS
(6 levels)
(HgCl2)
0,0.93, 1.9,
3.7, 7.4, 14.8
(HgCl2)
0, 1.9, 3.7
(HgCl2)
0, 3.7, 7.4,
14.8, 29, 59
(HgCl2)
Effects/Limitations/BML
LD50 = 25.8 mg/kg for 2-week old pups; older rats had
higher LD50 values.
Limitation: Incomplete data reporting (i.e., doses not
reported, toxic effects not specified)
BML not reported
2/5 males died at 14.8; no other animals died.
BML: 45.4 ug/g in kidney of males, 43.3 ug/g in kidney
of females
40/50 males died at 1.9 mg/kg, vs. 24/50 control males;
survival of dosed females was not significantly different
from controls.
BML not reported
9/10 died (LOAEL = 59).
Limitations: small number of animals
BML not reported
Reference
Kostial etal. 1978
Dieter et al. 1992
Dieter et al. 1992
NTP 1993
3.2.3.2 Neurological
Limited studies are available concerning neurological toxicity following oral exposure to
inorganic mercury. These studies are summarized below.
3-44
-------
Table 3-36
Neurotoxicity of Inorganic Mercury in Humans: Case Studies
Species/
No. per Sex
Human/2 F
Human/2 M (1
child), 1 adult F
Exposure
Duration
6-25 yr
3 mo
Dose
(mg/kg-day)
0.73
(Hg2Cl2)
NS
(Hg2Cl2)
(HgS)
Effects/Limitations/BML
Dementia, irritability, decreased cerebellar neurons, low
brain weight
Limitation: Case study
BML: 3.4-4.7 ug/g in frontal cortex
Drooling, dysphagia, irregular arm movements, impaired
gait, convulsions following ingestion of patent medicines
containing mercuric sulfide and mercurous chloride
Limitation: Case studies; concomitant exposure to other
metals; limited exposure data
BML: 39-2800 ug/L in 24 hr urine
Reference
Davis etal. 1974
Kang-Yum and
Oransky 1992
There are several animal studies in which inorganic mercury-induced neurotoxicity has been
reported.
Table 3-37
Neurotoxicity of Inorganic Mercury in Animals: Oral Exposure
Species/
Strain/
No. per Sex
per Group
Rat/
Holtzman/
8 M exposed,
8 M control
Rat/Sprague
Dawley/12 F
exposed, 10 F
controls
Mouse/C57BL
6J/NS
Exposure
Duration
11 wk
3 mo ad lib in
feed
17 mo
ad lib in
drinking
water
Dose
(mg/kg-day)
0, 0.74
(HgCl2)
0,2.2
(Hgcy
0.74 for 110 d,
then 7.4-14.8
for 400 d; 2.2
for 17 mo
(HgCl2)
Effects/Limitations/BML
Weakening of hind legs, crossing reflex of limbs, ataxia;
degenerative changes in neurons of dorsal root ganglia and
Purkinje and granule cells of cerebellum
Limitation: One dose level tested
BML not reported
Inactivity and abnormal gait
Limitation: One dose level tested
BML not reported
No clinical signs of neurotoxicity; no effect on optic or
peripheral nerve structure
Limitation: Lack of statistical analyses due to insufficient
number of animals tested; uncertainty of dosage due to
large variation in water consumption
BML not reported
Reference
Chang and
Hartmann 1972
Goldman and
Blackburn 1979
Ganser and
Kirschner 1985
3.2.3.3 Renal
The kidney appears to be the critical target organ for the effects of acute ingestion of inorganic
mercury. Case studies of poisonings by mercuric chloride report acute renal failure, including
3-45
-------
proteinuria, oliguria and hematuria, in people ingesting estimated doses of 3.5-37 mg Hg/kg (Afonso and
deAlvarez 1960; Pesce et al. 1977; Troen et al. 1951). These effects are attributed to tubular and
glomerular pathology.
Table 3-38
Renal Toxicity of Inorganic Mercury in Humans: Case Studies
Species/
No. per Sex
Human/25 M,
29 F
Human/ 1 F
(adult)
Human/ 1 M
Exposure
Duration
Once
Once
(tablet)
Once
Dose
(mg/kg-day)
3.5-37(est.)
(HgCl2)
30
(HgCl2)
21.4
(HgCl2)
Effects/Limitations/BML
Case studies of mercuric chloride poisonings in victims
age 2-60 yr; 18 cases resulted in renal effects
(albuminuria, anuria)
BML not reported
Oliguria; proteinuria; hematuria following ingestion of
mercuric chloride
Limitation: Case study
BML not reported
Proteinuria indicating glomerular and tubular damage
Limitation: Case study
BML: Avg 370 ug/L in blood
Reference
Troen etal. 1951
Afonso and
deAlvarez 1960
Pesce etal. 1977
There are numerous animal studies reporting kidney damage in rats and mice ingesting inorganic
mercury. Acute exposures result in increased kidney weight with at least 0.46 mg Hg/kg-day and tubular
necrosis at higher doses; males appear to have greater sensitivity for the histological changes than
females (Fowler 1972; NTP 1993). Similarly, longer-term studies have found histopathologic effects
affecting the tubules and glomeruli, including thickening of basement membranes and degeneration of
tubular cells (Carmignani et al. 1989; Jonker et al. 1993; NTP 1993). A study monitoring kidney
function reported ketonuria and proteinaceous casts (Jonker et al. 1993).
Table 3-39
Renal Toxicity of Inorganic Mercury in Animals: Oral Exposure
Species/
Strain/
No. per Sex
per Group
Rat/Sprague
Dawley/8 M
Exposure
Duration
350 d
ad lib in
drinking
water
Dose
(mg/kg-day)
0,7
(HgCl2)
Effects/Limitations/BML
Hydropic degeneration of tubular cells
Limitation: Only one dose tested
BML: 140 ug/g in kidney
Reference
Carmignani et al.
1989
3-46
-------
Table 3-39 (continued)
Renal Toxicity of Inorganic Mercury in Animals: Oral Exposure
Species/
Strain/
No. per Sex
per Group
Rat/Sprague-
Dawley/8 M
Rat/Wistar/5 M, 5
F exposed/10 M,
10 F controls
Rat/Wistar/
5M, 5F
exposed/10 M, 10
F controls
Rat/F344/
5M, 5F
Rat/F344/
10 M, 10 F
Rat/F344/
60 M, 60 F
Mouse/NMRI/20
(sexNS)
exposed/10
controls
Mouse/NMRI/24
(sexNS)
Mouse/BeCSF/
5M, 5F
Mouse/BeCSF/
10 M, 10 F
Exposure
Duration
180 d
ad lib in
drinking
water
4wk
ad lib in feed
4wk
ad lib in feed
14 d
5d/wk
Ix/d
(gavage)
6 mo
5d/wk
Ix/d
(gavage)
2yr
5d/wk
Ix/d
(gavage)
Once
(gavage)
Once
(gavage)
14 d
5d/wk
Ix/d
(gavage)
6 mo
5d/wk
Ix/d
(gavage)
Dose
(mg/kg-day)
0,28
(HgCl2)
0, 0.56, 4.4
(HgCl2)
2.8,5.6, 11.1
(HgCl2)
0,0.93, 1.9,
3.7, 7.4, 14.8
(HgCl2)
0, 0.23, 0.46,
0 93 1837
(HgCl2)
0, 1.9,3.7
(HgCl2)
0, 5, 10, 20, 40
(HgCl2)
0,20
(HgCl2)
0, 3.7, 7.4,
14.8, 29, 59
(HgCl2)
0,0.93, 1.9,
3.7, 7.4, 14.8
(HgCl2)
Effects/Limitations/BML
Hydropic degeneration of tubular cells, IgM deposition
in glomeruli, decreased urinary kallikrein and
creatinine, decreased plasma renin, increased plasma
angiotensin-converting enzyme
Limitation: Only one dose tested
BML: 0.94 ug/g in blood
Ketonuria in males at 4.4 mg/k-day; increased kidney
weight in males and females (LOAEL = 0.56 in
females, 4.4 in males)
BML not reported
Ketonuria in males at all levels; increased relative
kidney weight, increased nephrosis and proteinaceous
casts in males and females (LOAEL = 2.8)
BML not reported
Increased absolute and relative kidney weight in males
and females (LOAEL = 1.9); acute renal tubule necrosis
at >3.7 mg/kg-day in both sexes
BML: 43-46 ug/g in kidney at 14.8 mg/kg
Increased absolute and relative kidney weight in males
and females (LOAEL = 0.46); increased severity of
nephropathy in males (LOAEL = 0.93)
BML: 86.2-89.6 ug/g in kidney at 0.93 mg/kg
Increased severity of nephropathy in males (thickening
of glomerular and tubular basement membranes;
degeneration and atrophy of tubule epithelium)
(LOAEL =1.9)
BML not reported
Decreased selenium-dependent glutathione peroxidase
activity in kidney; minor renal tubular damage (LOAEL
= 10)
BML: 260 ug/L in blood at 10 mg/kg
Necrosis of proximal tubules
BML not reported
Increased absolute and relative kidney weight (LOAEL
= 3.7); acute renal tubular necrosis at 29 mg/kg-day in
males and at 59 mg/kg-day in males and females
BML: 1 16-171 ug/g in kidney at 29 mg/kg-day
Increased absolute and relative kidney weight of males
(LOAEL = 3.7); Cytoplasmic vacuolation of tubule
epithelium in males (LOAEL = 3.7)
BML: 36.1-40.6 ug/g in kidney at 3.7 mg/kg-day
Reference
Carmignani et al.
1992
Jonkeretal. 1993
Jonkeretal. 1993
NTP 1993
NTP 1993
NTP 1993
Nielsen etal. 1991
Nielsen etal. 1991
NTP 1993
NTP 1993
3-47
-------
Table 3-39 (continued)
Renal Toxicity of Inorganic Mercury in Animals: Oral Exposure
Species/
Strain/
No. per Sex
per Group
MOUS6/B6C3F/
60 M, 60 F
Exposure
Duration
2yr
5d/wk
Ix/d
(gavage)
Dose
(mg/kg-day)
0, 3.7, 7.4
(HgCl2)
Effects/Limitations/BML
Increased severity of nephropathy (foci of proximal
tubules with thickened basement membrane; basophilic
cells with scant cytoplasm (LOAEL = 3.7)
BML not reported
Reference
NTP 1993
Bernaudin et al. (1981) exposed male and female Brown Norway rats (number not specified) to
mercuric chloride via aerosols (4 hours/week) and intratracheal instillation for 2 months. The aerosol
exposures resulted in a retention of 0.05-0.06 mg HgCl2/kg/hour (based on radiolabeled mercury); the
parameters of the aerosol were not well characterized (e.g., no mass median aerodynamic diameter
(MMAD) or geometric standard deviation provided and the particle generation system was not
adequately described). The autoimmune response was typified by a linear pattern of IgG conjugate
fixation in kidney glomeruli and a granular pattern of fixation in kidney glomeruli and arteries, lung and
spleen; evidence of autoimmune disease was noted at all but the lowest intratracheal exposure level (60
(ig HgQ2/kg/week). In two of three rats exposed to aerosols and examined when sacrificed, weak
proteinuria (1, 28 and 47 mg/day) was detected. No significant proteinuria was observed in the animals
administered mercuric chloride by intratracheal instillations.
Table 3-40
Renal Toxicity of Inorganic Mercury in Animals: Inhalation Exposure
Species/
Strain/
No. per Sex
per Group
Rat/Brown
Norway/ 3-8 both
sexes
Rat/Brown
Norway/5 both
sexes
Exposure
Duration
2 mo, Ix/wk
(intra-
tracheal)
2 mo,
4 d/wk,
Ihr/d
(aerosol)
Dose
(mg/m3)
0,6, 11,47,
79 mg/kg-day
(HgCl,)
1
(HgCl2/m3)
(estimate of
minimum air
concentration)
Effects/Limitations/BML
Autoimmune effect in spleen at 6 mg/kg-day and in
spleen, lung and kidney at higher doses
BML not reported
Weak proteinuria; autoimmune effect in kidney, lung
and spleen
BML not reported
Reference
Bernaudin etal.
1981
Bernaudin etal.
1981
3.2.3.4 Cardiovascular
No studies were located regarding the cardiovascular toxicity of inorganic mercury in humans
following oral exposure.
3-48
-------
Limited information was located regarding the cardiovascular toxicity of inorganic mercury
following oral exposure in animals. Signs of cardiovascular toxicity in rats include increased blood
pressure and varying changes in the contractility of the heart (Carmignani et al. 1989, 1992). These signs
manifested after oral exposure to mercuric chloride in drinking water for 180 or 350 days. No other
animal studies were located.
Table 3-41
Cardiovascular Toxicity of Inorganic Mercury in Animals: Oral Exposure
Species/
Strain/
No. per Sex
per Group
Rat/Sprague
Dawley/8 M
Rat/Wistar/8 M
Exposure
Duration
350 d
ad lib in
drinking
water
180 d
ad lib in
drinking
water
Dose
(mg/kg-day)
0,7
(HgCl2)
0,28
(Hgcy
Effects/Limitations/BML
Increased blood pressure; positive inotropic response
(p<0.05)
Limitation: Only one dose tested; small number of
animals
BML: 0.9ug/ginheart
Increased blood pressure (p<0.05); negative inotropic
response (not significant)
Limitation: Only one dose tested; small number of
animals
BML: 940 ug/L in blood, 4.1 ug/g in heart
Reference
Carmignani et al.
1989
Carmignani etal.
1992
3.2.3.5 Gastrointe stinal
Irritation of the gastrointestinal mucosa is a common outcome of mercury toxicity following
ingestion of mercuric chloride (Murphy et al. 1979). Ingestion of inorganic mercury may also cause
vomiting, nausea, severe abdominal pain and diarrhea (Afonso and deAlvarez 1960; Murphy et al. 1979).
No studies were located regarding the gastrointestinal toxicity of inorganic mercury after ingestion in
humans for intermediate or chronic durations.
Table 3-42
Gastrointestinal Toxicity of Inorganic Mercury in Humans: Case Studies
Species/
No. per Sex
Human/25 M,
29 F
Human/ 1 F
(adult)
Exposure
Duration
Once
Once
(tablets)
Dose
(mg/kg-day)
3.5-37(est.)
(HgCl2)
30
(Hgcy
Effects/Limitations/BML
Case studies of mercuric chloride poisonings in victims
age 2-60 yr; effects ranged from nausea to severe corrosive
gastritis
Limitation: exposure data limited
BML not reported
Nausea; vomiting; abdominal cramps; diarrhea after
ingestion of mercuric chloride
Limitation: Case study
BML not reported
Reference
Troenetal. 1951
Afonso and
deAlvarez 1960
3-49
-------
Similar signs of gastrointestinal irritation appear in mice after intermediate duration oral
exposure to mercuric chloride (NTP 1993). Histopathologic analyses reveal inflammation and necrosis
of the stomach tissue. Further damage occurs to the gastrointestinal tract with continued dosing (NTP
1993). The incidence of hyperplasia of the forestomach epithelium increases in high-dose rats fed
mercuric chloride for two years (NTP 1993).
Table 3-43
Gastrointestinal Toxicity of Inorganic Mercury in Animals: Oral Exposure
Species/
Strain/
No. per Sex
per Group
Rat/F344/60
M, 60 F
Mouse/
B6C3F/5 M, 5
F
Exposure
Duration
2yr
5d/wk
Ix/d
(gavage)
14 d
5d/wk
Ix/d
(gavage)
Dose
(mg/kg-day)
0, 1.9,3.7
(HgCl2)
0, 3.7, 7.4,
14.8, 29, 59
(HgCl2)
Effects/Limitations/BML
Forestomach epithelial hyperplasia (LOAEL = 1.9 in
males, 3.7 in females).
BML not reported
Stomach inflammation and necrosis (LOAEL = 59).
BML: 116-171 ug/g in kidneys at 29 mg/kg-day
Limitation: small number of animals
Reference
NTP 1993
NTP 1993
3.2.3.6 Hepatic
Limited information is available regarding the hepatic toxicity of inorganic mercury in humans
after oral exposure. Murphy et al. (1979) reported on a man who died after ingesting an unspecified
amount of mercuric chloride. The man was jaundiced, and his liver enzymes were elevated prior to
death. Histopathological analyses revealed a softened and enlarged liver.
Liver enzymes were increased in rats and mice that ingested mercuric chloride for 4 or 6 weeks
(Dieter et al. 1983; Jonker et al. 1993; Rana and Boora 1992). In addition, liver weights were
significantly increased in mice after exposure to mercuric chloride in drinking water (Dieter et al. 1983).
No microscopic changes were seen, however, during the histopathological analyses. No other animal
data were available regarding oral exposure to inorganic mercury.
3-50
-------
Table 3-44
Hepatic Toxicity of Inorganic Mercury in Animals: Oral Exposure
Species/
Strain/
No. per Sex
per Group
Rat/Charles
Foster/5 M
Rat/Wistar/
5/sex
exposed/10/sex
controls
Mouse/
B6C3F/10 M
Exposure
Duration
30 d
Ix/d
feed
4wk
ad lib in feed
1-7 wk
ad lib in
drinking
water
Dose
(mg/kg-day)
NS
(HgCl2)
2.8,5.6, 11.1
(HgCl2)
0, 0.6, 2.9, 14.3
(HgCl2)
Effects/Limitations/BML
Increased lipid peroxidation (p<0.02)
Limitation: Only one dose (NS) tested
BML not reported
Increased serum alkaline phosphatase in males and females
(LOAEL = 5 .6 in females, 1 1 . 1 in males)
Limitation: small number of animals
BML not reported
Increased plasma cholinesterase (LOAEL = 2.9)
Limitation: small number of animals tested
BML: 0.6 ug/L in blood at 7 wk, at 2.9 mg/kg/d
Reference
Rana and Boora
1992
Jonkeretal. 1993
Dieter etal. 1983
3.2.3.7 Immunological
In addition to the inorganic mercury-induced autoimmune glomerulonephritis discussed earlier
(see discussion of renal effects in Section 3.2.1), several studies identified other immunotoxicity
endpoints in animals after oral exposure to inorganic mercury.
3.2.3.8 Developmental
No studies were located regarding the developmental toxicity of inorganic mercury in humans
after inhalation exposure.
The only information located regarding developmental toxicity in animals from inhalation
exposure to mercuric mercury comes from a study in which mice were exposed to aerosols containing
mercuric chloride during gestation (Selypes et al. 1984). Increases were observed in the incidence of
delayed ossification and dead or resorbed fetuses; the statistical significance of these effects was not
reported. In addition, at the highest concentration, a significant increase in weight retardation was also
observed. Interpretation of this study is limited, however, because the aerosols were not well
characterized, and it is not known to what extent the droplets were respirable or were cleared from the
upper respiratory tract and swallowed.
3-51
-------
Table 3-45
Immunotoxicity of Inorganic Mercury in Animals: Oral Exposure
Species/
Strain/
No. per Sex
per Group
Rat/Brown-
Norway/6 both
sexes
exposed/22
controls both
sexes
Mouse/
B6C3F/10 M
Mouse/SJL or
DBA/5 F
Mouse/SJL/7 F
Mouse/
B6C3F/5 M, 5
F
Exposure
Duration
2 mo
Ix/wk
(gavage)
7wk
ad lib in
drinking
water
2wk
ad lib in
drinking
water
10 wk
ad lib in
drinking
water
14 d
5d/wk
Ix/d
(gavage)
Dose
(mg/kg-day)
2.2
(HgCl2)
0, 0.6, 2.9, 14.3
(HgCl2)
0,0.7
(HgCl2)
0, 0.07, 0.14,
0.28, 0.56
(HgCl2)
0, 3.7, 7.4,
14.8, 29, 59
(HgCl2)
Effects/Limitations/BML
IgG deposits in glomerular capillary wall of kidney and
renal arteries, suggestive of autoimmune disease; similar
deposits also observed in lungs and spleen; no deposits
observed in controls
Limitation: Only one dose tested; small number of
animals tested
BML not reported
Suppression of lymphoproliferative response to T-cell,
concavalin A and phytohemagglutinin (LOAEL = 2.9
mg/kg-day; p<0.05)
BML: 600 ug/L in blood at 2.9 mg/kg-day
Increased lymphoproliferative response to concanavalin A
and E. coli lipopolysaccharide (p<0.02)
Limitations: only one dose tested; small number of
animals tested
BML not reported
Increased antinucleolar antibodies in IgG class (LOAEL =
0.28, p<0.05)
Limitation: small number of animals tested
BML: 5 .2 ug/g in kidney
Decreased thymus weight (LOAEL = 14.8)
Limitation: small number of animals tested
BML: 1 16-171 ug/g in kidney at 29 mg/kg-day
Reference
Bernaudin etal.
1981
Dieter etal. 1983
Hultman and
Johansson 1991
Hultman and
Enestrom 1992
NTP 1993
Table 3-46
Developmental Toxicity of Inorganic Mercury in Animals: Inhalation Exposure
Species/
Strain/
No. per Sex
per Group
Mice/CFLP/
No. F NS
Exposure
Duration
4d
4hr/d
Gd 9-12
Dose
(mg/m3)
0,0.17, 1.6
(Hgcy
Effects/Limitations/BML
Increased dead or resorbed fetuses; delayed ossification
(LOAEL = 0.1 7)
Limitations: Data were reported as number of embryos only,
not as number of affected litters; no statistical analysis; aerosol
exposure was not well characterized; maternal toxicity was not
evaluated
BML not reported
Reference
Selypesetal. 1984
3-52
-------
Developmental effects have been reported in animals following oral exposure to inorganic
mercury. These efforts include an increased incidence of abnormal fetuses in hamsters (Gale 1974),
growth retardation in rats (Rizzo and Furst 1972) and decreased body weights in several rat studies.
Gale (1974) administered 0, 4, 8, 25, 35, 50, 75, or 100 mg mercuric acetate/kg (0, 2.5, 5, 16, 22,
32, 47, or 63 mg Hg/kg) to pregnant golden hamsters (10/exposed group; 3/control group) by gavage in
distilled water on the 8th day of gestation. The pregnant animals were sacrificed on gestation day 12 or
14, and the uterine contents were examined. A statistically significant increase in the incidence of
abnormal fetuses (combined incidence of small, retarded, edematous, and/or malformed fetuses) was
observed at 16 mg Hg/kg. Statistically significant increases in the percentage of resorbed fetuses was
observed at 22 mg Hg/kg and in the percentages of small, retarded and edematous fetuses observed at 32
mg Hg/kg. No treatment-related effects were observed on the fetuses at 5 mg Hg/kg. Toxic effects
observed in maternal animals included weight loss, diarrhea, slight tremor, somnolence, tubular necrosis
in the kidneys and cytoplasmic vacuolization of hepatocytes.
Rizzo and Furst (1972) administered ~1 mg Hg/kg as mercuric oxide to pregnant Long-Evans
rats (5/group) by gavage in peanut oil on gestation day 5, 12, or 19 in a pilot study. On gestation day 20
or 21, the rats were sacrificed, and the uterine contents were examined. Rats administered mercury on
gestation day 5 had a higher percentage of fetuses with growth retardation and inhibition of eye
formation (statistical significance not reported). Similar increases in these effects were not observed
after administration on gestation day 12 or 19. No toxicity in maternal animals was reported.
McAnulty et al. (1982) administered 8, 12, 16, or 24 mg mercuric chloride/kg-day (6, 9, 12, or 18
mg Hg/kg-day) by gavage to pregnant rats (strain and number not specified) on gestation days 6-15 as
reported in an abstract. The abstract did not report whether controls were used. Fetal and placental
weights were decreased at 9 mg Hg/kg-day and above. At 12 and 18 mg Hg/kg-day, increased
postimplantation losses were reported. These effects were attributed to maternal toxicity and decreased
food intake. At 18 mg Hg/kg-day, increases in delayed ossification and malformations were reported.
Statistical analyses were not reported.
Pritchard et al. (1982a) administered 4, 8, or 16 mg mercuric chloride/kg-day (3, 6, or 12 mg
Hg/kg-day) by the oral route to pregnant rats (number and strain not specified) from gestation day 15
until postpartum day 25, as reported in an abstract. The abstract did not state whether controls were
used. At 6 and 12 mg Hg/kg-day, pup weight was decreased on postpartum day 1. Subsequent weight
gain in these groups was also decreased. No other effects on development or behavior were observed
postpartum. Females at 6 and 12 mg Hg/kg-day had a decreased rate of weight gain, and gestation time
was slightly extended. Statistical analyses were not reported.
Pritchard et al. (1982b) administered 12, 16, or 24 mg mercuric chloride/kg-day (9, 12, 18 mg
Hg/kg-day) to female rats (strain and number not reported) by gavage before mating and during gestation.
The abstract did not report whether controls were used. At 12 mg Hg/kg-day and above, females
exhibited weight loss and appeared unhealthy, estrous cycles became irregular, and high preimplantation
losses were observed. No effects on ovulation, estrous cycles, implantation, and fetal development were
observed at 9 mg Hg/kg-day. Statistical analyses were not reported.
3-53
-------
Table 3-47
Developmental Toxicity of Inorganic Mercury in Animals: Oral Exposure
Species/
Strain/
No. per Sex
per Group
Rat/Long-
Evans/5 F
Rat/Strain
NS/no. F NS
Rat/Strain
NS/no. F NS
Rat/Strain
NS/no. F NS
Hamster/
Golden/10 F
exposed/3 F
controls
Exposure
Duration
Once
Gd 5, 12, or
19
(gavage)
10 d
Ix/d
Gd6-15
(gavage)
Approx. 32 d
Ix/d
Gdl5-ppd25
(gavage)
Before
mating and
during
gestation
(gavage)
Once
Gd8
(gavage)
Dose
(mg/kg-day)
0,2.0
(HgO)
6, 9, 12, 18
(HgCl2)
3, 6, 12
(HgCl2)
9, 12, 18
(HgCl2)
0, 2.5, 5, 16, 22,
32, 47, 63
[Hg (CH3COO)2]
Effects/Limitations/BML
Growth retardation; inhibition of eye formation in group
treated on Gd 5, with some effect on Gd 12 group.
Limitations: No statistical analysis; small number of
litters in treated groups (and controls)
BML not reported
Decreased fetal and placental weights (LOAEL = 9);
malformations at 18.
Limitations: Reported as an abstract; few details reported
BML not reported
Decreased pup weight and weight gain (LOAEL = 6); no
effect in an unspecified developmental and behavioral
testing battery.
Limitations: Reported as an abstract; few details reported
BML not reported
High preimplantation loss (LOAEL = 12).
Limitations: Reported as an abstract; few details reported
BML not reported
Increased incidence of abnormal (small, retarded,
edematous, and/or malformed-exencephaly,
encephalocele, ectrodactyly, etc.) fetuses (LOAEL = 16,
p<0.05); maternal toxicity: weight loss, diarrhea, slight
tremors, somnolence, tubular necrosis, hepatocellular
necrosis.
Limitation: Small sample size; smaller control group;
insufficient detail about number of animals sacrificed at
Gd 12 or Gd 14; single day of treatment; incomplete
examinations reported (no visceral, only partial skeletal)
BML not reported
Reference
Rizzo and Furst
1972
McAnulty et al.
1982
Pritchard et al.
1982a
Pritchard etal.
1982b
Gale 1974
In addition to the oral and inhalation studies summarized above, several studies using other
routes of administration (i.p., s.c., i.v.) provide evidence of developmental toxicity associated with
exposure to mercury salts. These studies are summarized below.
Gale and Perm (1971) injected anesthetized pregnant golden hamsters (6-19/group)
intravenously with 0, 2, 3, or 4 mg mercuric acetate/kg (0, 1.3, 1.9, or 2.5 mg Hg/kg) on gestation day 8.
Controls were injected with vehicle (demineralized water). Maternal animals were sacrificed on
gestation day 12 or 14, and the uterine contents were examined. A significantly increased incidence of
resorptions was observed at all doses. In addition, increased incidences of retarded and edematous
fetuses were observed at all doses (statistical significance not reported). Toxic effects observed in
maternal animals included weight loss, diarrhea, slight tremor, somnolence and kidney lesions; however,
the report did not specify at which doses the maternal effects were observed.
3-54
-------
Gale (1974) compared the embryotoxicity of mercuric acetate administered by different routes in
pregnant golden hamsters (3-23/group). Subcutaneous administration of 0, 4, 8, 20, 35, or 50 mg
mercuric acetate/kg (0, 2.5, 5, 13, 22, or 32 mg Hg/kg) on gestation day 8 resulted in a significant
decrease in the percentage of normal embryos and a significant increase in the percentage of small
embryos at 2.5 mg Hg/kg. At 5 mg Hg/kg, significant increases in resorptions, abnormal, retarded,
edematous, and malformed fetuses were observed. Intraperitoneal administration of 0, 2, 4, or 8 mg
mercuric acetate (0, 1.3, 2.5, or 5 mg Hg/kg) on gestation day 8 resulted in significant increases in the
percentage of resorptions, abnormal, small, and edematous fetuses at 1.3 mg Hg/kg. Intravenous
administration of 0 or 4 mg mercuric acetate/kg (0 or 2.5 mg Hg/kg) on gestation day 8 resulted in
significant increases in resorptions, abnormal, small, retarded, edematous, and malformed fetuses at 2.5
mg Hg/kg. Comparison of the extent of the developmental toxicity demonstrated an effect of route of
administration: i.p. > i.v. > s.c. > oral.
Gale (1981) compared the embryotoxicity of mercuric acetate in 6 strains of hamsters
(LAK:LVG[SYR], CB/SsLak, LHC/Lak, LSH/SsLak, MHA/SsLak. PD4/Lak). Maternal animals of the
various strains (3-9/group) were injected subcutaneously with 0 or 15 mg/kg mercuric acetate (0 or 9.5
mg Hg/kg) on gestation day 8. Controls were injected with demineralized distilled water. Maternal
animals were sacrificed on gestation day 12 or 15, and the uterine contents were examined. The
percentage of resorptions was significantly increased in all strains examined on days 12 and 15. Strain-
specific variations were observed in the incidences of abnormal, edematous and retarded fetuses and in
the incidences of ventral wall defects, distension of the pericardial cavity, cleft palate, hydrocephalus and
cardiac abnormalities. Maternal toxicity was not described.
Kavlock et al. (1993) injected pregnant Sprague-Dawley rats (6-25/group) subcutaneously with
0, 1, 2, 3, or 4 mg mercuric chloride/kg (0, 0.7, 1.5, 2.2, or 3.0 mg Hg/kg) on gestation day 7, 9, 11, or 13.
On gestation day 21, rats were sacrificed, and the uterine contents were examined. No increase in
malformations was observed in fetuses from mercuric chloride-treated dams. Exposure on gestation day
7 resulted in a significant decrease in fetal weight and an increase in the number of supernumerary ribs
(statistical significance not reported) at 2.2 mg Hg/kg. Exposure on gestation day 9 resulted in
significantly decreased live fetuses/litter and increased resorptions at 3 mg Hg/kg. Exposure on gestation
days 11 or 13 resulted in no significant differences in fetal parameters. Maternal toxicity (increased
mortality, decreased body weight, increased kidney weight, increased urine osmolality, and/or increased
serum urea or creatinine) were observed at 1.5 mg Hg/kg and above. No consistent correlations were
observed between maternal and fetal toxicity.
Kajiwara and Inouye (1986) injected Kud:ddY mice (10/group) intravenously with 0, 0.5, 1.0,
1.5, 2.0, and 2.5 mg Hg/kg as mercuric chloride on gestation day 0. Controls were injected with vehicle
(physiological saline). Maternal animals were sacrificed on gestation day 3.5, and the oviducts and
uterus were flushed to obtain preimplantation embryos. At 1.5 mg Hg/kg and above, the number of
abnormal embryos was significantly increased. Maternal animals at 1.5 mg Hg/kg and above exhibited a
decrease in body weight (statistical significance not determined). The study authors suggested that fetal
toxicity may have been related to maternal toxicity.
Kajiwara and Inouye (1992) injected Kud:ddY mice (5-15/group) intravenously with 0, 1, 2, or
2.5 mg Hg/kg as mercuric chloride on gestation day 0. Controls were injected with vehicle
(physiological saline). Maternal animals were sacrificed on gestation day 5 or 12, and the oviducts and
uterine contents were examined. The animals sacrificed at gestation day 5 showed statistically
significant decreases in the number of embryos at all doses and an increase in blastocysts without
decidua (delay of implantation) at 2 and 2.5 mg Hg/kg. The animals sacrificed at gestation day 12
showed a statistically significant decrease in the number of implants, number of living fetuses and
3-55
-------
average fetal weight at 2 and 2.5 mg Hg/kg. Maternal toxicity was not well-described, but 7 females at
2.5 mg Hg/kg and 2 females at 2 mg Hg/kg died. The study did not determine whether the failure to
implant was due to fetal toxicity or maternal uterine dysfunction.
A study by Bernard et al. (1992) was performed to assess whether prenatal and early postnatal
exposure to inorganic mercuric salts can produce nephrotoxic effects. They found that s.c. injection of
dams with just 1 mg/kg-day during pregnancy caused renal effects in the offspring. Of note is that these
effects did not appear to be significantly different in the group of dams dosed throughout the gestation
period compared to dams dosed only during the last 8 days of gestation.
3.2.3.9 Reproductive
A single case study was located concerning reproductive toxicity in humans exposed to inorganic
mercury; however, it is not clear whether the effects were compound-related. No information was
identified regarding the reproductive toxicity of inorganic mercury following inhalation exposure.
Table 3-48
Reproductive Toxicity of Inorganic Mercury in Humans: Case Study
Species/
No. per Sex
Human/ 1 F
Exposure
Duration
Once
(tablet)
Dose
(mg/kg-day)
30
(HgCl2)
Effects/Limitations/BML
Spontaneous abortion 13 days after ingestion of mercuric
chloride
Limitation: Case study; abortion may have been unrelated
to mercury exposure
BML not reported
Reference
Afonso and
deAlvarez 1960
In animals orally exposed to inorganic mercury compounds, changes in the estrous cycle and
ovulation and/or increased resorptions were reported (Pritchard et al. 1982b).
In male mice administered a single i.p. dose of 0.74 mg Hg/kg as mercuric chloride, fertility
decreased between days 28 and 49 post-treatment with no obvious histological effects noted in the sperm
(Lee and Dixon 1975). The period of decreased fertility indicated that spermatogonia and premeiotic
spermatocytes were affected. The effects were less severe than those noted after treatment with a similar
dose of methylmercury. A single i.p. dose of 1.5 mg Hg/kg as mercuric chloride administered 1-5 days
prior to mating in female mice resulted in a significant decrease in the total number of implants, number
of living embryos and a significant increase in the percentage of dead implants (Suter 1975). These
effects suggest that mercury may be a weak inducer of dominant lethal mutations. In female golden
hamsters administered 6.4 or 12.8 mg Hg/kg subcutaneously, there was no observed increase in
chromosomal aberrations in metaphase II oocytes (Watanabe et al. 1982). At the first estrous cycle post
treatment, there was a significant increase in the number of degenerated oocytes in animals at the high-
dose group. At the second estrous cycle both treatment groups had increased numbers of degenerated
oocytes, suggesting an effect of mercuric chloride on ovulation.
3-56
-------
Table 3-49
Reproductive Toxicity of Inorganic Mercury in Animals; Oral Exposure
Species/
Strain/
No. per Sex
per Group
Rat/strain
NS/NSno. ofF
Exposure
Duration
NS "before
mating and
during
gestation"
(gavage)
Dose
(mg/kg-day)
9, 12, 18
(HgCl2)
Effects/Limitations/BML
Irregular estrous cycles and high preimplantation loss
(LOAEL = 12); decreased ovulation (LOAEL = 18)
Limitations: Limited details; reported as an abstract; no
statistical analysis reported
BML not reported
Reference
Pritchardetal.
1982b
3.2.3.10 Genotoxicity
Two occupational studies (Anwar and Gabal 1991; Popescu et al. 1979) reported on workers
inhaling inorganic mercury; the data were inconclusive regarding the clastogenic activity of inorganic
mercury. Workers involved in the manufacture of mercury fulminate (HgfOCN]^ had a significant
increase in the incidence of chromosomal aberrations and micronuclei in peripheral lymphocytes when
compared to unexposed controls (Anwar and Gabal 1991). There was no correlation between urinary
mercury levels or duration of exposure to the increased frequency of effects; the study authors concluded
that mercury may not have been the clastogen in the manufacturing process. In a study by Popescu et al.
(1979), 18 workers exposed to a mixture of mercuric chloride, methylmercuric chloride and
ethylmercuric chloride had significant increases in the frequency of acentric fragments (chromosome
breaks). The findings, however, are suspect because the control group was not matched for sex, smoking
habits or sample size.
Table 3-50
Genotoxicity of Inorganic Mercury in Humans
Species/
No. per Sex
Human/29 M
exposed/ 29 M
control
Human/1 8 M
exposed/
10 control
Exposure
Duration
20.8yr(avg)
(occup)
10.5 yr
(occup)
Dose
(mg/m3)
NS
Hg(OCN)2
0.15-0.44
(HgCl2)
Effects/Limitations/BML
Increased incidence of chromosomal aberrations (p<0.001)
and micronuclei (p<0.01) in lymphocytes of workers exposed
to mercury fulminate compared with age-matched controls; no
correlation between frequency of chromosome and exposure
duration or urinary mercury level.
BML: 123.2 ug/L in urine (avg)
Increased frequency of chromosomal breaks.
Limitations: Workers also exposed to methylmercuric
chloride and ethylmercuric chloride, and one worker had
history of benzene poisoning; control group was not matched
for sex, smoking habits, or sample size.
BML: =890 ug/L in urine (avg)
Reference
Anwar and Gabal
1991
Popescu etal. 1979
3-57
-------
Exposure to inorganic mercury may produce an increase in chromosomal aberrations in mice
following oral and inhalation exposure (Ghosh et al. 1991; Selypes et al. 1984). Mercuric chloride
administered to mice by gavage induced a dose-related increase in chromosome aberrations and aberrant
cells in the bone marrow (Ghosh et al. 1991); however, mice given i.p. doses of mercuric chloride have
shown no increase in chromosomal aberrations in bone marrow cells (Poma et al. 1981) and no increase
in aneuploidy in spermatogonia (Jagiello and Lin 1973). Similarly, an increased incidence of
chromosomal aberrations (primarily deletion and numeric aberrations) was observed in livers of fetal
mice exposed to mercury in utero as the result of maternal inhalation of aerosols of mercuric chloride
(Selypes et al. 1984). Female golden hamsters injected s.c. with mercuric chloride were observed to have
increased incidence of chromosome aberrations in bone marrow cells but not in metaphase II oocytes
(Watanabe et al. 1982). Mercuric chloride concentrations in the ovaries were low but had an inhibiting
effect on ovulation. Verschaeve et al. (1984) reported that in vitro exposure of human lymphocytes and
muntjac fibroblasts to mercuric chloride resulted in segregation abnormalities; namely, c-mitotic figures.
The effects of mercuric chloride on genetic material has been suggested to be due to the ability of
mercury to inhibit of the formation of the mitotic spindle, which can result in c-mitotic figures. Mercuric
chloride has also been shown to inhibit nucleolus organizing activity in human lymphocytes (Vershaeve
etal. 1983).
Positive dominant lethal results have been obtained in studies in which rats were administered
mercuric chloride orally (Zasukhina et al. 1983). Suter (1975) observed a small, but significant increase
in the number of non-viable implants when female mice were administered mercuric chloride by
intraperitoneal injection; this effect was not observed when males were treated. It was not clear whether
the increase in non-viable implants was due to maternal toxicity or to a true dominant lethal effect of the
treatment.
Sex-linked recessive lethal mutations were not observed as a consequence of exposure of male
Drosophila melanogaster by either feeding or injection (NTP 1993).
As summarized in NTP (1993) and U.S. EPA (1985), mercuric chloride has produced some
positive results for clastogenicity in a variety of in vitro and in vivo genotoxicity assays, but mixed
results regarding its mutagenic activity have been reported. Mercuric chloride was negative in gene
mutation tests with Salmonella typhimurium strains TA1535, TA1537, TA98 and TA102 with or without
hepatic microsomal preparations (S9) (Arlauskas et al. 1985; Marzin and Phi 1985; Wong 1988).
Mercuric chloride has shown evidence of DNA damage in the Bacillus subtilis rec assay (Kanematsu et
al. 1980) but did not induce lytic phage in a lysogenic E. coll strain (Rossman et al. 1984).
DNA damage (single strand breaks) has also been observed in assays using rat and mouse
embryo fibroblasts (Zasukhina et al. 1983) and Chinese hamster ovary (CHO) cells and human KB cells
(Cantoni and Costa 1983; Cantoni et al. 1982, 1984a,b; Christie et al. 1984, 1986; Robison et al. 1982,
1984; Williams et al. 1987). Mercuric chloride also produced chromosome aberrations and sister
chromatid exchange (SCE) in CHO cells (Howard et al. 1991) and SCE in human leucocytes (Morimoto
et al. 1982). Negative results for chromosomal aberrations were reported for FM3A cells (from a mouse
mammary carcinoma) (Umeda and Nishimura 1979) and for two human diploid lines, WI38 and MRC5
(Paton and Allison 1972). Negative results for SCE were reported for don cells (Ohno et al. 1982) and
for P388D, mouse cells and CHO cells (Anderson 1983). Evidence of gene mutations (considered
weakly positive) was observed in L5178Y mouse lymphoma cells in the presence of microsomal
preparations (Oberly et al. 1982).
NTP (1993) reached the following conclusions from their In vitro testing of mercuric chloride:
not mutagenic for Salmonella typhimurium in preincubation protocols with and without rat and hamster
3-58
-------
liver preparations; positive for L5178Y cells without addition of hepatic preparations; negative for SCE
in CHO cells without addition of S9 but weakly positive when rat S9 was added; positive for
chromosomal aberrations in CHO cells in the absence but not the presence of liver preparations; it was
not clear what role was played by cytotoxicity in the generation of these chromosomal aberrations.
Table 3-51
Genotoxicity of Inorganic Mercury in Animals
Species/
Strain/
No. per Sex
per Group
Mice/CFLP
NS/No. F NS
Exposure
Duration
4d
4hr/d
Gd 9-12
Dose
(mg/m3)
0.17, 1.6
(HgCl2)
Effects/Limitations/BML
Increased incidence of chromosomal aberrations in fetal
hepatocytes
Limitations: The number of mothers corresponding to the 10
fetuses examined was not reported; no statistical analysis
BML not reported
Reference
Selypesetal. 1984
3.3 Methylmercury
Organic mercury compounds have been used as fungicides and as pharmaceutical agents
(diuretics). Organic mercurials including Metaphen, Merthiolate and Mercurochrome still find use as
topical antiseptics. Phenylmercury salts are used in pharmaceutical, ophthalmic and cosmetic
preparations to control growth of microbial organisms (Joklik et al. 1984). Other organic mercury
compounds include methylmercuric chloride (MMC), methylmercuric hydroxide (MMH) and
phenylmercuric acetate (PMA). Nearly all of the available toxicity studies for organic mercury
compounds, however, are for methylmercury. Unless otherwise noted, all studies summarized in tables
in this section are for methylmercury. All oral doses were converted to mg Hg/kg-day, and all inhalation
doses were converted mg Hg/m3 using the method shown in Appendix A.
3.3.1 Critical Noncancer Data
This section provides descriptions of studies considered by U.S. EPA in evaluation of systemic
health endpoints, largely neurotoxicity in exposed adults and in children exposed in utero. Chapter 6
describes the derivation of an RfD for methylmercury based on developmental neurologic abnormalities
in human infants. For completeness some of these studies are also presented in subsequent sections in
tabular form.
3.3.1.1 Human Data
Several studies of methylmercury poisonings in humans have been reported (see discussion on
Neurologic Effects). CNS effects were observed in several studies summarized by Clarkson et al. (1976),
Nordberg and Strangert (1976), and WHO (1976). CNS effects including ataxia and paresthesia have
been observed in subjects with blood mercury concentrations as low as 200 (ig/L.
3-59
-------
The original epidemiologic report of methylmercury poisoning involved 628 human cases that
occurred in Minamata, Japan, between 1953 and 1960. The overall prevalence rate for the Minamata
region for neurologic and mental disorders was 59%. Among this group 78 deaths occurred, and hair
concentrations of mercury ranged from 50-700 ppm. These hair mercury concentrations were
determined through the use of less precise analytic methods than were available for later studies. The
most common clinical signs observed in adults were paresthesia, ataxia, sensory disturbances, tremors,
impairment of hearing and difficulty in walking. Examination of the brains of severely affected patients
that died revealed marked atrophy of the brain (55% normal volume and weight) with cystic cavities and
spongy foci. Microscopically, entire regions were devoid of neurons, granular cells in the cerebellum,
golgi cells and Purkinje cells. Extensive investigations of congenital Minamata disease were undertaken,
and 20 cases that occurred over a 4-year period were documented. In all instances the congenital cases
showed a higher incidence of symptoms than did the cases wherein exposure occurred as an adult.
Severe disturbances of nervous function were described, and the affected offspring were very late in
reaching developmental milestones. Hair concentrations of mercury in affected infants ranged from 10 to
100 ppm. Hair mercury levels for the mothers during gestation were not available (Tsubaki 1977).
In 1971, an unknown number of people in Iraq were exposed to methylmercury-treated seed
grain that was used in home-baked bread. Studies conducted on this population include that of Bakir et
al. (1973), Marsh et al. (1987) and others. Toxicity was observed in many adults and children who had
consumed this bread over a three-month period, but the population that showed greatest sensitivity were
offspring of pregnant women who ate contaminated bread during gestation. The predominant symptom
noted in adults was paresthesia, and it usually occurred after a latent period of from 16 to 38 days. In
adults symptoms were dose-dependent, and among the more severely affected individuals ataxia, blurred
vision, slurred speech and hearing difficulties were observed. Signs noted in the infants exposed during
fetal development included cerebral palsy, altered muscle tone and deep tendon reflexes, as well as
delayed developmental milestones (e.g., walking before 18 months and talking before 24 months). Some
information indicated that male offspring were more sensitive than females. The mothers experienced
paresthesia and other sensory disturbances but at higher doses than those associated with their children
exposed in utero. Unique analytic features of mercury (analysis of segments of hair correlated to specific
time periods in the past) permitted approximation of maternal blood levels that fetuses were exposed to
in utero. The data collected by Marsh et al. (1987) summarized clinical neurologic signs of 81 mother
and child pairs. From x-ray fluorescent spectrometric analysis of selected regions of maternal scalp hair,
concentrations ranging from 1 to 674 ppm were determined, then correlated with clinical signs observed
in the affected members of the 81 mother-child pairs. Among the exposed population there were affected
and unaffected individuals throughout the dose-exposure range. (See also Bakir et al. (1973) in sections
on Death and Neurological Effects).
McKeown-Eyssen et al. (1983) provided a report of neurologic abnormalities in four
communities of Cree Indians in northern Quebec. A group of 241 children between 12 and 30 months of
age was evaluated for clinical signs consistent with methylmercury exposure. A pediatric neurologist
evaluated the children for the following: height, weight, head circumference, dysmorphic and congenital
features and the presence of acquired disease. In addition to the DDST the following assessments were
done: special senses, cranial nerve function, sensory function, muscle tone, stretch reflexes, co-
ordination, persistence of Babinski response, and a summary of signs for absence or presence of
neurologic abnormality. An attempt was made to account for possible confounding factors; the
interviewers determined alcohol and tobacco consumption patterns among the mothers of affected
children. Age of the mothers and multiparity was also taken into account in analysis of the data.
Maternal hair mercury was used as the exposure measure. The average maternal hair mercury was the
same for boys and girls (6 ppm); only 6% of the population had exposure above 20 ppm. The prevalence
of multiple abnormal neurologic findings was about 4% for children of both sexes. The most frequently
3-60
-------
observed abnormality was delayed deep tendon reflexes. This was seen in 11.4% of the boys and 12.2%
of the girls. Abnormality of muscle tone or reflexes showed a significant positive association with
maternal mercury exposure for boys, but not for girls. A consistent dose-response relationship for this
effect was not observed; however, the greatest prevalence of the effect in boys occurred for those with
mothers in the highest exposure group (13.0-23.9 ppm mercury in hair). No other measure of abnormal
or decreased neurologic function or development showed a significant positive association with maternal
hair mercury. The prevalence of abnormality of muscle tone or reflexes was found to increase 7 times
with each increase of 10 ppm of the prenatal exposure index. There was possible influence of alcohol
consumption and smoking among mothers on the effects observed in their children.
Studies performed in New Zealand investigated the development of children who had prenatal
exposure to methylmercury (Kjellstrom et al. 1986a, 1989). A group of 11,000 mothers who regularly
ate fish was initially screened by survey; of these, about 1000 had consumed 3 fish meals per week
during pregnancy. Working from these two large groups, 31 matched pairs were established. A
reference child matched for mother's ethnic group, age, child's birthplace and birth date was located for
each child in the high fish consumption group. Mercury exposure during gestation was determined from
maternal hair analysis. The average hair concentration for high exposure mothers was 8.8 ppm and for
the reference group was 1.9 ppm. At 4 years of age, the children were tested using the Denver
Developmental Screen Test (DDST). This is a standardized test of a child's mental development and can
be administered in the child's home. It consists of four major function sectors: gross motor, fine motor,
language and personal-social. A developmental delay in an individual item is scored when the child has
failed in his/her response and at least 90% of children can pass this item at a younger age. The whole test
is scored as abnormal, questionable, or normal. Standardized vision tests and sensory tests were also
performed to measure development of these components of the nervous system. The prevalence for
developmental delay in children was 52% for progeny of high mercury mothers and 17% for progeny of
mothers of the reference group. The hair mercury concentrations of the mothers in this study were lower
than those associated with CNS effects in children exposed in Japan and Iraq. The results of the DDST
included 2 abnormal scores and 14 questionable scores in the high mercury-exposed group and 1
abnormal and 4 questionable scores in the control group. The results remained statistically significant
after the 8 pairs where ethnic group matching was not successful and twins were excluded. Analysis of
the DDST results by sector showed that developmental delays were most commonly noted in the fine
motor and language sectors, but the differences between the experimental and control groups were not
significant. The differences noted in performance of the DDST between high mercury-exposed and
referent children could be due to confounding variables. Infants of the mercury-exposed group more
frequently had low birth weights and were more likely to be born prematurely.
A second stage follow-up of the original Kjellstrom study was carried out when the children
were 6 years old (Kjellstrom et al. 1986b). In this later study the high exposure children were compared
with 3 control groups with lower prenatal mercury exposure. During pregnancy, mothers in two of these
control groups had high fish consumption and average hair mercury concentrations of 3-6 ppm and
0-3ppm, respectively. The high exposure group was matched with controls for maternal ethnic group,
age, smoking habits, residence, and sex of the child. For the second study, 61 of 74 high exposure
children were available for study. Each child was tested with an array of scholastic, psychological and
behavioral tests which included Test of Language Development (TOLD), the Wechsler Intelligence Scale
for Children and McCarthy Scale of Children's Abilities. The results of the tests were compared between
groups. Confounding was controlled for by a modelling procedure using linear multiple regression
analysis. A principal finding was that normal results of the psychological test variables were influenced
by ethnic background and social class. High prenatal methylmercury exposure decreased performance in
the tests, but it contributed only a small part of the variation in test results. It was found that an average
hair mercury level of 13-15 ppm during pregnancy was consistently associated with decreased test
3-61
-------
performance. Size of the experimental groups limited the power of the study to determine if lower
exposure levels might have had a significant effect on test results. The studies are limited for assessing
methylmercury toxicity because the intelligence tests used may not be the most appropriate for defining
the effects of methylmercury. Also, greater significance was seen in differences of cultural origins of the
children than the differences in maternal hair methylmercury concentrations.
A prospective study (Marsh et al. 1995) of fetal exposure to methylmercury through maternal
consumption offish was conducted in the Peruvian fishing village of Mancora between 1981-1984. An
account of the study was written in 1985 and subsequently published in 1995. Hair samples were
obtained from 369 pregnant women, and neurologic examinations were performed on 194 of the children
from these pregnancies. Of this cohort there were 131 mother-infant pairs with complete clinical data
and hair samples that provided quantitative measures of methylmercury exposure during pregnancy. The
peak hair mercury levels ranged from 1.2 to 30 ppm with a mean of 8.3 ppm. Testing of hair segments
and statistical analyses confirmed that the women had reached a relatively steady state in terms of
methylmercury. No significant correlation was shown between increasing maternal hair methylmercury
and effects on the developmental determinations assessed in this study. Measures included the
following: perinatal factors (labor or delivery difficulty, abnormal respiration or color), maternal
paresthesia, speech retardation, muscle tone evaluation, determination of primitive and tendon reflexes,
ataxia and mental and motor retardation.
There has been an increasing concern about the likelihood of exposure of people residing in the
Amazonian watershed to methylmercury in fish and other sources. The Amazon River valley is the site
of many small gold mining operations which use metallic mercury in the extraction process; it is
estimated that 55-60% is released to the atmosphere and 40-45% enters the aquatic environment (Pfeiffer
et al. 1989; Malm et al 1990). Lebel et al (1996) have published a study of 29 adult residents of two
villages located on the Tapajos River, a tributary of the Amazon located about 200 Km from the mining
sites. There were 14 women and 15 men aged 35 and younger who were randomly chosen from a
previous survey. Total hair mercury ranged from 5.6 to 38.4 ppm; methylmercury constituted between
72.2% and 93.3% of the sampled mercury. A quantitative behavioral neurophysiologic battery was
modified for administration to persons with minimal formal education in an area without electricity. For
women only there was a decrease with increasing hair mercury in manual dexterity as measured in the
Santa Ana test, Helsinki version. For both men and women there was a statistically significant decrease
with increasing mercury in color discrimination capacity (as measured by the Lanthony D-15 desaturated
panel). Near visual contrast sensitivity profiles (measured with the Vistech 6000) and peripheral visual
field profile (Goldman Perimetry with Targets I and IV) were both reduced in the individuals with the
highest hair mercuries. The authors noted that constriction of the visual field has been observed in other
instances of mercury intoxication and that changes in contrast sensitivity has been noted in non-human
primates exposed to methylmercury (Rice and Gilbert 1982; Rice and Gilbert 1990).
Lonky et al. (1996) have reported on studies of behavioral effects in newborns as a consequence
of maternal consumption offish from Lake Ontario in the U.S.. Fish from the U.S. Great Lakes have
been shown to be contaminated with a number of environmental pollutants including PCBs and
methylmercury. A total of 559 children were tested. Exposure was measured as fish consumption which
was determined by interviews taking place during pregnancy. Information was collected on fish species,
number of meals, serving size, and method of preparation. Exposure was scaled in PCB equivalent
weights. Women were assigned to the high dose group if they reported eating at least 40 PCB-equivalent
pounds offish in their lifetime (n=152); the low fish consumption group numbered 243, and there were
164 no-fish-consumption controls. Infants born to the high fish consumption group scored significantly
more poorly on the Reflex, Autonomic and Habituation clusters of the Neonatal Behavioral Assessment
Scale. The authors do not attribute these developmental effects to exposure to any one compound;
3-62
-------
analyses of cord blood and maternal hair are being done for PCBs, DDE, hexachlorobenzene, lead and
mercury.
In 1981 a group of researchers in conjunction with the Seychelles Island government initiated a
large study on developmental effects of low level methylmercury exposure from consumption of marine
fish. The Seychelles is an island country in the Indian Ocean near the coast of Africa where fish are
consumed by the population on a daily basis. A pilot study (essentially a cross-sectional study) was
initiated which focused on all children born between February 1989 and February 1990. A total of 804
mother-infant pairs were enrolled, which was 48% of those eligible. The main study was designed to be
prospective and involved 779 mother-child pairs. In both studies maternal hair samples and umbilical
cord blood were measured for total mercury content using atomic absorption spectroscopy. Children
were enrolled in the pilot study during the years 1987-1988; the main study enrolled children from 1989
to 1990. Both the pilot and main studies involved about 50% of all children born during the year, 804
and 779 children respectively. (Shamlaye et al. 1995)
The authors have noted that the pilot study was not as well-controlled as the main or longitudinal
study: there were fewer covariates, medical records were not reviewed as carefully, there was less
information on socio-economic status. Subsets of enrolled children were tested, and the main purpose of
testing was to pilot the test batteries. Multiple hair samples collected for the pilot study during gestation
had methylmercury ranging from 0.6 to 36.4 ppm with a median of 6.6 ppm (Myers et al. 1995b). The
endpoints evaluated during the pilot study included a general neurologic evaluation and the DDST-R in
addition to examinations of physical development. The age of each child was known; children were
tested once between the ages of 5 and 109 weeks of age. In addition to DDST-R testing, a medical
history was obtained, and general and neurologic examinations were also performed. Statistical analysis
included the following covariates: gender, birth weight, one- and five-minute Apgar score, age at testing,
and medical problems. Covariates for the mothers were age, tobacco and alcohol consumption during
pregnancy and medical problems. An association between in utero mercury exposure was found for
DDST-R abnormal plus questionable scores combined. (Myers et al 1995b)
A subset of the pilot cohort was evaluated at 66 months (Myers et al. 1995a). This group of 217
children was administered the McCarthy Scales of Children's Abilities, the Preschool Language Scale,
and the Letter-Word Recognition and Applied Problems subtests of the Woodcock-Johnson Tests of
Achievement that were appropriate to the children's age. The median maternal hair mercury for this
group was 7.11 ppm. Mercury exposure (measured as maternal hair mercury) was negatively associated
with four endpoints: The McCarthy General Cognitive Index and Perceptual Performance subscale; and
the Preschool Language Scale Total Language and Auditory Comprehension subscale. When statistically
determined outliers and points considered to be influential were removed from the analyses, statistical
significance of the association remained only for auditory comprehension.
The prospective or main study, involved evaluation of children at 6.5, 19, 26 and 66 months of
age. Age-appropriate tests administered included the following: Infantest (or Pagan's test of visual
recognition memory), Bayley Scales of Infant Development (BSID), McCarthy Scales of Children's
Abilities, the Preschool Language Scale and the DDST (6.5 months only). Maternal intelligence and
home environment were also assessed (Marsh et al. 1995a). In the group (n = 740) evaluated at 6.5
months, median maternal hair mercury was 5.9 ppm with a range of 0.5 ppm to 26.7 ppm. No association
with maternal hair mercury was found for any of six endpoints in six children tested at six months.
(Myers etal. 1995c).
Evaluations at 19 and 29 months were done on groups of 738 and 736 individuals, respectively
(Davidson et al. 1995). Median maternal hair mercury was 5.9 ppm and the range was 0.5 to 26.7 ppm.
3-63
-------
Children were evaluated with the BSID at 19 months of age. At 29 months, children were administered
the BSID as well as the Bayley Infant Behavior Record. Mean BSID mental indices were comparable to
scores for children in the U.S. at both 19 and 29 months. The BSID Psychomotor Scale was reflective of
the rapid development of motor skills by children reared in African cultures. No effects of mercury
exposure were seen on outcome of five test endpoints at 19 months. At 29 months there was an
association between mercury exposure and decreased activity level in male children only. The authors
point out that the activity level observed during the testing session may not reflect the child's activity
level in other settings. There was no association with maternal hair mercury in 15 other endpoints.
Delayed onset of walking and talking were among the measures of toxicity used in the evaluation
of Iraqi children in utero (Marsh et al. 1987.) When the children in the Seychelles study reached the age
of 19 months, caregivers were queried as to the time at which they walked and talked (n=760 and 680,
respectively). Onset of walking was defined as walking without support and age at talking as the age at
which the child said words other than "mama" or "dada." The mean age for walking was 10.7 months for
girls and 10.6 for boys; for talking it was 10.5 for girls and 11.0 for boys. For female children, the mean
age for walking was similar across all mercury exposures. In boys, the mean age for walking increased
between 0.3 and 0.6 months from the lowest to highest mercury exposure groups. Statistical analyses
adjusted covariates and outlying points data. With these adjustments there was no statistically significant
association between maternal hair mercury and age at which the child walked or talked.
The overall conclusion of the studies published to date is that it is yet unclear whether an
association exists between low level mercury exposure and neurologic deficits in children. The study
does show a close correlation between maternal hair mercury and neonatal levels of mercury in brain
tissue (Cernichiari et al. 1995). The authors cautioned in several papers that subtle neurologic and
neurobehavioral effects are more likely to be detected in older rather than younger children. The overall
conclusion of the authors is that their results require careful interpretation, and that an association
between relatively low level mercury exposure in utero and neurologic deficits has not been conclusively
demonstrated.
A large study was initiated in the Faroe Islands in 1986 on neurologic developmental effects of
methylmercury and PCB exposure in utero (Grandjean et al. 1997). The population of the Faroes is
relatively homogeneous. During pregnancy consumption of alcoholic beverages is uncommon. As with
other fishing communities, seafood is large part of the Faroese diet. Increased mercury exposure,
however, is largely attributed to the eating of pilot whale, which is traditionally hunted and shared among
the population (Grandjean et al. 1992a). Subjects were a group of 917 (of an initial cohort of 1022)
children born between 1986 and 1987 and evaluated at about 7 years of age. Mercury was measured in
maternal hair and cord blood, and a subset of cords was evaluated for PCBs. Of the initial cohort the
median blood mercury was 24.2 (ig/L with 25% of the samples above 40 (ig/L. The median maternal hair
mercury concentration was 4.5 ppm, and 13% were greater than 10 ppm (Grandjean et al. 1992a). Cord
blood mercury was found to be most closely associated with maternal hair mercury; the association with
hair measurements in the children at 12 months and 7 years was not as strong (Grandjean et al. 1997).
At seven years children received a physical examination including a functional neurological
examination which emphasized motor co-ordination and perceptual-motor performance. Visual acuity
was measured using Snellen's board; contrast sensitivity was assessed using the Functional Acuity
Contrast Test. Standard hearing tests were done. Neurophysiological tests included the following:
pattern-reversal visual evoked potentials; brainstem auditory evoked potentials; postural sway under four
conditions; and coefficient of variation for R-R intervals on electrocardiogram as a measure of autonomic
nervous system function. Neuropsychological tests included these: motor tests~the Neurobehavioral
3-64
-------
Evaluation System (NES) finger tapping test and the NES Hand-Eye Coordination test; tactile processing
~ Tactual Performance Test; vigilance/attention~NES Continuous Performance Test; attention and
tracking ~ Wechsler Intelligence Scale for Children-revised (WISC-r) digit span forward; reasoning and
cognitive flexibility~WISC-r similarities; visuospatial~WISC-r Block Designs and Bender Visual Motor
Gestalt Test; language—Boston Naming Test; memory—California Verbal Learning Test; and mood--
Nonverbal Analogue Profile of Mood States.
Three neurological tests were found to be difficult for 7-year-old children; fewer than 60% of the
children performed optimally. On one of these tests, finger opposition, the group of 465 with optimal
performance had a geometric mean cord blood mercury of 21.8 (ig/L of 21.8 by contrast to 23.9 (ig/L for
the 425 children with questionable or deficient performance; this was a statistically significant
difference. The neurophysiological tests showed no indication of mercury-associated dysfunction.
Significant negative associations were seen on several neuropsychological tests. Even with inclusion of
covariates with uncertain influence on these tests results, multiple regression analysis indicated that 9/20
measures showed mercury related decrements (p< 0.05, one tailed). Application of a Peters-Belson
adjustment resulted in significant mercury associations for 11/20 measures.
Pilot whale fat (blubber) is consumed in the Faroe Islands, and this could result in increased
exposure to PCBs. PCB concentrations in Faroese breast milk has been shown to be higher than in other
Scandinavian countries (Grandjean et al. 1995b). PCB determinations were done on a total of 436 cord
bloods, and PCB exposure was included as a covariate in the regression analyses. This had an effect only
on the regression for the Boston Naming Test. The authors concluded that in utero exposure to
methylmercury affects several domains of cerebral function. After exclusion of children with maternal
hair mercury concentrations above 10 ppm, the association between mercury exposure and
neuropsychological dysfunction remained unchanged. The authors, therefore, concluded that adverse
effects are observed at exposures below 10 ppm maternal hair (Grandjean et al. 1997).
3.3.1.2 Animal Data
Rice (1989b) dosed five cynomolgus monkeys (Macaca fasclculans) from birth to 7 years of age
with 0.05 mg Hg/kg-day as methylmercuric chloride and performed clinical and neurologic examinations
during the dosing period and for an additional 6 years. As a sensitive indicator of the latent effects of
methylmercury, neurologic examinations performed at the end of the observation period revealed
insensitivity to touch and loss of tactile response. In the later stages of the observation period monkeys
dosed with methylmercury were clumsier and slower to react when placed in the exercise cage than were
unexposed monkeys.
Gunderson et al. (1986) administered daily doses of 0.04-0.06 mg Hg/kg as methylmercuric
hydroxide to 11 crab-eating macaques (Macaca fasclculans) throughout pregnancy, resulting in maternal
blood levels of 1080-1330 //g/L in mothers and 1410-1840 //g/L in the offspring. When tested 35 days
after birth, the infants exhibited visual recognition deficits.
Groups of 7 or 8 female crab-eating macaques (Macaca fasclculans) were dosed with 0.05 or
0.09 mg/kg-day of methylmercury through 4 menstrual cycles (Burbacher et al. 1984). They were mated
with untreated males, and clinical observations were made for an additional 4 months. Two of 7 high-
dose females aborted, and 3 did not conceive during the 4-month mating period. The other two females
delivered live infants. Two of 7 females exposed to 0.05 mg/kg-day aborted; the remaining 5 females
delivered live infants. All control females conceived, and 6 delivered live infants. These reproductive
results approached but did not reach statistical significance. Reproductive failure within dose groups
could be predicted by blood mercury levels. The dams did not show clinical signs of methylmercury
3-65
-------
poisoning during the breeding period or gestation, but when females were dosed with 0.09 mg/kg-day for
a year, 4 of 7 did show adverse neurologic signs.
Bornhausen et al. (1980) has reported a decrease in operant behavior performance in 4-month-old
rats whose dams had received methylmercuric chloride on gestation days 6-9. A statistically significant
effect was seen in offspring whose dams had received 0.01 and 0.05 mg/kg five times during gestation.
The authors postulated that more severe effects of in utero exposure would be seen in humans since the
biological half-time of mercury in the brain of humans is 5 times longer than the rat. In addition, much
longer in utero exposure to mercury would occur in humans since gestation is much longer.
Groups of Wistar rats (50/sex/group) were administered daily doses of 0.002, 0.01, 0.05, and
0.25 mg Hg/kg-day as methylmercuric chloride for 26 months (Munro et al. 1980). Female rats that
received 0.25 mg/kg-day had reduced body weight gains and showed only minimal clinical signs of
neurotoxicity; however, male rats that received this dose did show overt clinical signs of neurotoxicity,
had decreased hemoglobin and hematocrit values, had reduced weight gains and showed increased
mortality. Histopathologic examination of rats of both sexes receiving 0.25 mg/kg-day revealed
demyelination of dorsal nerve roots and peripheral nerves. Males showed severe kidney damage, and
females had minimal renal damage. This study showed aNOAEL of 0.05 mg/kg-day and a LOAEL of
0.25 mg/kg-day.
A 2-year feeding study of methylmercuric chloride was conducted in B6C3F1 mice
(60 mice/sex/group) at doses of 0, 0.4, 2, and 10 ppm (0, 0.03, 0.15, and 0.73 mg Hg/kg-day in males; 0,
0.02, 0.11, and 0.6 mg Hg/kg-day in females) to determine chronic toxicity and possible carcinogenic
effects (Mitsumori et al. 1990). Mice were examined clinically during the study, and neurotoxic signs
characterized by posterior paralysis were observed in 33 males after 59 weeks and 3 females after 80
weeks in the 0.6-mg Hg/kg-day group. A marked increase in mortality and a significant decrease in body
weight gain were also observed in the high-dose males, beginning at 60 weeks. Post-mortem
examination revealed toxic encephalopathy consisting of neuronal necrosis of the brain and toxic
peripheral sensory neuropathy in both sexes of the high-dose group. An increased incidence of chronic
nephropathy was observed in the 0.11- and 0.6-mg Hg/kg-day males. These results indicated that
B6C3F1 mice are more sensitive to the neurotoxic effects of methylmercury than ICR mice.
Ultrastructural renal changes were also observed in rhesus monkeys treated with
0.08-0.12 mg/kg of methylmercury although clinical changes were not observed (Chen et al. 1983).
3.3.2 Cancer Data
3.3.2.1 Human Data
Three studies were identified that examined the relationship between methylmercury exposure
and cancer. No persuasive evidence of increased carcinogenicity attributable to methylmercury exposure
was observed in any of the studies. Interpretation of these studies, however, was limited by poor study
design and incomplete descriptions of methodology and/or results.
Tamashiro et al. (1984) evaluated the causes of death in 334 subjects from the Kumamoto
Prefecture who had been diagnosed with Minamata disease and died between 1970 and 1980. Minamata
disease was used as a surrogate for methylmercury exposure. The cases were fishermen and their
families who had been diagnosed with methylmercury poisoning (Minamata disease); thus, Minamata
disease was used as a surrogate for methylmercury exposure. The controls were selected from all deaths
that had occurred in the same city or town as had the cases and were matched on the basis of sex, age at
3-66
-------
death (within 3 years) and year of death; two controls were matched to each case. Malignant neoplasms
were designated as the underlying cause of death in 14.7% (49/334) of the cases and 20.1% (134/668) of
the controls. For 47 cases in which Minamata disease was listed as the underlying cause of death, the
investigators reanalyzed the mortality data and selected one of the secondary causes to be the underlying
cause of death in order to allow examination of the cases and controls under similar conditions and
parameters. The three cases for which Minamata disease was listed as the only cause of death were
excluded from further analysis. Using the Mantel-Haenzel method to estimate odds ratios, no significant
differences were observed between the cases and controls with respect to the proportion of deaths due to
malignant neoplasms among males, females, or both sexes combined. The estimated odds ratios and 95%
confidence intervals were 0.84 (0.49-1.43), 0.58 (0.28-1.21), and 0.75 (0.50-1.11) for males, females,
and both sexes combined. Similarly, no increases were observed among the cases relative to the controls
when malignant neoplasms were identified as a secondary cause of death or were listed on death
certificates as one of multiple causes of death. These data suggest that cancer incidence is not increased
in persons with overt signs of methylmercury poisoning when compared to persons for whom no
diagnosis of methylmercury poisoning had been made. Interpretation is limited, however, by potential
bias in designating the cause of death among patients with known Minamata disease and by the
uncertainty regarding the extent of methylmercury exposure and undiagnosed Minamata disease among
the controls.
In a subsequent study, Tamashiro et al. (1986) compared the mortality patterns (between 1970
and 1981) among residents of Fukuro and Tsukinoura districts (inhabited mainly by fishermen and their
families) in the Kumamoto Prefecture with age-matched residents of Minamata city (also in the
Kumamoto Prefecture) who died between 1972 and 1978. In this study, high exposure to methylmercury
was inferred from residence in a district believed to have higher intake of local seafood. By contrast, in
the 1984 study described above, high methylmercury exposure was inferred from a diagnosis of
Minamata disease. A total of 416 deaths were recorded in the Fukuro and Tsukinoura districts in
1970-1981, and 2,325 deaths were recorded in Minamata City in 1972-1978. No statistically significant
increase in the overall cancer mortality rate was observed; however, an increase in the mortality rate due
to liver cancer was observed (SMR, 207.3; 95% C.I. 116.0-341.9). Analysis of mortality by sex showed
a statistically significant increase in the rate of liver cancer only among males (SMR, 250.5; 95% C.I.
133.4-428.4). Males also had statistically significantly higher mortality due to chronic liver disease and
cirrhosis. The authors note that these results should be interpreted with caution because the population
of Fukuro and Tsukinoura districts had higher alcohol consumption and a higher prevalence of hepatitis
B (a predisposing factor for hepatocellular cancer). Interpretation of these results is also limited by an
incomplete description of the methodology used to calculate the SMRs; it is unclear whether the study
authors used appropriate methods to compare mortality data collected over disparate time frames (i.e. 12
years for exposed and seven years for controls).
In a study from Poland, Janicki et al. (1987) reported a statistically significant (p<0.02) increase
in the mercury content of hair in leukemia patients (0.92 ± 1.44 ppm; n=47) relative to that in healthy
unrelated patients (0.49 ± 0.41 (ig/g; n=79). Similarly, the mercury content in the hair of a subgroup of
leukemia patients (0.69 ±0.75 (ig/g; n=19) was significantly (p<0.05) greater than that in healthy
relatives who had shared the same residence for at least 3 years preceding the onset of the disease (0.43 ±
0.24 (ig/g; n=52). When patients with specific types of leukemia were compared with the healthy
unrelated subjects (0.49 ± 0.41 (ig/g; n=79), only those with acute leukemia (type not specified; 1.24 ±
1.93 (ig/g; n=23) had a significantly increased hair mercury content. No significant differences in hair
mercury content were observed in 9 patients with chronic granulocytic leukemia or 15 patients with
chronic lymphocytic leukemia when compared to the unrelated, healthy controls. This study is of limited
use for cancer risk assessment because of the following: small size of population studied; inadequate
description of the leukemia patients or healthy controls (e.g., age distribution, length of residence in the
3-67
-------
region, criteria for inclusion in the study); uncertainty regarding the source of mercury exposure (the
authors presumed that exposure was the result of use of methylmercury-containing fungicides);
uncertainty regarding the correlation between the chronology of incorporation of mercury in the hair and
onset of the disease; and the failure to address exposure to other chemicals or adjust for other leukemia
risk factors. Furthermore, the variability of hair mercury content was large, and the mean hair mercury
levels were within normal limits for all groups. One cannot rule out the likelihood that the observed
correlation of leukemia incidence with mercury in hair is due to chance alone.
The carcinogenic effects of organomercury seed dressing exposure were investigated in a series
of case-control studies for incidence of soft-tissue sarcomas (Eriksson et al. 1981; Hardell and Eriksson,
1988; Eriksson et al. 1990) or malignant lymphomas (Hardell et al. 1981). These studies were conducted
in Swedish populations exposed to phenoxyacetic acid herbicides or chlorophenols (the exposures of
primary interest in the studies), organomercury seed dressings, or other pesticides. Exposure frequencies
were derived from questionnaires and/or interviews. Control groups from the same region of the country
were matched to cases based on vital status. There were 402 total cases of soft-tissue sarcoma, and
(among persons not exposed to phenoxyacetic acid herbicides) there were 128 cases of malignant
lymphoma. In each study, the odds ratio for exposure to organomercury in seed dressings and sarcoma or
lymphoma was either less than 1.0, or the range of the 95% confidence interval for the odds ratio
included 1.0; therefore, no association was indicated for organomercury exposure and soft-tissue sarcoma
or malignant lymphoma. The conclusions from these studies are limited, however, due to the study
subjects' likely exposures to the other pesticides and chemicals.
Table 3-52
Carcinogenic Effects of Methylmercury in Humans: Epidemiological Studies
Species/
No. per Sex
Human/334
exposed (M+F),
668 control
Human/412
exposed (M+F)
Human/47 w/
leukemia (sex
not specified)
control 79
Exposure
Duration
NS
NS
NS
Dose
(mg/kg-day)
NS
NS
NS
Effects/Limitations/BML
No increase in cancer mortality among Minamata exposure
victims (i.e., with overt methylmercury poisoning).
Minamata disease was used as a surrogate for
methylmercury exposure.
Limitations: Exposure levels or number of undiagnosed
cases among controls not known.
Increased incidence of liver cancer in males living in the
vicinity of Minamata Bay.
Limitations: "Exposed" districts had higher alcohol
consumption and higher prevalence of hepatitis B.
Increased mercury in hair of leukemia patients; however,
mean hair mercury levels in the leukemia patients was
within the normal range.
Limitations: Small study population; source of
methylmercury exposure not clear; failure to address other
leukemia risk factors or exposure to other chemicals.
Reference
Tamashiro et al. 1984
Tamashiro et al. 1986
Janickietal. 1987
3-68
-------
3.3.2.2 Animal Data
The results from three dietary studies in two strains of mice indicate that methylmercury is
carcinogenic. A fourth dietary study in mice, three dietary studies in rats and a dietary study in cats
failed to show carcinogenicity of methylmercury. Interpretation of two of the positive studies was
complicated by observation of tumors only at doses that exceeded the MTD. Interpretation of four non-
positive studies was limited because of deficiencies in study design or failure to achieve an MTD.
Methylmercuric chloride was administered in the diet at levels of 0, 0.4, 2, or 10 ppm (0, 0.03,
0.14 and 0.69 mg Hg/kg-day in males and 0, 0.03, 0.13, and 0.60 mg Hg/kg-day in females) to B6C3F1
mice (60/sex/group) for 104 weeks (Mitsumori et al. 1990). In high-dose males, a marked increase in
mortality was observed after 60 weeks (data were presented graphically; statistical analyses not
performed). Survival at study termination was approximately 50%, 60%, 60%, and 20% in control, low-,
mid-, and high-dose males, respectively, and 58%, 68%, 60%, and 60% in control, low-, mid-, and high-
dose females, respectively. The cause of the high mortality was not reported. At study termination, the
mean body weight in high-dose males was approximately 67% of controls and in high-dose females was
approximately 90% of controls (data presented graphically; statistical analyses not performed). Focal
hyperplasia of the renal tubules was significantly (p<0.01) increased in high-dose males (14/60; the
incidence was 0/60 in all other groups). The incidence of renal epithelial carcinomas (classified as solid
or cystic papillary type) was significantly (p<0.01) increased in high-dose males (13/60; the incidence
was 0/60 in all other groups). The incidence of renal adenomas (classified as solid or tubular type) was
also significantly (p<0.05) increased in high-dose males; the incidence was 0/60, 0/60, 1/60, and 5/60 in
control, low-, mid-, and high-dose males, respectively, and 0/60, 0/60, 0/60, and 1/60 in control, low-,
mid-, and high-dose females, respectively. No metastases were seen in the animals. The incidences of a
variety of nonneoplastic lesions were increased in the high-dose rats including these: sensory
neuropathy, neuronal necrosis in the cerebrum, neuronal degeneration in the cerebellum, and chronic
nephropathy of the kidney. Males exhibited tubular atrophy of the testis (1/60, 5/60, 2/60, and 54/60 in
control, low-,
mid-, and high-dose, respectively) and ulceration of the glandular stomach (1/60, 1/60, 0/60, and 7/60 in
control, low-, mid-, and high-dose males, respectively). An MTD was achieved in mid-dose males and
high-dose females. High mortality in high-dose males indicated that the MTD was exceeded in this
group.
Mitsumori et al. (1981) administered 0, 15, or 30 ppm of methylmercuric chloride (99.3% pure)
in the diet (0, 1.6 and 3.1 mg Hg/kg-day) to ICR mice (60/sex/group) for 78 weeks. Interim sacrifices of
up to 6/sex/group were conducted at weeks 26 and 52. Kidneys were microscopically examined from all
animals that died or became moribund after week 53 or were killed at study termination. Lungs from
mice with renal masses and renal lymph nodes showing gross abnormalities were also examined.
Survival was decreased in a dose-related manner; at week 78 survival was 24/60, 6/60 and 0/60 in
control, low- and high-dose males, respectively, and 33/60, 18/60 and 0/60, in control, low- and high-
dose females, respectively (statistical analyses not performed). The majority of high-dose mice (51/60
males and 59/60 females) died by week 26 of the study. Examination of the kidneys of mice that died or
were sacrificed after 53 weeks showed a significant (p<0.001) increase in renal tumors in low-dose males
(13/16 versus 1/37 in controls). The incidence of renal epithelial adenocarcinomas in control and low-
dose males was 0/37 and 11/16, respectively (p<0.001). The incidence of renal epithelial adenomas in
control and low-dose males was 1/37 and 5/16, respectively (p<0.01). No renal tumors were observed in
females in any group. No metastases to the lung or renal lymph nodes were observed. Evidence of
neurotoxicity and renal pathology was observed in the treated mice at both dose levels. The high
mortality in both groups of treated males and in high-dose females indicated that the MTD was exceeded
in these groups.
3-69
-------
A follow-up study to the Mitsumori et al. (1981) study was reported by Hirano et al. (1986).
Methylmercuric chloride was administered in the diet to ICRmice (60/sex/group) at levels of 0, 0.4, 2, or
10 ppm (0, 0.03, 0.15, and 0.73 mg Hg/kg-day in males and 0, 0.02, 0.11, and 0.6 mg Hg/kg-day in
females) for 104 weeks. Interim sacrifices (6/sex/group) were conducted at 26, 52, and 78 weeks.
Complete histopathological examinations were performed on all animals found dead, killed in extremis,
or killed by design. Mortality, group mean body weights and food consumption were comparable to
controls. The first renal tumor was observed at 58 weeks in a high-dose male, and the incidence of renal
epithelial tumors (adenomas or adenocarcinomas) was significantly increased in high-dose males (1/32,
0/25, 0/29, and 13/26 in the control, low-, mid-, and high-dose groups, respectively). Ten of the 13
tumors in high-dose males were adenocarcinomas. These tumors were described as solid type or cystic
papillary types of adenocarcinomas. No invading proliferation into the surrounding tissues was seen.
The incidence of renal epithelial adenomas was not significantly increased in males, and no renal
adenomas or adenocarcinomas were observed in any females. Focal hyperplasia of the tubular
epithelium was reported to be increased in high-dose males (13/59; other incidences not reported).
Increases in nonneoplastic lesions in high-dose animals provided evidence that an MTD was exceeded.
Nonneoplastic lesions reported as increased in treated males included the following: epithelial
degeneration of the renal proximal tubules; cystic kidney; urinary cast and pelvic dilatation; and
decreased spermatogenesis. Epithelial degeneration of the renal proximal tubules and degeneration or
fibrosis of the sciatic nerve were reported in high-dose females.
Table 3-53
Carcinogenic Effects of Methylmercury in Animals: Oral Exposure
Species/
Strain/
No. per Sex
per Group
Rat/strain NS/ 25
M, 25 F
Rat/
Sprague
Dawley/56 M, 56
F
Mice/Swiss/
54 M, 54 F
Mouse/ICR/
60 M, 60 F
Mouse/ICR/
60 M, 60 F
Mouse/
B6C3F1/
60 M, 60 F
Exposure
Duration
2yr
ad lib in feed
130 wk
ad lib in FEFD
weaning until
death in
drinking water
78 wk ad lib
in feed
104wkadlib
in feed
2yr
ad lib in feed
Dose
(mg/kg-day)
0, 0.004, 0.02,
0.1
0.011,0.05,
0. 28 (M); 0.0 14,
0.064, 0.34 (F)
0,0.19,0.19-
0.95
(MMA)
0, 1.6,3.1
0, 0.02, 0.03,
0.11,0.15,
0.6, 0.73
0.03,0.14,0.69
(M); 0.03, 0.13,
0.6 (F)
Effects/Limitations/BML
Tumors at comparable incidence in all groups
Limitations: Small sample size; failure to achieve MTD
BML avg: 850 ug/L in blood at 0.004, 6,500 ug/L at
0.02, and 36,000-39,000 ug/L at 0.1
No increase in tumor incidence
No increase in gross tumor incidence
Limitation: Histological examination not performed.
Increased incidence renal adenomas and adenocarcinomas
in low-dose males.
Limitations: Very poor survival in both male dose
groups.
Incidence of renal epithelial adenocarcinoma significantly
increased in males at 0.73; not invasive.
Limitations: MTD exceeded (including severe renal
damage in high-dose males)
Renal epithelial carcinomas and adenomas in males at
0.69.
Limitation: MTD exceeded in high-dose males.
Reference
Verschuuren et al.
1976
Mitsumori et al.
1983, 1984
Schroeder and
Mitchener 1975
Mitsumori et al. 1981
Hirano etal. 1986
Mitsumori et al. 1990
3-70
-------
Table 3-53 (continued)
Carcinogenic Effects of Methylmercury in Animals: Oral Exposure
Species/
Strain/
No. per Sex
per Group
Mice/Swiss/
NS
Cat/domestic/
4-5 M, 4-5 F
Exposure
Duration
15 wk ad lib
in drinking
water
2yr
ad lib in feed
Dose
(mg/kg-day)
0, 0.03, 0.07,
0.27
(MMC)
0, 0.0084, 0.02,
0.046, 0.074,
0.176
Effects/Limitations/BML
Number of lung adenomas/mouse and tumor size/mouse
increased with dose
No increase in tumor incidence
Limitations: Small group size, short exposure duration,
no pathological data for 3 lowest doses.
Reference
Blakley 1984
Charbonneau et al.
1976
No increase in tumor incidence was observed in a study using white Swiss mice (Schroeder and
Mitchener 1975). Groups of mice (54/sex/group) were exposed from weaning until death to
methylmercuric acetate in the drinking water at two doses. The low-dose group received 1 ppm
methylmercuric acetate (0.19 mg Hg/kg-day). The high-dose group received 5 ppm methylmercuric
acetate (0.95 mg Hg/kg-day) for the first 70 days and then 1 ppm, thereafter, due to high mortality (21/54
males and 23/54 females died prior to the dose reduction). Survival among the remaining mice was not
significantly different from controls. Significant (p<0.001) reductions in body weight were reported in
high-dose males (9-15% lower than controls) and high-dose females (15-22% lower than controls)
between 2 and 6 months of age. Mice were weighed, dissected, gross tumors were detected, and some
sections were made of heart, lung, liver, kidney and spleen for microscopic examination. No increase in
tumor incidence was observed. This study is limited because complete histological examinations were
not performed, and pathology data other than tumor incidence were not reported.
Mitsumori et al. (1983, 1984) conducted a study in Sprague-Dawley rats. They administered
diets containing 0, 0.4, 2, or 10 ppm of methylmercuric chloride (0, 0.011, 0.05, and 0.28 mg Hg/kg-day
in males; 0, 0.014, 0.064 and 0.34 mg Hg/kg-day in females) to Sprague-Dawley rats (56
animals/sex/group) for up to 130 weeks. Interim sacrifices of 10/group (either sex) were conducted at
weeks 13 and 26 and of 6/group (either sex) at weeks 52 and 78. Mortality was increased in high-dose
males and females. At week 104, survival was approximately 55%, 45%, 75% and 10% in control, low-,
mid-, and high-dose males, respectively, and 70%, 75%, 75% and 30% in control, low-, mid-, and high-
dose females, respectively (data presented graphically). Body weight gain was decreased in high-dose
animals (approximately 20-30%; data presented graphically). No increase in tumor incidence was
observed in either males or females. Noncarcinogenic lesions that were significantly increased (p< 0.05)
in high-dose rats included the following: degeneration in peripheral nerves and the spinal cord (both
sexes); degeneration of the proximal tubular epithelium of the kidney (both sexes); severe chronic
nephropathy (females); parathyroid hyperplasia (both sexes); polyarteritis nodosa and calcification of the
abdominal arterial wall (females); bone fibrosis (females); bile duct hyperplasia (males); and
hemosiderosis and extramedullary hematopoiesis in the spleen (males). In addition, mid-dose males
exhibited significantly increased degeneration of the kidney proximal tubular epithelium and hyperplasia
of the parathyroid. An MTD was achieved in mid-dose males and in high-dose females; the MTD was
exceeded in high-dose males.
No increase in tumor incidence or decrease in tumor latency was observed in another study using
rats (strain not specified) (Verschuuren et al. 1976). Groups of 25 female and 25 male rats were
3-71
-------
administered methylmercuric chloride at dietary levels of 0, 0.1, 0.5 and 2.5 ppm (0, 0.004, 0.02 and 0.1
mg Hg/kg-day) for 2 years. No significant effects were observed on growth or food intake except for a
6% decrease (statistically significant) in body weight gain at 60 weeks in high-dose females. Survival
was 72%, 68%, 48% and 48% in control, low-, mid- and high-dose males, respectively; and 76%, 60%,
64% and 56% in control, low-, mid- and high-dose females, respectively (statistical significance not
reported). Increases in relative kidney weights were observed in both males and females at the highest
dose. No effects on the nature or incidence of pathological lesions were observed, and tumors were
reported to have been observed with comparable incidence and latency among all of the groups. This
study was limited by the small sample size and failure to achieve an MTD.
No tumor data were reported in a study using Wistar rats (Munro, 1980). Groups of 50 Wistar
rats/sex/dose were fed diets containing methylmercury; doses of 2, 10, 50, and 250 micrograms Hg/kg-
day were fed for 26 months. High-dose female rats exhibited reduced body weight gains and showed
minimal clinical signs of neurotoxicity; however, high-dose male rats showed overt clinical signs of
neurotoxicity, decreased hemoglobin and hematocrit values, reduced weight gains and significantly
increased mortality. Histopathologic examination of the high-dose rats of both sexes revealed
demyelination of dorsal nerve roots and peripheral nerves. Males showed severe dose-related kidney
damage, and females had minimal renal damage.
No increase in tumor incidence was observed in a multiple generation reproduction study using
Sprague-Dawley rats (Newberne et al. 1972). Groups of rats (30/sex) were given semisynthetic diets
supplemented with either casein or a fish protein concentrate to yield dietary levels of 0.2 ppm
methylmercury (0.008 mg Hg/kg-day). Another group of controls received untreated rat chow. Rats that
received diets containing methylmercury during the 2-year study had body weights and hematology
comparable to controls. Detailed histopathologic analyses revealed no lesions of the brain, liver, or
kidney that were attributable to the methylmercury exposure. Mortality data were not presented.
Interpretation of these data is limited by the somewhat small group sizes and failure to achieve an MTD.
No increase in tumor incidence was observed in a study using random-bred domestic cats
(Charbonneau et al. 1976). Groups of cats (4-5/sex/group) were given doses of 0.0084, 0.020, 0.046,
0.074 or 0.176 mg Hg/kg-day either as methylmercury-contaminated seafood or as methylmercuric
chloride in the diet for up to two years. Controls were estimated to have received 0.003 mg Hg/kg-day.
Food consumption and body weight were not affected by treatment with methylmercury. Due to
advanced signs of neurotoxicity (loss of balance, ataxia, impaired gait, impaired reflexes, weakness,
impaired sensory function, mood change and tremor), cats at the highest dose tested were sacrificed after
approximately 16 weeks, and cats at the next highest dose were sacrificed after approximately 54-57
weeks. Cats at the next highest dose generally exhibited mild neurological impairment (altered hopping
reaction and hypalgesia). One cat at this dose was sacrificed after 38 weeks because of neurotoxicity,
and one cat died of acute renal failure after 68 weeks. Cats at the two highest doses had pathological
changes in the brain and spinal cord, but no histopathological changes were noted in other tissues
examined. Interpretation of the results of this study is limited because of the small group sizes, early
sacrifice of cats at the two highest dose levels and no available data regarding pathological changes in
cats at the three lowest dose levels. This study was also limited by its short duration when compared to
the lifespan of a cat.
Blakley (1984) administered methylmercuric chloride to female Swiss mice (number/group not
specified) in drinking water at concentrations of 0, 0.2, 0.5 or 2.0 mg/L for 15 weeks. This corresponded
to approximately 0, 0.03, 0.07 and 0.27 mg Hg/kg-day. At the end of week 3, a single dose of 1.5 mg/kg
of urethane was administered intraperitoneally to 16-20 mice/group. No effects on weight gain or food
consumption were observed. Lung tumor incidence in mice not administered urethane (number/group
3-72
-------
not specified) was less than 1 tumor/mouse in all groups. Statistically significant trends for increases in
the number and size of lung adenomas/mouse with increasing methylmercury dose were observed; the
tumor number/mouse was 21.5, 19.4, 19.4 and 33.1 in control, low-, mid- and high-dose mice,
respectively, and the tumor size/mouse was 0.70, 0.73, 0.76 and 0.76 mm in control, low-, mid- and high-
dose mice, respectively. The study authors suggest that the increase in tumor number and size may have
been related to immunosuppressive activity of methylmercury. It should be noted that this is considered
a short term assay and that only pulmonary adenomas were evaluated.
3.3.3 Other Data
3.3.3.1 Death
Methylmercury is a potent toxin that causes impairment of the CNS and developmental toxicity
in humans. Ingestion of methylmercury from treated seed grain or contaminated fish has resulted in
death. An outbreak of methylmercury poisoning in Iraq caused deaths in people who consumed
methylmercury from bread made with grain treated with a fungicide (Al-Saleem and the Clinical
Committee on Mercury Poisoning 1976; Bakir et al. 1973). The deaths were attributed to impaired CNS
function. A syndrome known as Minamata disease has been characterized by nervous system impairment
following consumption of methylmercury-contaminated fish from Minamata Bay in Japan. Symptoms of
Minamata disease include the following: prickling; tingling sensation of extremities; impaired peripheral
vision, hearing, taste and smell; slurred speech; muscle weakness; irritability; memory loss; depression;
and sleeping difficulties (Kutsuna 1968; Takeuchi et al. 1962; Tsubaki and Takahashi 1986). Deaths
from Minamata disease can be broken into two categories: deaths occurring from the beginning of the
outbreak (1954) to 1969, and deaths occurring from 1970 to 1980 (Tamashiro et al. 1984). Over half of
the deaths in the first group were attributed to Minamata disease and/or noninflammatory disease of the
central nervous system, or pneumonia, whereas deaths in the second group were attributed to
cerebrovascular disease with underlying Minamata disease. The mean age at death for the first group
was 45.4 years for males and 26.4 years for females, and the mean age at death for the second group was
70.0 years for males and 72.7 years for females (Tamashiro et al. 1984).
Table 3-54
Lethality of Methylmercury in Humans: Case Study of Oral Exposure
Species/
No. per Sex
Human/6,530
both sexes
Human/1,422
both sexes
Exposure
Duration
43-68 d
(feed)
NS
Dose
(mg/kg-day)
0.71-5.7 (est.)
NS
Effects/Limitations/BML
Of 6,350 cases admitted to hospitals, 459 died after eating
bread made from grain treated with methylmercury
fungicide
BML: <100-5,000 ug/L in blood
Of 1,422 patients from the Minamata disease outbreak in
1959, 378 died by 1980.
Limitation: exposure data limited
BML not reported
Reference
Bakir etal. 1973
Tamashiro et al.
1984
Very little information regarding death after inhalation exposure to methylmercury was located.
One study reported a man who died after being exposed for three years to alkylmercury particles from
seed dressings (Hook et al. 1954). Prior to death, the man experienced increased symptoms of
neurotoxicity. A case study reported on the deaths of two women exposed to diethylmercury vapors for
3-73
-------
3-5 months (Hill 1943). Gastrointestinal effects and neurological symptoms occurred prior to deaths.
No other human studies were located regarding death after inhalation exposure to methylmercury.
Table 3-55
Lethality of Methylmercury in Humans: Case Studies of Inhalation Exposure
Species/
No. per Sex
Human/2 F
Human/ 1 M
Exposure
Duration
3-5 mo
(occup)
3yr
(occup)
Dose
(mg/m3)
NS
NS
Effects/Limitations/BML
Death following exposure to diethylmercury vapors
Limitation: Case study; concomitant dermal exposure likely;
limited exposure data
BML not reported
Death following exposure to pesticide containing
methylmercury
Limitations: Case study; concomitant dermal exposure likely;
limited exposure data
Range: 500-640 ug/L in urine
Reference
Hill 1943
Hooketal. 1954
Mice given a single oral dose of methylmercury had an increased incidence of death compared to
controls (Yasutake et al. 1991). Male mice appear to be more sensitive to the effects of methylmercury
than females, possibly due to the effect of mercury on the male kidneys. Mice exposed for 26 weeks to
3.1 mg Hg/kg-day as methylmercury in drinking water also showed an increase in mortality compared to
controls (51/60 males and 59/60 females of exposed group died versus 1/60 males and 1/60 females in
controls) (Mitsumori et al. 1981). Longer studies (78 and 104 weeks) confirm that methylmercury causes
significantly increased mortality in mice compared to controls (Mitsumori et al. 1981, 1990). No animal
studies were located on death after inhalation exposure to methylmercury.
Table 3-56
Lethality of Methylmercury in Animals
Species/
Strain/
No. per Sex
per Group
Mouse/ICR/60
M, 60 F
Mouse/
B6C3F/60 M,
60 F
Exposure
Duration
78 wk
ad lib in feed
104 wk
ad lib in feed
Dose
(mg/kg-day)
0, 1.6,3.1
(MMC)
0,0.03,0.13,
0.14, 0.60, 0.69
(MMC)
Effects/Limitations/BML
51/60 males and 59/60 females receiving 3.1 mg/kg/d
died by week 26, vs 7 males and 6 females at 1 .6, and 1
control males and 1 control female; death at 52 wk was
also elevated at 1.6
Limitation: No statistical analysis
BML not reported
50/60 males treated with 0.69 mg/kg/d died vs. 31/60
control males; survival of females and males at lower
doses was unaffected
BML not reported
Reference
Mitsumori et al.
1981
Mitsumori et al.
1990
3-74
-------
Table 3-56 (continued)
Lethality of Methylmercury in Animals
Species/
Strain/
No. per Sex
per Group
Mouse/
C57BL/6 M, 6
F
Exposure
Duration
Once
Dose
(mg/kg-day)
4, 8, 16, 24, 32,
40
Effects/Limitations/BML
4/6 males died (LOAEL = 16); LOAEL for females was
40 (4/6 died); no statistical analysis or LD50 calculated
Limitation: small number of animals tested
BML: 2.45 ug/g in kidney of males at 16 mg/kg
Reference
Yasutake et al. 1991
3.3.3.2 Neurological
The nervous system is the primary target organ for methylmercury toxicity. Information from the
large-scale poisonings in Japan (Niigata and Minamata) and Iraq provide substantial information
regarding the neurotoxicity of methylmercury in humans (Bakir et al. 1973, 1980; Berglund et al. 1971;
Harada 1978; Marsh et al. 1987; Rustam and Hamdi 1974). In Japan, poisonings occurred between 1953
and 1960 when people consumed seafood that had been contaminated by methylmercury released by a
chemical plant into Minamata Bay and the Agano river near Niigata. In Iraq, poisonings occurred in the
winter of 1971 to 1972 when people ate bread made from seed grain that had been treated with a
mercury-containing fungicide. In all of these episodes, neurotoxicity was the most prominent effect
observed in the exposed populations. In the Iraqi incident, more than 6000 patients were hospitalized,
and more than 500 deaths occurred, usually due to CNS failure.
The least severely affected persons from the poisonings in Japan and Iraq experienced numbness
or tingling (paresthesia) of the extremities and/or perioral area. Additional symptoms frequently
experienced by more severely affected individuals included the following: ataxia (gait impairment
ranging from mild incoordination or unsteadiness to complete inability to walk); blurred vision;
constriction of visual fields (in extreme cases blindness); slurred speech; and hearing difficulties
(deafness in extreme cases). Less frequently observed symptoms associated with the methylmercury
poisonings included tremors, muscular weakness, abnormal reflexes, increased muscle tone, and clouded
memory or stuperousness. A long latent period (16-38 days in the Iraqi episode and up to several years
in the Japanese episodes) between exposure and onset of symptoms of neurotoxicity was observed. The
cause for the latent period is unknown. It is thought that latency may be related to cellular repair
mechanisms dealing with damage from lipid peroxidation. At the point when repair processes are
overwhelmed tissue damage and accompanying symptoms become apparent. The possible ameliorating
effect of selenium in the diet has also been hypothesized to play a part in latency.
Similar neurological symptoms have been observed in persons ingesting meat contaminated with
ethylmercuric chloride (Cinca et al. 1979). Two boys who ultimately died from exposure exhibited
neurological signs including gait disturbance, ataxia, dysarthria, dysphagia, aphonia, hyperactive tendon
reflexes, hypotonia, mydriasis and agitation. In the surviving members of the family, ataxia, gait
impairment, spasticity, drowsiness, intention tremor, agitation, speech difficulties and visual disturbances
were reported. All effects except the narrowing of the visual fields disappeared after exposure
termination.
3-75
-------
Histopathologic analyses of nervous system tissue taken from poisoning victims show neuronal
degeneration in the cerebrum and cerebellum (Bakir et al. 1980; Swedish Expert Group 1971; Takeuchi
et al. 1962). In the cerebral cortex, the calcarine area was most regularly affected with varying degrees
of damage in the pre- and postcentral cortices, superior temporal gyrus, and basal ganglia. In the
cerebellar cortex, granule cell loss predominated, but this was usually less severe than cerebral damage.
An autopsy of two boys who ingested ethyl mercury contaminated meat revealed nerve cell loss and glial
proliferation in the cerebral cortex, demyelination, granule cell loss in the cerebellum, and motor neuron
loss in the ventral horns of the spinal cord (Cinca et al. 1979). Less information is available regarding
the histopathology of peripheral nerve involvement, but sural nerves taken from two victims of the
Minamata episode showed evidence of peripheral nerve degeneration and regeneration (Miyakawa et al.
1976). Fourteen Iraqi patients who developed ataxia and "pins and needles" and could not perform heel-
to-toe walk were examined for impaired peripheral nerve function (Von Burg and Rustam 1974a, 1974b).
Determinations of motor and sensory conduction velocities, sensory threshold and latency, reflex of the
tibial nerve and myoneural transmission were performed, but there were no statistical significances
between exposed and unexposed control groups; the mean values of the experimental group, however,
were somewhat lower than those of the controls. There was also no consistent correlation between
clinical or electrophysiological observation on the peripheral nervous system and blood mercury levels.
In two patients who were hospitalized 10 days after ingestion of ethyl mercury-contaminated meat,
sensory nerve conduction velocity was decreased immediately after admission but was found to be
normal six months later (Cinca et al. 1979).
3-76
-------
Table 3-57
Neurotoxicity of Methylmercury in Humans: Case Studies of Oral Exposure
Species/
No. per Sex
Human/ 14
cases
Human/No. NS
Human/2 M, 2
F
Human/6530
cases both
sexes
Human/81 F
Exposure
Duration
NS
NS
Once
43-68 d
<5 mo
Dose
(mg/kg-day)
NS
NS
NS
(ethyl mercury
chloride)
0.71-5.7 (est.)
NS
(MMC)
Effects/Limitations/BML
Ataxia; impaired heel-to-toe walk; complaints of "pins and
needles" . Sensory and motor peripheral nerves were not
affected. Clinical and electrophysiological observations
did not correlate with blood concentration
Limitation: Exposure concentration and duration not
known
BML: blood Hg levels were 138-878 ug/L
Paresthesia/numbness; constriction of visual field;
incoordination; difficulty speaking; tremor in consumers
of contaminated fish
Limitation: Limited details reported
BML not reported
Gait disturbance, ataxia, dysarthria, speech difficulties,
visual disturbances, hyperactive tendon reflexes,
mydriasis, agitation, coma; nerve degeneration in cerebral
cortex, cerebellum, and ventral horns of spinal cord;
decreased sensory nerve conduction velocity. Ingestion of
ethyl mercury chloride-contaminated meat.
Limitation: Exposure concentration not known
BML: hair Hg levels of 152-542 ug/g
Paresthesia/numbness in extremities and perioral area;
ataxia; constriction of visual field or blindness; slurred
speech; hearing difficulties following ingestion of grain
contaminated with methylmercury. Incidence and severity
of effect correlated with blood concentration
BML: Total body burden > 50 mg at time of onset
Paresthesia and "other neurological symptoms"
BML Range: 1-674 ug/g Hg in hair; one woman with 14
ug/g (maximum in strand) had paresthesia and a woman
with 10 ug/g had other symptoms. However, others with
levels as high as 600 ug/g had no symptoms. (This is a
follow-up study to Bakir et al. 1980)
Reference
Von Burg and
Rustam 1974a,
1974b
Harada 1978
Cincaetal. 1979
Bakir et al. 1973,
1980
Marsh etal. 1987
There are two case studies that report neurotoxicity in humans following inhalation of
methylmercury (Hook et al. 1954; Hunter et al. 1940); however, no quantitative data were available. The
two studies described in Table 3-58 demonstrated the spectrum of neurotoxic effects that occur following
occupational exposure to methylmercury. Weiss and Simon (1975) have suggested that such changes in
function in the general population, particularly at relatively low doses, may not be clinically detectable as
a loss of function but may be unmasked by the normal processes of aging.
3-77
-------
Table 3-58
Neurotoxicity of Methylmercury in Humans: Case Studies of Inhalation Exposure
Species/
No. per Sex
Human/5 M
Human/ 1 M
Exposure
Duration
5 mo-2 yr
(occup)
3yr
(occup)
Dose
(mg/m3)
NS
NS
Effects/Limitation/BML
Tingling of limbs; unsteady gait; difficulty performing fine
movements; constricted visual field following exposure to
methylmercury nitrate, methylmercury iodide, of
methylmercury phosphate in chemical factories
Limitations: Case studies; concomitant dermal exposure and
exposure to other chemicals likely; limited exposure data
BML not reported
Weakness in arms and legs; irregular EEG; sensory and speech
disorders following exposure to pesticide containing
methylmercury
Limitations: Case study; concomitant dermal exposure likely;
limited exposure data
BML Range: 500-640 ug/L in urine
Reference
Hunter etal. 1940
Hooketal. 1954
As a result of the methylmercury poisonings in Japan and Iraq, substantial information on the
neurotoxicity of methylmercury has been generated from animal studies. Relatively brief, high level
exposures in rats have been shown to cause characteristic signs of neurotoxicity (flailing and hindlimb
crossing when the animal is lifted by the tail) and neuronal degeneration in the cerebellum, cerebral
cortex and dorsal root ganglia (Inouye and Murakami 1975; Leyshon and Morgan 1991; Magos et al.
1985; Yip and Chang 1981). As with humans there is a latency period; the effects frequently are not
observed or do not show maximal severity until several days after the cessation of dosing. In an acute
study, exposure of rats to a single gavage dose of 19.9 mg Hg/kg as methylmercuric chloride resulted in
impaired open-field tests such as decreases in standing upright, area traversed and activity compared to
the control group (Post et al. 1973). Animals were lethargic and ataxic initially, but symptoms
disappeared within 3 hours.
Longer-term, lower-level exposures revealed that evidence of neuronal degeneration may be
observed prior to the onset of overt signs of toxicity. Degeneration in the cerebellum was found in rats
given 10 mg Hg/kg as methylmercuric chloride once every 3 days for 15 days (Leyshon and Morgan
1991) while severe degenerative changes in the dorsal root fibers were observed in rats given 1.6 mg
Hg/kg-day as methylmercuric chloride for 8 weeks (Yip and Chang 1981). Munro et al. (1980) observed
demyelination of dorsal nerve roots and damage in sciatic nerves with oral exposure to 0.25 mg Hg/kg-
day as methylmercuric chloride for up to 26 months. In mice given 1.9 mg Hg/kg-day as methylmercury,
cerebellar lesions were observed as early as eight days after the start of dosing, but changes in motor
activity did not develop until 24 weeks of exposure (MacDonald and Harbison 1977). Similarly, cats
receiving methylmercury in the diet for 11 months displayed degenerative changes in the cerebellum and
cerebral cortex, but incoordination or weakness was observed in only a small number of the animals with
histopathological changes (Chang et al. 1974).
The molecular basis for methylmercury neurotoxicity is likely to be complex and multifactorial.
The broad affinity of mercury for -SH groups leads to membrane, enzyme and cytoplasmic organelle
interaction. Major mechanistic pathways have been proposed to include the following: inhibition of
macromolecular metabolism, especially that of protein translation and nucleic acid biogenesis; oxidative
3-78
-------
injury; disturbance in Ca2+ hemostasis; aberrant protein phosphorylation. The mechanisms underlying
inhibition of protein and RNA synthesis are multiple. Depending upon the systems used with in vitro, in
vivo or neuronal cell suspensions, evidence for inhibition of translation associated with a change in
ATP/ADP concentration has been found. On the other hand, direct inhibition of elongation was
documented secondary to the selective inhibition of certain aminoacyl-tRNA synthetase (Cheung and
Verity, 1985). Syversen (1977) investigated the effects of methylmercury on protein synthesis in rats
using techniques which allowed analysis of different cell populations from the central nervous system.
Results of this study indicated selective irreversible damage to granule cells of the cerebellum, whereas
damage to the other neurons, such as Purkinje cells was reversible. Such selectivity of toxicity is a
feature of the neuronal loss seen in human and experimental disease. Methylmercury has also been
suggested to cause neuronal degeneration by promoting the formation of reactive oxygen species (Ali et
al. 1992; Le Bel et al. 1990, 1992; Verity and Sarafian 1991). While contributory, such oxidative injury
does not appear primary to the site of toxicity as appropriate protective measures blocking oxidative
stress and lipoperoxide formation are only minimally cytoprotective.
A recent review by Atchison and Hard (1994) discusses several proposed mechanisms of action
of methylmercury on Ca2+ hemostasis and ion channel function. Individual studies have demonstrated
that the neuromuscular actions of methylmercury occur predominantly at the presynaptic site (Atchison
et al. 1984). Methylmercury may interfere with acetylcholine neurotransmitter release and subsequently
synaptic transmission (Atchison et al. 1986; Barrett et al. 1974; Schafer et al. 1990; Schafer and Atchison
1989, 1991). Finally, Sarafian (1993) demonstrated that the methylmercury-induced stimulation of
protein phosphorylation in cerebellar granule cell culture is coupled to Ca2+ uptake, changed intracellular
Ca2+ hemostasis and inositol phosphate metabolism. These latter observations invoke the activation of
the protein kinase C pathway.
Cats and monkeys appear to be more sensitive to the neurotoxic effects of methylmercury than
rodents. Long-term studies in primates and in cats have shown neurological impairment at doses as low
as 0.05 mg Hg/kg-day. In cats, mild impairment of motor activity and decreased pain sensitivity was
observed at 0.046 mg Hg/kg-day as methylmercury after 60 weeks of exposure (Charbonneau et al.
1976). In cynomolgus monkeys given methylmercury from birth until approximately 7 years of age,
impairment of spatial visual function was observed after 3 years, and decreased fine motor performance,
touch and pinprick sensitivity and impaired high frequency hearing were observed six to seven years after
cessation of dosing (Rice 1989b; Rice and Gilbert 1982, 1992). Exposure of cynomolgus monkeys to
0.03 mg Hg/kg-day as methylmercury for approximately 4 months caused no detectable changes in motor
activity or effects on vision or hearing, but degenerative changes were observed in neurons of the
calcarine cortex and sural nerve when these were examined electron microscopically (Sato and Ikuta
1975). At higher doses (0.08 mg Hg/kg-day), slight tremor, motor incoordination and blindness were
observed in Macaca fascicularis monkeys after four months of exposure (Burbacher et al. 1988).
The developing organism is generally at higher risk of neurotoxicity than adults. The section on
developmental effects of methylmercury lists studies wherein animals were observed with neurological
or neurobehavioral deficits as a consequence of in utero or perinatal methylmercury exposure.
3-79
-------
Table 3-59
Neurotoxicity of Methylmercury in Animals
Species/
Strain/
No. per Sex
per Group
Rat/Wistar/
10 F
Rat/Wistar/
50 F, 50 M
Rat/Charles
River/6 M
Rat/Wistar/24
M, 18 F
Rat/Wistar/ 15
M
Swiss origin
Mouse
M
Cat/Breed
NS/15-16both
sexes
Cat/Breed
NS/8-10 NS
Monkey/
Macaca
fascicularis/l-2
both sexes
Monkey/
Macaca
artoides,
Macaca
nemestrinall
both sexes
Exposure
Duration
0-12 or
12-20 d, Ix/d
(gavage)
up to 26 mo
ad lib in feed
8wk
7d/wk
Ix/d
(gavage)
5d
Ix/d
(gavage)
5x/15d
(gavage)
28 wk
(ad lib
drinking
water)
11 mo
(ad lib in
feed)
2yr
7d/wk
(feed)
36-132 d
Ix/d
(feed)
90-270 d
Ix/wk
(gavage)
Dose
(mg/kg-day)
2,4
(MMC)*
0.002, 0.01,
0.05, 0.25
(MMC)*
0, 1.6
(MMC)*
8
(MMC)*
0, 10
(MMC)*
1.9,9.5
(MMC)*
0,0.015
(MM)
0.003, 0.008,
0.020, 0.046,
0.074, 0.176
(MMC)*
0.02, 0.03,
0.04, 0.07, 0.21
1 for 5 doses,
then
0.4,0.5,0.6
Effects/Limitations/BML
Hindlimb crossing (LOAEL = 4) after 0-12 days
BML not reported
Ruffled fur, loss of balance, hindlimb crossing, paralysis
(LOAEL = 0.25) after 6 mo (males more affected);
demyelination of dorsal nerve roots and damage in teased
sciatic nerves at 0.25
Avg. BML at 0.25: 1 15 ppm in blood
Degeneration of dorsal root fiber
BML not reported
Cerebellar granule cell and dorsal root ganglion cell
degeneration; flailing and hind leg crossing following
administration of methylmercuric chloride
Limitations: Only one level tested; no controls
Avg BML: 150,000 ug/L in blood
Granule cell degeneration in cerebellum
BML: 60 ug/g dry cerebellar weight
Ataxia; degenerative changes of Purkinje cells; granule
cell loss in cerebellum; (LOAEL = 1.9)
BML not reported
Degeneration of cerebellum and cerebral cortex; necrosis
of dorsal root ganglia of kittens fed mercury-contaminated
tuna
BML not reported
Impaired hopping reaction; decreased pain sensitivity;
degeneration of dorsal root ganglia (LOAEL = 0.046)
Avg BML: 9,000 ug/L in blood at 0.046 mg/kg-day
Atrophy of neurons in calcarine cortex; focal degeneration
in sural nerves (LOAEL=0.03); ataxic gait, myoclonic
seizures at 0.21 mg/kg-day
Limitation: small number of animals tested
BML: Maximal at 0.03 mg/kg-day of 460 ug/L in blood
and 62 ug/g in hair
Tremor; visual impairment (LOAEL = 0.5 mg/kg)
Limitations: Small number of animals tested, limited
description of effects
Avg BML: 2,900 ug/L in blood
Reference
Inouye and
Murakami 1975
Munro et al. 1980
Yip and Chang 1981
Magosetal. 1985
Leyshon and
Morgan 1991
MacDonald and
Harbison 1977
Chang etal. 1974
Charbonneau et al.
1976
Sato and Ikuta 1975
Evans et al. 1977
3-80
-------
Table 3-59 (continued)
Neurotoxicity of Methylmercury in Animals
Species/
Strain/
No. per Sex
per Group
Monkey/
Macaca
fascicularis!5
exposed, 2
control (sex
NS)
Monkey/
Macaca
fascicularis/7-8
F
Monkey/
Macaca
fascicularis/4
M, IF
exposed, 1 M, 2
F controls
Exposure
Duration
3-4 yr
7d/wk
Ix/d
(NS)
~3yr
Ix/d
(oral route
NS)
6.5-7 yr
7d/wk
Ix/d
(capsule;
gavage)
Dose
(mg/kg-day)
0, 0.05
(MMC)*
0, 0.04, 0.06,
0.08
(MMC)*
0, 0.05
(MMC)*
Effects/Limitations/BML
Spatial visual impairment
Limitation: One dose level tested
BML: 600-900 ug/L in blood
Slight tremor; motor incoordination; blindness (LOAEL =
0.04) following administration of methylmercury
hydroxide; time to onset was 177-395 d
Avg BML: 2,030 ug/L in blood at highest dose
Six years after end of dosing (follow-up study to Rice and
Gilbert 1982): decreased fine motor performance;
diminished touch and pinprick sensitivity; impaired high
frequency hearing (p<0.05)
Limitations: small number of animals tested; one dose
level tested
BML: Not detectable at time of testing
Reference
Rice and Gilbert
1982
Burbacher etal.
1988
Rice 1989b; Rice
and Gilbert 1992
*MMC = methylmercuric chloride
3.3.3.3 Renal
No studies were located regarding the renal toxicity of methylmercury in humans following oral
exposure. Renal histopathology and decreased function have been observed following acute or chronic
oral exposure of rats and mice to methylmercury. Renal tubule vacuolation was observed in rats
receiving 8 mg Hg/kg-day for 5 days (Magos et al. 1985), and decreased phenolsulfonphthalein excretion
occurred in male mice receiving a single dose of 16 mg Hg/kg-day or greater and females at 32 mg
Hg/kg-day or greater as methylmercuric chloride (Yasutake et al. 1991). Chronic nephropathy, including
epithelial degeneration of proximal tubules and interstitial fibrosis, was observed at longer durations
(Fowler 1972; Hirano et al. 1986; Mitsumori et al. 1990). Males were more sensitive than females to
renal effects (Mitsumori et al. 1990).
3-81
-------
Table 3-60
Renal Toxicity of Methylmercury in Animals
Species/
Strain/
No. per Sex
per Group
Rat/Wistar/3
M, 6F
exposed/16
controls (sex
NS)
Rat/Wistar/24
M, 18 F
Mouse/ICR/60
M, 60 F
Mouse/
B6C3F/60 M,
60 F
Mouse/
C57BL/6 M, 6
F
Exposure
Duration
12 wk
ad lib in feed
5d
Ix/d
(gavage)
26 wk
ad lib in feed
104 wk
ad lib in feed
Once
(gavage)
Dose
(mg/kg-day)
0, 0.08 (M)
0, 0.09 (F)
(MMC)
8
0,0.03,0.15,
0.72 (M); 0.02,
0. 11, 0.62 (F)
0, 0.03, 0.14,
0.68 (M); 0.03,
0.13, 0.6 (F)
(MMC)
4, 8, 16, 24, 32,
40 (MMC)
Effects/Limitations/BML
Cytoplasmic mass in proximal tubule cells
Limitation: Only one level tested; small number of treated
animals
BML not reported
Renal tubule vacuolation and dilation
Limitation: One level tested, no controls
Avg. BML: 150,000 ug/L in blood
Toxic epithelial degeneration of renal proximal tubules
(LOAEL = 0.62 F; 0.72 M)
BML not reported
Chronic nephropathy (epithelial cell degeneration,
regeneration of proximal tubules, interstitial fibrosis) in
males at >0.14 and in females at 0.60 (p<0.01)
BML not reported
Decreased phenolsulfonphthalein excretion and increased
serum creatinine in males (LOAEL = 16 in males, 32 in
females); swollen epithelial cells in proximal tubules
Limitation: No statistical analysis; small number of
treated animals
BML: 2.45 ug/g in kidneys of males and 1.9 ug/g in
kidneys of females at 16 mg/kg
Reference
Fowler 1972
Magosetal. 1985
Hiranoetal. 1986
Mitsumori et al.
1990
Yasutake et al. 1991
3.3.3.4 Cardiovascular
Only one study was located regarding the cardiovascular toxicity of methylmercury in humans.
Hook et al. (1954) reported two men with elevated blood pressure after inhalation exposure to organic
mercury particulates from seed dressings. Other neurotoxic effects were also present at the time of
examination, and one man subsequently died.
3-82
-------
Table 3-61
Cardiovascular Toxicity of Methylmercury in Humans: Case Study
Species/
No. per Sex
Human/ 1 M
Exposure
Duration
3yr
(occup)
Dose
(mg/m3)
NS
Effects/Limitations/BML
Elevated blood pressure
Limitations: Case study; concomitant dermal exposure likely
BML Range: 500-640 ug/L in urine
Reference
Hook et al. 1954
Very little information was located regarding the effects of oral methylmercury exposure on the
cardiovascular system. Rats given two daily doses of methylmercuric chloride exhibited decreases in
heart rates (Arito and Takahashi 1991). Rats treated with methylmercuric chloride for one month had
increased systolic blood pressures beginning 42 days after cessation of dosing (Wakita 1987). This effect
persisted for more than a year.
Table 3-62
Cardiovascular Toxicity of Methylmercury in Animals
Species/
Strain/
No. per Sex
per Group
Rat/Wistar/10
(sexNS)
Rat/Sprague-
Dawley/5-6
(sexNS)
Exposure
Duration
23-28 d
7d/wk
(gavage)
2d
Ix/d
(gavage)
Dose
(mg/kg-day)
0.4, 1.2
(MMC)
12
(MMC)
Effects/Limitation/BML
Increased systolic pressure beginning 42 d after the end of
treatment (p<0.05)
BML not reported
Decreased heart rate (p<0.05)
Limitation: Only one dose tested for this parameter
BML: 10 ug/g in brain
Reference
Wakita 1987
Arito and Takahashi
1991
3.3.3.5 Gastrointe stinal
No information was located regarding the gastrointestinal toxicity of methylmercury in humans.
Only one study was located regarding the gastrointestinal toxicity of methylmercury following oral
exposure in animals. Mitsumori et al. (1990) reported an increased incidence of stomach ulceration in
mice following a 2-year exposure to 0.69 mg Hg/kg-day as methylmercuric chloride in drinking water.
3-83
-------
Table 3-63
Gastrointestinal Toxicity of Methylmercury in Animals
Species/
Strain/
No. per Sex
per Group
Mouse/
B6C3F/60 M,
60 F
Exposure
Duration
104 wk
ad lib in feed
Dose
(mg/kg-day)
0, 0.03, 0.14,
0.69 (M); 0.03,
0.13, 0.6 (F)
(MMC)
Effects/Limitations/BML
Stomach ulceration in males at 0.69 (p<0.05)
BML not reported
Reference
Mitsumori et al.
1990
3.3.3.6 Immunological
Suppression of the humoral and cellular immune responses have been observed in animals after
oral exposure to methylmercury or methylmercuric chloride. Both decreases in the production of
antibody-producing cells and decreased antibody titre in response to inoculation with immune-
stimulating agents (such as sheep red blood cells) have been observed (Blakley et al. 1980; Koller et al.
1977; Ohi et al. 1976). Decreases in natural killer T-cell activity have been observed in animals after
exposure to methylmercury (Ilback 1991).
Table 3-64
Immunotoxicity of Methylmercury in Animals
Species/
Strain/
No. per Sex per
Group
Rat/Brown Norway/6
both sexes exposed/22
both sexes/controls
Mouse/ICR/6 M
Mouse/Swiss/8-10 M
Exposure
Duration
NSx/wk
2 mo
5d
Ix/d
(gavage)
3 wk
ad lib in
drinking water
Dose
(mg/kg-day)
0,4.8
(MMC)
0.27, 2.7
(MMC)
0.076,0.3, 1.52
(MMC)
Effects/Limitations/BML
IgG deposits along the glomerular capillary wall of
the kidney, not in arteries, suggestive of an
autoimmune disease; no effect seen in controls.
Limitation: only one level tested
BML not reported
Decreased production of antibody-producing cells
(LOAEL = 2.7;p<0.01).
Limitation: small number of animals, only males
tested
BML not reported
Decreased production of antibody-producing cells
and decreased antibody liter (LOAEL = 0.076;
p<0.01).
Limitation: small number of animals, only males
tested
BML not reported
Reference
Bernaudin et al 1981
Ohietal. 1976
Blakley etal. 1980
3-84
-------
Table 3-64 (continued)
Immunotoxicity of Methylmercury in Animals
Species/
Strain/
No. per Sex per
Group
Mouse/Balb/c CUM/
8F
Rabbit/New Zealand
white/ 10 M, 10 F
Exposure
Duration
12 wk
ad lib in feed
14 wk
Ix/d
in feed
Dose
(mg/kg-day)
0,0.5
0.04, 0.4, 0.8
(MMC)
Effects/Limitations/BML
Reduced natural killer T-cell activity; decreased
thymus weight and cell number (p<0.01).
Limitation: small number of animals treated, only
females tested
BML not reported
Decreased antibody liter (LOAEL = 0.4) (26% of
the animals at 0.4 and no controls died by wk 14).
Limitations: No statistical analysis
BML: 2,240 ug/L in blood at 0.4 mg/kg/d at wk
14
Reference
Ilback 1991
Koller et al. 1977
3.3.3.7 Dermal
Al-Mufti et al. (1976) studied the effects of methylmercury in humans who ate contaminated
bread; a correlation between bread consumption and a history of rash was reported. No other information
was located regarding dermal effects of organic mercury following oral exposure.
Table 3-65
Dermal Toxicity of Methylmercury in Humans: Epidemiological Study
Species/
No. per Sex
Human/415
exposed/1012
controls (sex
NS)
Exposure
Duration
= 1-3 mo
(feed)
Dose
(mg/kg-day)
NS
(MMC)
Effects/Limitations/BML
"History of rash" in 14% of exposed group, compared with
-------
appeared normal at birth but within several months exhibited mental retardation, retention of primitive
reflexes, cerebellar symptoms, dysarthria, hyperkinesia, hypersalivation, strabismus and pyramidal
symptoms (Harada 1978). Similarly, infants born to mothers who had ingested bread made with seed
grain treated with methylmercury-containing fungicides in Iraq during 1971 to 1972 exhibited symptoms
ranging from delays in speech and motor development to mental retardation, reflex abnormalities and
seizures (Amin-Zaki et al. 1974, 1978). Histopathologic analyses of brain tissues from infants that died
in the Iraqi (Choi et al. 1978) and Minamata (Harada 1978) episodes showed atrophy and hypoplasia of
the cerebral cortex, corpus callosum and granule cell layer of the cerebellum; dysmyelination of the
pyramidal tracts; and/or abnormal neuronal cytoarchitecture characterized by ectopic cells and
disorganization of cellular layers.
A number of studies have attempted to evaluate developmental neurotoxicity in populations with
elevated methylmercury exposure from consumption offish as a major component of the diet but for
whom massive poisonings have not been reported. Kjellstrom et al. 1989) observed a higher incidence of
abnormal scoring on tests designed to assess intelligence and development among children from New
Zealand whose mothers had high levels of hair mercury. Also a study by McKeown-Eyssen et al. (1983)
of a Cree population from northern Quebec revealed a correlation between maternal exposure (as
determined using hair levels) and abnormal muscle tone or reflexes in male children. A dose-response
for this effect was not observed.
Dose-response analyses of human data from the Iraqi epidemic of 1971 to 1972 have indicated
correlations between maximal maternal hair levels during pregnancy and the severity of the neurological
deficits seen in the children (Cox et al. 1989; Marsh et al. 1981, 1987). An evaluation of a calculated
threshold for response is presented in Section 6.3.1 of this volume.
Table 3-66
Developmental Toxicity of Methylmercury in Humans: Case Studies
Species/
No. per Sex
Human/8 M, 7
F infants
Human/ 1 F
Exposure
Duration
~2mo.
(feed)
6 mo.
3 mo.
postcoital-
term
(feed)
Dose
(mg/kg-day)
NS
NS
Effects/Limitations/BML
Assessment of 15 mother-infant pairs where the mothers
ate grain treated with methylmercury fungicide during
pregnancy. Motor and mental development were impaired
(blindness, impaired hearing) in 6 infants; there were no
congenital malformations.
BML: Affected infants: -3,000 ug/L in blood at 2 months;
Affected mothers: >400 ug/L in blood
Severe neurological impairment (blindness, myoclonic
seizures, spastic quadriparesis) of male infant born to a
mother eating meat from pigs that had eaten grain treated
with methylmercury fungicide.
Limitation: Case report
BML not reported
Reference
Amin-Zaki et al.
1974
Snyder and
Seelinger 1976
3-86
-------
Table 3-67
Developmental Toxicity of Methylmercury in Humans: Epidemiologic Studies
Species/
No. per Sex
Human/220 F
Human/84
mother-child
pairs
Human/243
exposed (sex
NS) aged 12-30
mo.
Human/8 1
mother-child
pairs
Exposure
Duration
NS
(food)
few days to
several mo.
(food)
Gestation and
lactation
(food)
few days to
several mo.
(food)
Dose
(mg/kg-day)
NS
NS
NS
NS
Effects/Limitations/BML
Mental retardation, atrophy of brain and degeneration of
cerebellum in offspring. Of 220 infants born in Minamata
(to mothers eating contaminated fish), 13 had severe
symptoms; the number with less severe symptoms was not
reported.
Limitations: Few details on methods or results
BML not reported
Assessment of mother-infant pairs where mothers ate grain
treated with methylmercury fungicide during pregnancy
(same Iraqi population as reported by Amin-Zaki et al.
1974). Severe psychomotor retardation in infants.
BML Range: 37-293 ug/g in hair (maximum in segment of
maternal hair)
Abnormal tendon reflexes or muscle tone in male offspring
correlated with methylmercury exposure (p<0.05).
Conducted as a case-control study after potential affected
measures were identified.
Limitation: Author reported that the statistical method
could have led to an association by chance.
BML avg: 6 ug/g in maternal hair
Assessment of mother-infant pairs where mothers ate grain
treated with methylmercury fungicide during pregnancy
(same Iraqi population as reported by Amin-Zaki et al.
1974). Delayed walking and talking; seizures; mental
retardation.
BML Range: —18-598 ug/g (maximum in strand) in hair
of mothers of affected infants
Reference
Harada 1978
Marsh etal. 1981
McKeown-Eyssen et
al. 1983
Marsh etal. 1987
The developmental toxicity of oral exposure to methylmercury has been extensively studied in
animals. In rodents exposed in utero, a spectrum of effects has been observed ranging from decreases in
fetal weight and skeletal ossification and increases in skeletal variations and malformations (brain
lesions, hydrocephalus, cleft palate, micrognathia, edema, subcutaneous bleeding, hydronephrosis,
hypoplasia of the kidneys, dilation of the renal pelvis) to increased resorptions and fetal deaths (Fuyuta et
al. 1978, 1979; Inouye and Kajiwara 1988a; Inouye and Murakami 1975; Khera and Tabacova 1973;
Nolen et al. 1972; Reuhl et al. 1981; Yasuda et al. 1985). The severity of the effects generally increased
with dose, and the incidence of malformations increased with exposures that occurred later in gestation
(Fuyuta et al. 1978; Inouye and Murakami 1975). Brain lesions have been observed in a variety of areas
including the brain mantle, corpus callosum, caudate putamen and cerebellum. In guinea pigs, early
gestational exposures (weeks 3-5 of pregnancy) resulted primarily in developmental disturbances of the
brain (smaller brains, dilated lateral ventricles and reduced size of caudate putamen), whereas later
gestational exposures (>week 6 of pregnancy) resulted in widespread neuronal degeneration (Inouye and
Kajiwara 1988b).
In addition to structural changes, functional changes have been observed in animals after
gestational exposures. Such functional effects include abnormal tail position during walking; flexion;
3-87
-------
hindlimb crossing; decreased locomotor activity, responding in an avoidance task and righting response;
increased passiveness, startle-response and sensitivity to pentylenetetrazol-induced convulsions; and
impaired maze performance, operant behavior, swimming behavior, tactile-kinesthetic function, visual
recognition memory, temporal discrimination, and subtle learning deficits such as insensitivity to
changing reinforcement contingencies (Bornhausen et al. 1980; Buelke-Sam et al. 1985; Burbacher et al.
1990; Eisner 1991; Geyer et al. 1985; Gunderson et al. 1988; Hughes and Annau 1976; Inouye et al.
1985; Musch et al. 1978; Olson and Boush 1975; Rice 1992; Rice and Gilbert 1990; Stoltenburg-
Didinger and Markwort 1990; Suter and Schon 1986; Newland et al. 1994).
Overt neurological impairment is the endpoint used to document methylmercury poisonings;
however, as shown in animal studies, methylmercury may produce more subtle neurodevelopmental
effects such as impairment of sensory or cognitive systems. Schreiner et al. (1986) exposed rats to 0, 0.2
or 0.6 mg Hg/kg-day as methylmercuric chloride in utero and during lactation to evaluate pup
performance on visual discrimination reversal task. While no overt signs of neurotoxicity were evident,
subtle differences between the control and high-dose group were observed during more difficult tasks. A
stressful or highly demanding situation appears to be necessary for the expression of these sensory
effects, wherein the decreased ability to adapt to the altered conditions became manifest. Spyker et al.
(1972) reported that although no signs of neurological toxicity was observed in mouse pups exposed to
methylmercury in utero, open field and swimming tests revealed subtle neurological effects in the
exposed pups. Newland et al. (1994) administered methylmercury by gavage to pregnant squirrel
monkeys between weeks 11 and 14.5 of gestation. Doses were adjusted to maintain 0.7 to 0.9 ppm Hg in
the maternal blood. There were three controls and three methylmercury-treated offspring. Offspring
were evaluated at 5-6 on a lever pressing test which required discrimination between degrees of
reinforcement. At steady state, monkeys exposed to methylmercury in utero were less sensitive to
differences in reinforcement rates. When reinforcement rates changed, exposed animals either changed
their behavior slowly in response to the altered reinforcement or not at all.
The developmental toxicity of methylmercury may be attributable to the ability of
methylmercury to bind to sulfhydryl-rich tubulin (a protein component of microtubules) and cause its
depolymerization (Falconer et al. 1994; Sager et al. 1983). Both cell division and cell migration require
intact microtubules for normal functioning. Disruption of microtubule function could result in the
derangement of cell migration (Choi et al. 1978; Falconer et al. 1994; Matsumoto et al. 1965) and
arrested cell division (Reuhl et al. 1994; Sager et al. 1984).
Table 3-68
Developmental Toxicity of Methylmercury in Animals
Species/
Strain/
No. per Sex
per Group
Rat/Charles
River/20 F
Exposure
Duration
9d
Gd6-14
ad lib in
drinking
water
Dose
(mg/kg-day)
0, 0.02, 0.2, 4
Effects/Limitations/BML
Increased number of fetuses with soft tissue variations of
the urinary system and incomplete ossification or
calcification (LOAEL = 4; p<0.05).
BML not reported
Reference
Nolenetal. 1972
3-S
-------
Table 3-68 (continued)
Developmental Toxicity of Methylmercury in Animals
Species/
Strain/
No. per Sex
per Group
Rat/Wistar/35 F
Rat/Wistar/10 F
Rat/Holtzman/5
F
Rat/Charles
River CD/20 F
Rat/Long-
Evans/4 exposed,
6 control
Rat/Wistar/20 F
Rat/Wistar-
Neuherberg/
No. F. NS
Rat/Wistar/10 F
Exposure
Duration
52 d
ad lib in feed
8, 12, or 20 d
Ix/d
Gd 12-20,
0-12, or 0-20
(gavage)
during gesta-
tion, during
lactation, or
postnatal
days 21-30 in
drinking
water
47 d prior to
and during
gestation
ad lib in
drinking
water
Once
Gd7
(gavage)
8d
Ix/d
Gd 7-14
(gavage)
4d
Gd6-9
(gavage)
4d
Ix/d
Gd6-9
(gavage)
Dose
(mg/kg-day)
0, 0.002, 0.01,
0.05, 0.25
(MMC)
2,4
(MMC)
0,2.5
(MMC)
0.42,0.7, 1.4
(MMH)
0,4
(MMC)
0, 2, 4, 6
(MMC)
0,0.04, 1.6
(MMC)
0, 0.004, 0.008,
0.035
(MMC)
Effects/Limitations/BML
Increased incidence of eye defects (in harderian and
lachrymal glands) and salivary glands in fetuses (LOAEL
= 0.25); significant dose response (p = 0.01). Mothers
were treated from immaturity through weaning or later.
Limitations: Incomplete reporting; of results
BML not reported
Increased brain lesions and generalized edema
(Gd 0-20) (LOAEL = 2).
Limitations: Limited data reporting; no statistical
analysis; small number of treated animals
BML not reported
Decreased visual evoked potential latencies for peaks Nl
(p<0.05), PI (p<0.01) and P2 (p<0.01) in 30-day old
pups exposed during gestation, during lactation, or
during postnatal days 21-30.
BML not reported
Ultrastructural changes, dose-related decrease in
biochemical activity in mitochondria of fetal hepatocytes
(p<0.01) following administration of methylmercury
hydroxide to mothers (LOAEL = 1 .4).
BML: 40 ug/g (organic and inorganic) in liver of fetuses
at 1 .4 mg/kg-day
Increased P1-N1 amplitudes and decreased P2 and N2
latencies of cortically visual evoked potential (p<0.05).
BML not reported
At 6 mg/kg-day, decreased maternal weight gain,
increased resorptions and fetal deaths (p<0.001);
decreased fetal body weight increased skeletal and
visceral malformations (hydrocephaly, wavy ribs).
(LOAEL = 4;p<0.01)
BML not reported
Impaired ability to perform operant conditioning
procedures (number of responses on lever required in
specified period of time) (LOAEL = 0.05).
Limitation: Statistical analyses not reported
BML not reported
Reduction in behavioral performance in offspring of
treated mice following operant conditioning
(LOAEL = 0.008; p<0.01).
BML not reported
Reference
Khera and Tabacova
1973
Inouye and
Murakami 1975
Zenick 1976
Fowler and Woods
1977
Dyeretal. 1978
Fuyutaetal. 1978
Muschetal. 1978
Bornhausen et al.
1980
3-89
-------
Table 3-68 (continued)
Developmental Toxicity of Methylmercury in Animals
Species/
Strain/
No. per Sex
per Group
Rat/Sprague-
Dawley/No. F NS
Rat/Sprague-
Dawley/No. F NS
Rat/Sprague-
Dawley/
15-19 F
Rat/Wistar/
38 M, 38 F
Rat/HAN-
Wistar/10 F
Rat/Wistar/No. F
NS
Rat/Wistar/16 F
Exposure
Duration
Once
Gd8
(gavage)
10 d
Ix/d
Gd6-15
(gavage)
4d
Ix/d
Gd6-9
(gavage)
during
gestation and
lactation ad
lib in
drinking
water
13 days prior
to mating
until post-
natal day 21
in drinking
water
4d
Ix/d
Gd6-9
(gavage)
2 wk prior to
mating
through
weaning
ad lib in
drinking
water
Dose
(mg/kg-day)
0,6.3
(MMC)
0, 0.2, 1, 2, 4
(MMC)
0, 1.6,4.8
(MMC)
0, 0.2, 0.6
(MMC)
0,0.2,0.6, 1.7
(MMC)
0, 0.02, 0.04,
0.4,4
(MMC)
0, 0.08-0.38,
0.34-0.95
(MMC)
Effects/Limitations/BML
Shorter avoidance latency in 60-day old offspring
(LOAEL = 6.3).
BML not reported
Delayed sexual development (vaginal patency and testes
descent), reduced pivoting, delayed surface righting,
partially retarded swimming development, increased
activity in center of open field, impaired startle reflex
response. Reduced maternal weight gain and litter
weight. No live offspring were produced at 4 mg/kg-day
(LOAEL = 2; p<0.05).
BML not reported
Delayed vaginal patency, delayed surface righting,
retarded swimming development, lower activity,
impaired complex water maze performance. Increased
mortality of pups at 1-21 days of age (LOAEL = 4.8;
p<0.05).
BML not reported
Increase in response latency in male (p<0.05) and female
pups (p<0.01) and in passiveness (p<0.05) in visual
discrimination reversal task at 0.6 mg/kg-day (LOAEL =
0.6).
BML not reported
Reduced weight gain, ataxia and inability to give birth in
dams at 1 .7. High mortality in pups at 1 .7. Impaired
swimming behavior and righting reflex, delayed sexual
maturity (vaginal opening and testes descent) at 0.2 and
0.6. (LOAEL = 0.2; p<0.05).
BML = 9,700-191,000 ug/L in dams and 10,000-
127,000 ug/L in pups at birth
Increased startle response; impaired swimming behavior,
decreased locomotor and nose-poking behavior;
alteration of dendritic spine morphology (LOAEL = 4).
Limitations: Limited data reporting; no statistical
analysis
BML not reported
Impaired tactile-kinesthetic function (p<0.05) (LOAEL =
0.08-0.38).
BML not reported
Reference
Cuomo etal. 1984
Geyeretal. 1985
Vorhees 1985
Schreineretal. 1986
Suter and Schon
1986
Stoltenburg-
Didinger and
Markwort 1990
Eisner 1991
3-90
-------
Table 3-68 (continued)
Developmental Toxicity of Methylmercury in Animals
Species/
Strain/
No. per Sex
per Group
Rat/Sprague-
Dawley/No. and
sexNS
Mouse/SvSl/
No. F NS
Mouse/CFW/No.
FNS
Mouse/
129/Svsl/
No. F NS
Mouse/C57BL/
10 F
Mouse/
DUB/ICR/8 F
exposed, 7 F
controls
Mouse/
C3H/HeN/10 F
Exposure
Duration
Once
Gdl5
(gavage)
Once
Gd 7 or 9
(i.p.)
Once
Gd8
(i.v.)
Once
GdlO
(s.c.)
8d
Ix/d
Gd6-13
(gavage)
Once
Gdl2
(gavage)
Once
Gd 13, 14,
15, 16, or 17
(gavage)
Dose
(mg/kg-day)
0,6.4
(MMC)
0,
0.16mg
MMD/20 g
0, 1, 2, 3, 5, 10
(MMH)
0, 5, 7, 10
0, 2, 4, 4.8, 6
(MMC)
0,8
0, 16
(MMC)
Effects/Limitations/BML
Increased GABAA receptors in prenatally exposed pups
sacrificed at 14 or 21 days postpartum; increased
behavioral depression after diazepam.
Limitations: Only one treatment level; no data on
number of animals
BML not reported
Impaired swimming ability and open-field behavior
(p<0.05) in 30-day old pups. Dose administered as
methylmercury dicyandiamide (MMD)
BML not reported
Increased number of trials to criterion (p<0.05) and
increased number that failed to attain criterion in 2-way
avoidance test conducted on 56-day old pups (LOAEL =
3).
BML not reported
Longer center square latency at 10 (once) and 3.5 (3 d),
decreased rearings and increased backings at 3.5;
decreased locomotor activity at 7 and 10; postnatal
growth retardation at 7 and 10 (LOAEL = 7; p<0.05).
BML not reported
Increased resorptions and fetal deaths at 4.8 and 6
(p<0.01); increased malformations (cleft palate, fused
vertebrae) at 2 and higher (p<0.05); increased skeletal
variations; decreased maternal weight gain at 4.8 mg/kg-
day (LOAEL = 2).
Limitation: small number of treated animals
BML not reported
Arrest of brain cells during mitosis (p<0.01).
Limitations: Only one dose tested; small number of
animals tested
BML not reported
Decreased neonatal survival and weight gain; impaired
righting response; decreased locomotor activity;
abnormal gait; crossing of hindlimbs; decreased brain
weight in groups treated on Gd 13 or 14 (p<0.01); dilated
lateral ventricles; slightly simplified cerebellar pattern.
Effects were seen in groups dosed on all days, but
somewhat stronger in those treated on Gd 13 or 14.
Limitations: Incomplete reporting of data; most
parameters were not analyzed statistically; only one dose
tested
BML: -20 ug/g in brain of fetuses
Reference
Guidetti et al. 1992
Spykeretal. 1972
Hughes and Annau
1976
Su and Okita 1976
Fuyuta et al. 1978
Rodieretal. 1984
Inouye et al. 1985
3-91
-------
Table 3-68 (continued)
Developmental Toxicity of Methylmercury in Animals
Species/
Strain/
No. per Sex
per Group
Guinea pig/
Hartley/5-9 F
Hamster/
Golden/10 F
Monkey/Macaca
fascicularisi^ F
exposed, 8 F
control
Monkey/Macaca
fascicularisl
12 F exposed, 13
F control
Monkey/Macaca
fascicularisi '5
mothers
Monkey/
Macaca
fascicularisi '4 M,
1 F exposed, 1
M, 2 F controls
Monkey/Macaca
fasicularis/13
total
Exposure
Duration
Once
Gd 21, 28,
35, 42, or 49
(gavage)
Once at
Gd 10, or
6d
Ix/d
Gd 10-15
(gavage)
approx. 1-3
yr 1 x/d prior
to mating
through
gestation (in
apple juice)
approx. 4 mo
to 2 yr 1 x/d
prior to
mating
through
gestation (in
apple juice)
4-4.5 yr
1 x/d in utero
and
postnatally
(gavage)
6.5-7 yr
7d/wk
Ix/d
(capsule;
gavage)
4-4.5 yr Ixd
in utero and
postnatally
(gavage)
Dose
(mg/kg-day)
9.4-15
7.5 mg/animal
(wt 500-800 g)
(MMC)
0, 1.6, 8
(MMC)
0, 0.04, 0.06
0, 0.04
0, 0.01, 0.025,
0.5
(MMC)
0, 0.05
(MMC)
0, 0.01, 0.025,
or 0.05
Effects/Limitations/BML
Aborted litters and retarded fetal brain development at all
treatment times.
Limitations: No statistical analysis; small number of
treated animals, only 1 day of dosing
Avg BML over treatment time: Fetal: 2,600 u/L in
blood; Maternal: 1,800 ug/g in blood
Degeneration of cerebellar neurons in rats born to
mothers treated with 1.6 mg/kg/d on Gd 10-15 or a
single dose of 8 mg/kg on Gd 10 and sacrificed
neonatally or as adults.
Limitation: small number of treated animals
BML not reported
Impaired visual recognition memory (data pooled from
both groups of infants of exposed mothers) compared to
unexposed controls; test performed at 50-60 days of age.
Limitation: small number of treatment animals
BML Range: 880-2,450 ug/L in blood of infants at birth;
280-830 ug/L at testing
Decrease in social play behavior and concomitant
increase in nonsocial passive behavior compared to
unexposed controls; tests performed at 2 weeks to
8 months of age.
Limitation: small number of treatment animals
BML Range: 1,565 ug/L in blood of infants at birth
Spatial visual impairment (LOAEL = 0.01).
Limitation: Small number of infants (5 high-dose; 2
mid-dose; 1 low-dose)
BML not reported
Six years after end of dosing (follow-up study to Rice
and Gilbert 1982); decreased fine motor performance;
diminished touch and pinprick sensitivity; impaired high
frequency hearing (p<0.05).
Limitations: small number of animals tested; one dose
level tested
BML: Not detectable at time of dosing
Monkeys tested as juveniles showed no gross intellectual
impairment; some indication of decreased temporal
discrimination. BML in treated animals at birth
averaged 0.46, 0.93, or 2.66 ppm; decreased to steady-
state of 0.20, 0.25 or 0.60 ppm.
Reference
Inouye and Kajiwara
1988b
Reuhletal. 1981
Gunderson et al.
1988
Gunderson etal.
1988
Rice and Gilbert
1990
Rice 1989b; Rice
and Gilbert 1992
Rice 1992.
3-92
-------
Table 3-68 (continued)
Developmental Toxicity of Methylmercury in Animals
Species/
Strain/
No. per Sex
per Group
Monkey/Macaca
fascicularis/23 F
Monkey/Sazmzri
sciureus/3 F
Exposure
Duration
unspecified
period prior
to mating
through
gestation
week 1 1 or
14.5 until
parturition
(gavage)
Dose
(mg/kg-day)
0.04, 0.06, 0.08
0.7to0.9ppm
methylmercury
in maternal
blood
Effects/Limitations/BML
No effect on spatial memory of adult offspring of
animals treated with methylmercury hydroxide (data
pooled from 24 animals, all treated groups).
BML Range: 1,040-2,460 ug/L in blood of infants at
birth
Monkeys exposed in utero tested (on learned lever
pulling activity) at ages 5-6 yr. Methylmercury
treatment resulted in decreased sensitivity to degrees in
reinforcement; change in reinforcement degree resulted
in either no behavior change or slow change by
comparison to controls. Limitations: small number of
animals tested; incomplete reporting on treatment.
Reference
Gilbert et al. 1993
Newlandetal. 1994
3.3.3.9 Reproductive
Although no data were located regarding the reproductive effects of oral exposure to
methylmercury in humans, animal data suggest that, at sufficiently high doses, methylmercury may
adversely affect reproductive function in both males and females. When male rats were given
methylmercury for several days prior to mating, mated females were observed with increased
preimplantation losses (Khera 1973). Exposure of male monkeys to methylmercury for longer durations
has been shown to adversely affect sperm motility and speed and to result in increased incidences of
sperm tail defects (Mohamed et al. 1987). Decreases in spermatogenesis and tubular atrophy of the testes
have been observed upon histopathological analyses of the testes of mice exposed to methylmercury
chronically (Hirano et al. 1986; Mitsumori et al. 1990).
Less information is available regarding the effects of methylmercury on female reproductive
function. Exposure of female monkeys to methylmercury for 4 months prior to mating produced no
effects on the length of the menstrual cycle but resulted in decreased conceptions and increased early
abortions and stillbirths (Burbacher et al. 1988). Several studies have shown increased rates of
resorptions and abortions after exposure during gestation (Fuyuta et al. 1978; Hughes and Annau 1976;
Inouye and Kajiwara 1988a); however, it is unclear from these studies whether the effects observed are
the result of maternal reproductive failure or fetal toxicity.
3-93
-------
Table 3-69
Reproductive Toxicity of Methylmercury in Animals
Species/
Strain/
No. per Sex per Group
Rat/Wistar/10-20 M
Rat/Wistar/14-19 M
Mouse/Swiss Webster/10-20
M
Mouse/ICR/60 M, 60 F
MOUS6/B6C3F/60 M, 60 F
Monkey/Macaca
fascicularis/3 M
Monkey/Macaca
fascicularisll-9 F
Exposure
Duration
7d
Ix/d
(gavage)
95-125 d
Ix/d
5-7 d
Ix/d
(gavage)
104 wk
ad lib in feed
104 wk
ad lib in feed
20 wk
7d/wk
Ix/d
(gavage)
4 mo prior to
mating
Ix/d
(gavage)
Dose
(mg/kg-day)
0, 1,2.5,5
(MMC)
0.1,0.5, 1
(MMC)
0, 1,2.5,5
(MMC)
0,0.03,0.15,
0.72 (M); 0.02,
0. 11, 0.62 (F)
(MMC)
0, 0.03, 0.14,
0.68 (M); 0.03,
0.13,
0.6 (F)
(MMC)
0, 0.047, 0.065
0, 0.04, 0.06,
0.08
(MMH)
Effects/Limitations/BML
Reduced mean litter size after male exposure
(LOAEL = 5; p<0.01) in sequential mating trials
with unexposed females
BML not reported
Males were mated to unexposed females concurrent
with dosing. Reduced mean litter size (LOAEL =
0.5)
BML not reported
No effect on number of viable embryos, dead
embryos, or percent pregnancy (NOAEL = 5)
BML not reported
Significantly decreased spermatogenesis (LOAEL =
0.73; significance level not reported)
BML not reported
Tubular atrophy of the testes (LOAEL = 0.69;
p<0.01)
BML not reported
Decreased sperm motility and speed; increased
sperm tail defects (LOAEL = 0.065; p<0.05)
BML: -2200 ug/L in blood at 0.065 mg/kg-day,
approaching steady state
Abortion; stillbirth; decreased conception in exposed
females (LOAEL = 0.06); no effect on menstrual
cyclicity
Avg. BML: 1,600 ug/L in blood at equilibrium at
0.06 mg/kg
Reference
Khera 1973
Khera 1973
Khera 1973
Hirano et al.
1986
Mitsumori et
al. 1990
Mohamed et
al. 1987
Burbacher et
al. 1988
3.3.3.10 Genotoxicity
Data from several studies in humans suggest that ingesting methylmercury may cause
chromosomal aberrations and sister chromatid exchange (Skerfving et al. 1970, 1974; Wulf et al. 1986;
Franchietal. 1994).
A study of nine Swedish subjects who consumed mercury-contaminated fish and 4 controls
showed a statistically significant rank correlation between blood mercury and percentage of lymphocytes
with chromosome breaks (Skerfving et al. 1970). An extension of this study (Skerfving et al. 1974)
included 23 "exposed" (5 females and 18 males) and 16 "controls" (3 females and 13 males). The authors
3-94
-------
reported a significant correlation between blood mercury level and frequency of chromatid changes and
"unstable" chromosome aberrations; there was no correlation with "stable" chromosome aberrations.
The Wulf et al. (1988) study was of 92 Greenlander Eskimos. Subjects were divided into three
groups based on intake of seal meat (6 times/week; 2-5 times/week; once/week or no consumption of seal
meat). Higher frequency of SCE in lymphocytes was correlated with blood mercury concentration; an
increase of 10 (ig Hg/L in blood was associated with an increase of 0.3 SCE/cell. Positive correlations
were also found for smoking, diet, living district and cadmium exposure.
Franchi et al. (1994) evaluated formation of micronuclei in peripheral blood lymphocytes of
Mediterranean fishers, a group with presumed high exposure to methylmercury. Fifty-one subjects were
interviewed on age, number of seafood-based meals/week and habits such as smoking and alcohol
consumption. Total blood mercury was measured; the range was 10.08 - 304.11 ng/g with a mean of
88.97 + 54.09 ng/g. There was a statistically significant correlation between blood mercury concentration
and micronucleus frequency and between age and micronucleus frequency.
Table 3-70
Genotoxicity of Methylmercury in Humans: Case Study
Species/
No. per Sex
per Group
Human/6 M, 3
F exposed; 3
M, 1 F control
Exposure
Duration
>5yr
>3x/wk
Dose
(mg/kg-day)
NS
Effects/Limitations/BML
Correlation between blood mercury concentration and
chromosome breaks in lymphocytes cultured from people
who ate mercury-contaminated fish
Limitation: Small sample size; limited exposure data
BML Range: 4-650 ug/L in blood
Reference
Skerfvingetal. 1970
Table 3-71
Genotoxicity of Methylmercury in Humans: Epidemiology Study
Species/
No. per Sex
Human/24-63
(both sexes)
Exposure
Duration
NS
Dose
(mg/kg-day)
NS
Effects/Limitations/BML
Incidence of sister chromatid exchanges (SCEs) in
cultured peripheral lymphocytes correlated with intake of
seal meat in an Eskimo population (as a surrogate for
mercury intake); p = 0.001. Other factors also correlated
with SCEs, but multiple regression analysis found that
some of the effect was attributable to mercury.
Limitation: Limited exposure data
BML not reported
Reference
Wulfetal. 1986
3-95
-------
Table 3-71 (continued)
Genotoxicity of Methylmercury in Humans: Epidemiology Study
Species/
No. per Sex
Human / 5 1 M
Human/1 8M
exposed/10
control
Exposure
Duration
measured as
seafood
meals/ week.
Range 2 - 14.
10.5 yr
(occup)
Dose
(mg/kg-day)
NS
0.15-0.44
(HgCl2)
Effects/Limitations/BML
Incidence of micronuclei positively correlated with blood
mercury concentration and with age. No correlation with
smoking or number of seafood meals /week. Limitation:
no control group.
BML range: 10.08 - 403.11 ,ug/g blood.
Increased frequency of chromosomal breaks.
Limitations: Workers also exposed to mercuric chloride
and one worker had history of benzene poisoning; control
group was not matched for sex, smoking habits, or sample
size.
BML: = 890 ,ug/L in urine (avg)
Reference
Franchietal. 1994.
Popescuetal. 1979
In a study with cats (Charbonneau et al. 1976), methylmercury did not induce dose-related
unscheduled DNA synthesis in lymphocytes or chromosomal aberrations in bone marrow cells after oral
exposure to methylmercury for up to 39 months (Miller et al. 1979). Statistically significant decreases in
unscheduled DNA synthesis and increases in chromosomal aberrations were observed, but there was no
dose-response.
Table 3-72
Genotoxicity of Methylmercury in Cats
Species/
Strain/
No. per Sex
per Group
Cat/Breed and
sexNS
Exposure
Duration
39 mo
7d/wk
Dose
(mg/kg-day)
0.008, 0.020,
0.046
Effects/Limitations/BML
No dose-related changes in unscheduled DNA synthesis in
cultured lymphocytes or frequency of chromosomal
aberrations in bone marrow of cats fed mercury-
contaminated fish or a fish diet supplemented with
methylmercuric chloride
Limitations: No positive control; no assessment of
cytotoxicity
BML Range: 500-13,500 ug/L Hg in blood
Reference
Miller etal. 1979
Strain-specific differences exist with respect to the ability of methylmercury to produce dominant
lethal effects in mice (Suter 1975). When (SEC x CsyB^Fj males were injected with 10 mg/kg
methylmercury hydroxide, there was a slight reduction in the total number of implantations and a
decrease in the number of viable embryos. This was not observed when (101 x C^Fj males were
3-96
-------
exposed in a similar fashion. When female (10 x C3H)Fj mice were treated with methylmercuric
hydroxide, no increase in the incidence of dead implants was observed (unlike the case for mercuric
chloride). Changes in chromosome number but no increase in chromosome aberrations were observed in
oocytes of Syrian hamsters treated with one i.p injection of 10 mg/kg methylmercuric chloride (Mailhes
1983). Methylmercury was administered s.c. to golden hamsters at doses of 6.4 mg or 12.8 mg
Hg/kg/body weight. Polyploidy and chromosomal aberrations were increased in bone marrow cells, but
there was no effect on metaphase II oocytes. There was an inhibitory effect on ovulation which the
authors noted was not as severe as that induced by mercuric chloride in the same study (Watanabe et al.
1982). Non-dysjunction and sex-linked recessive lethal mutations were seen in Drosophila melanogaster
treated with methylmercury in the diet (Ramel 1972).
As reviewed in WHO (1990), methylmercury is not a potent mutagen but is capable of causing
chromosome damage in a variety of systems. In vitro studies have generally shown clastogenic activity
but only weak mutagenic activity. Methylmercuric chloride and dimethylmercury were both shown to
induce chromosome aberrations and aneuploidy in primary cultures of human lymphocytes;
methylmercuric chloride was the more potent clastogen at equally toxic doses (Betti et al. 1992). Both
methylmercury and mercuric chloride induced a dose dependent increase in SCE in primary human
lymphocytes and muntjac fibroblasts; methylmercury was about five time more effective in this regard
(Verschaeve et al. 1984; Morimoto et al. 1982).
Methylmercury has been shown to inhibit nucleolus organizing activity in human lymphocytes
(Verschaeve et al. 1983). Methylmercury can induce histone protein perturbations and has been reported
to interfere with gene expression in cultures of glioma cells (WHO 1990). Impaired growth and
development was noted in cultured mouse embryonic tissue treated in vitro with methylmercuric
chloride, but there was no increase in SCE (Matsumoto and Spindle 1982). Costa et al. (1991) showed
that methylmercuric chloride caused DNA strand breaks in both V79 and rat glioblastoma cells treated in
vitro. Methylmercuric chloride produced more strand breaks than did mercuric chloride.
Evidence of DNA damage has been observed in the Bacillus subtilis rec-assay (Kanematsu et al.
1980). These authors reported negative results for methylmercury in spot tests for mutagenicity in the
following bacterial strains: E. coll B/r WP2 and WP2; and Salmonella typhimurium strains TA1535,
TA1537, TA1538, TA98 and TA100. Jenssen and Ramel (1980) in a review article indicated that
methylmercury acetate was negative in both micronucleus assays and in mutagenicity tests in Salmonella;
the article referred to Heddle, J.R. and W.R. Bruce (1977) and provided no experimental details. Weak
mutagenic responses for methylmercuric chloride and methoxyethyl mercury chloride were observed in
Chinese hamster V79 cells at doses near the cytotoxic threshold (Fiskesjo 1979), and methylmercury
produced a slight increase in the frequency of chromosomal nondisjunction in Saccharomyces cerevisiae
(Nakai and Machida 1973). Methylmercury, however, caused neither gene mutations nor recombination
in S. cerevisiae (Nakai and Machida 1973). Methylmercury retarded DNA synthesis and produced single
strand breaks in DNA in L5178Y cells (Nakazawa et al. 1975).
3-97
-------
4. SUSCEPTIBLE POPULATIONS
A susceptible population is a group who may experience more severe adverse effects at
comparable levels or adverse effects at lower exposure levels than the general population. The greater
response of these sensitive subpopulations may be a result of a variety of intrinsic or extrinsic factors.
Volume V describes populations that may be at increase risk because of higher exposure to mercury and
mercury compounds. Additional factors that may be important include, but are not limited to, the
following: an impaired ability of the detoxification, excretory, or compensatory processes in the body to
protect against or reduce toxicity; differences in physiological protective mechanisms (e.g., blood brain
barrier); or unique toxic reactions that are specific to the genetic makeup, developmental stage, health
status, gender or age of the individual.
The nervous and renal systems are the primary targets for mercury-induced toxicity. Data are
also available indicating some effects to the respiratory, cardiovascular, gastrointestinal, hematologic,
immune, and reproductive systems. The developing organism appears to be particularly sensitive to
methylmercury exposure. In addition, it is probable that individuals with preexisting damage or disease
in target organs for mercury-induced toxicity may experience more severe effects upon exposure to
mercury. The populations listed below may be highly susceptible to mercury toxicity.
• Developing Organisms. Data from epidemic poisonings in Japan (Harada 1978) and Iraq
(Marsh et al. 1987) indicate that infants exposed in utero to methylmercury developed
marked neurological development delays while their mothers experienced little or no
overt signs of toxicity. Data indicate that the developing fetus may be 5 to 10 times
more sensitive than the adult (Clarkson, 1992). This difference in sensitivity is believed
to be due, in part, to the high sensitivity of developmental processes (i.e., cellular
division, differentiation, and migration) to disruption by mercury (Choi et al. 1978;
Sager et al. 1982). One factor that may account for this difference in sensitivity is the
presence of an incomplete blood brain barrier in the fetus. Another important factor may
be the lack of methylmercury excretion in the fetus (Grandjean et al. 1994).
• Age - Infants and Other Age Groups. Available data indicate that neonates are at
increased risk to inorganic mercury and methylmercury. Both inorganic and organic
forms of mercury are excreted in breast milk (Sundberg and Oskarsson 1992; Yoshida et
al. 1992; Grandjean et al. 1994); thus, neonates in an exposed population may experience
increased mercury exposure. Animal data for rats indicate that suckling infants retain a
higher percentage of ingested inorganic mercury than do adults (Kostial et al. 1978).
The most significant difference in organ retention (neonates > adults) was
methylmercury in the brain following exposure to methylmercury (Yang et al. 1973;
Kostial et al. 1978) and inorganic mercury retained in the kidney following exposure to
elemental mercury (Yoshida et al. 1992). These differences may be associated with an
increased absorption of mercury with a milk diet, a decrease in excretion, or an
incomplete blood brain barrier (Kostial et al. 1978, Grandjean et al. 1994).
Signs of toxicity may begin to be manifested several years after the cessation of dosing,
possibly related to subclinical effects being unmasked by aging. Rice (1989b) dosed
monkeys with methylmercury from birth to 6.5-7 years of age. Although there were no
overt signs of neurotoxicity during dosing, neurological deficits were observed at 13
4-1
-------
years of age, 6-7 years following cessation of exposure. Similarly, a small human
population with Minamata disease has been identified in Japan as experiencing new or
worsening neurological effects a few years following termination of mercury exposure.
This late-onset Minamata disease may be related to several factors including aging (Igata
1993).
Gender. Sex-related differences in mercury toxicokinetics and sensitivity to mercury
have been observed, although data indicate that the more sensitive sex may differ by
species and strain. Using death as the critical endpoint, in one strain of mice,
C57BL/6N, males were less sensitive to methylmercury following daily dosing than
females while, in contrast, male mice were more sensitive than females in another strain,
BALB/cA (Yasutake and Hirayama 1988). In humans, although the ratio of males to
females with Minamata disease has been reported to be 1.2:1, the ratio of deaths was
recorded at 1.8:1 (Tamashiro et al. 1984).
Other studies are in general agreement that male rats (Thomas et al. 1986) and mice
(Nielsen and Andersen 1991a, 1991b) eliminate mercury faster and have lower tissue
levels than females following dosing with methylmercury. Part of the difference in
whole-body retention of mercury in methylmercury-exposed mice has been associated
with varying degrees of deposition of mercury in the carcass, including the skin and hair
(Nielsen and Andersen 1991b). This difference is thought to be due in part to
differences in glutathione metabolism and renal excretion of mercury, which is affected
by the hormonal status of testosterone (Nielsen et al. 1994). Hirayama et al. (1987)
have reported that the toxicokinetics of methylmercury in castrated male mice was very
similar to that in female mice, and that the male pattern of methylmercury toxicokinetics
could be restored by testosterone treatment. Such differences were not observed in a
small set of similarly tested human volunteers (Miettinen et al. 1971).
Dietary Insufficiencies of Zinc. Glutathione. or Antioxidants. Mercury has been
suggested to cause tissue damage by increasing the formation of reactive oxygen species
and activation of lipoperoxidation, calcium-dependent proteolysis, endonuclease activity,
and phospholipid hydrolysis (Ali et al. 1992; LeBel et al. 1990, 1992; Gstraunthaler et
al. 1983; Verity and Sarafian 1991). Zinc, glutathione, and antioxidant deficiencies
would be expected to exacerbate mercury-induced damage by limiting cellular defenses
against the oxidative processes. Animal data support the importance of zinc,
glutathione, and antioxidants in limiting mercury-induced damage (Fukino et al. 1992;
Girardi and Elias 1991; Yamini and Sleight 1984) (see also Section 5, Interactions).
Predisposition for Autoimmune Glomerulonephritis. Autoimmune glomerulonephritis is
a form of renal toxicity characterized by proteinuria, deposition of immune material (i.e.,
autoantibodies and complement C3) in the renal mesangium and glomerular blood
vessels and glomerular cell hyperplasia (Bigazzi 1992; Goldman et al. 1991; Mathieson
1992). Limited human data suggest that certain individuals may develop this
autoimmune response when exposed to inorganic or elemental mercury (Cardenas et al.
1993; Langworth et al. 1992b; Tubbs et al. 1982). While the etiology of this syndrome
has not been completely elucidated, data from susceptible and resistant strains of animals
indicate that susceptibility is governed by both major histocompatibility complex (MHC)
4-2
-------
genes and non-MHC genes (Aten et al. 1991; Druet et al. 1978; Hultman and Enestrom
1992; Hultman et al. 1992; Michaelson et al. 1985; Sapin et al. 1984).
Predisposition for Acrodynia. Acrodynia, also known as "pink disease," is a
hypersensitive response following exposure to elemental or inorganic mercury and is
characterized by the following signs and symptoms: irritability; marked mood swings;
restlessness; itching; flushing, swelling, and/or desquamation of the palms of the hands
and soles of the feet (the tip of the nose, ears, and cheeks may also be affected);
excessive perspiration; loss of appetite; tachycardia; hypertension; joint pains and
muscle weakness; photophobia; and sleeplessness. Acrodynia, which is more likely
related to exposure level rather than any inherent, genetic sensitivity, rarely occurs in the
general population.
Limited reports indicate that acrodynia has been almost exclusively observed in children,
affecting approximately 1 in 500 exposed children (Blondell and Knott 1993; Warkany
and Hubbard 1953). This disease was recently observed in a 4-year-old Michigan boy
who was exposed to mercury vapor released from paint in which mercury had been used
as a fungicide (Aronow et al. 1990). In this case, family members (i.e., both parents and
two siblings) were also exposed to the mercury vapors but remained asymptomatic
(Aronow et al. 1990). This case study supports the hypothesis that there is no genetic
predisposition to acrodynia.
Acrodynia was more frequently observed in the past when mercury-containing laxatives,
worming medications, teething powders and diaper rinses were widely used (Gotelli et
al. 1985; Warkany and Hubbard 1953). The physiological basis for this hypersensitivity
has not been identified. It does not appear, however, to be an allergic reaction to
mercury or to occur in the most highly exposed individuals (Warkany and Hubbard
1953).
4-3
-------
5. INTERACTIONS
A number of interactions have been identified for chemicals that affect the pharmacokinetics
and/or toxicity of mercury compounds. Table 5-1 summarizes interactions in which potentiation or
protection from the toxic effects of mercury have been observed. Interactions that affect mercury
toxicokinetics are also shown. The effect on toxicity, however, can not be predicted based on changes in
distribution or excretion. For example, zinc pretreatment increases renal mercury levels but decreases
toxicity because it alters the distribution within the kidney (Zalups and Cherian 1992).
Only the interaction of selenium with mercury will be discussed in detail here. Selenium is
known to bioaccumulate in fish, so exposure to methylmercury in fish is associated with exposure to
increased levels of selenium. Where the main source of dietary mercury is fish, the diet is naturally
enriched with selenium relative to mercury. Increased selenium has been suspected of providing some
degree of protection, either by preventing oxidative damage or by forming a methylmercury-selenium
complex (Grandjean 1992a). It does not appear that the population in the Iraqi poisoning incident was
selenium deficient. Animal studies have demonstrated that simultaneous ingestion of selenium may be
protective against toxicity of methylmercury based upon its antioxidant properties (see Table 5-1). This
may explain why the latent period in Japan, where the population was exposed to methylmercury in fish,
was longer than that in Iraqi, where the exposure was to methylmercury in grain.
A common association between the metabolism of selenium and methylmercury is the thiol-
containing peptide glutathione (GSH). The metabolic cycling and oxidation-reduction of GSH are
integral processes coupled to the activation and metabolism of selenium (Hill and Burke 1982) and the
metabolism and detoxification of methylmercury (Ballatori and Clarkson 1982; Thomas and Smith
1982).
There are data to indicate that selenium co-administered with methylmercury can form selenium-
methylmercury complexes (Magos et al 1987). The formation of these complexes appeared temporarily
to prevent methylmercury-induced tissue damage but also apparently delayed excretion of the
methylmercury in the urine. Thus, formation of selenium-methylmercury complexes may not reduce
methylmercury toxicity but may rather delay the onset of symptoms.
In support of the protective role of biological selenium, several investigators have found that a
diet supplemented with seafood high in selenium delayed the onset of methylmercury intoxication in rats
(Ganther 1980; Ohi et al. 1976). Ganther (1980) has observed that rats given selenium plus
methylmercury show increased body burdens of both selenium and methylmercury without signs of
toxicity. The accumulation of these elements may lead to mutual detoxification, but such
coaccumulation is not always linked to protection. Fair and associates (1985) have examined renal
ultrastructure changes along with changes in gamma glutamyl transferase activity in mice coadministered
both selenium and methylmercury in diet for 7 or 20 days or given a single i.p. dose. The results of this
study indicated that dietary selenium had only an initial protective effect against mercury accumulation
in the kidney; injected selenium offered longer protection.
Selenium has been shown to protect against the developmental toxicity of methylmercury in mice
(Nishikido et al. 1987; Satoh et al. 1985) and protects against oxidative damage by free radicals (Cuvin-
Aralar and Furness 1991; DiSimplicio et al. 1993; Ganther 1978). Further studies reported by
Fredricksson et al. (1993) indicated that dietary selenium supplementation during pregestation through
lactation in rats resulted in reduction of some adverse effects (hypoactivity) in neonates of the
5-1
-------
methylmercury administered to mothers by gavage during organogenesis period. Significant increases in
glutathione peroxidase activity were noticed in animals fed selenium supplemented diet.
Kosta et al. (1975) have observed a coaccumulation of mercury and selenium in the organs and
tissues of mineworkers at an approximate molar ratio of 1:1. In these circumstances, the abnormally high
mercury levels detected in the tissues were without apparent adverse effects on the miners. After
exposure to mercury in the mines, several of the miners had been retired 10-16 years when the study was
conducted. The selenium intake from the diet was not reported but was said not to be abnormally high,
suggesting that the co-accumulation with mercury is a natural and autoprotective effect. It is plausible
that in areas naturally low in selenium, individuals would be at greater risk from methylmercury
poisoning than those in areas of high selenium concentration.
A group of 21 workers with no previous history of mercury exposure were monitored for urinary
mercury and selenium after their employment in the demolition of a chlor-alkali plant. Pre-exposure
urinary mercury ranged from 0.3 - 1.9 nmol/ mmol creatinine (mean = 0.8); urinary selenium was 13.9 -
89.5 (mean = 39.1). Post-exposure urinary mercury was significantly increased; 1.2 - 10.0 nmol/mmol
creatinine (mean = 4.8). Selenium in the urine was decreased post exposure to 10.1 - 52.9 nmol/ mmol
creatinine (mean = 29.0). The authors did not speculate on the biological significance of the change in
urinary selenium.
Co-administration of methylmercury and selenium apparently results in decreased
methylmercury concentrations in kidney; mercury levels in brain and liver, however, are increased
(Suzuki and Yamamoto, 1984; Brzenickaand Chmielnicka, 1985; Komsta-Szumskae^a/., 1983).
Selenium has also been observed to increase methylmercury staining in spinal cord and nerve cell bodies
(M011er-Madsen and Danscher, 1991). A positive correlation between brain mercury and selenium
levels was observed in monkeys exposed to methylmercury with no additional exposure to selenium other
than that in a standard diet (Bjorkman et al., 1995). The apparent protective effect of selenium against
overt high-dose methylmercury toxicity has been attributed to the decreased accumulation of
methylmercury in kidney in the presence of selenium (Stillings et al., 1974). It is doubtful that this effect
on kidney is relevant at environmental levels of methylmercury. It has also been suggested that the
formation of bis (methylmercury) selinide may render methylmercury less toxic (Naganuma and Imura,
1980), but there is no direct evidence for this. In addition, although increased fish consumption was
associated with a very modest increase in (cord) blood selenium levels in a fish-eating population, the
blood mercury levels increased much more dramatically (Grandjean et al., 1992). Grain may also
contain substantial levels of selenium, depending on the soil in which it is grown. Based on the
questionable relevance of any protective effect of selenium against high-dose methylmercury
nephrotoxicity, the fact that increased selenium intake results in increased brain mercury levels following
methylmercury ingestion, and a complete lack of data on the selenium status of the Iraqi population
exposed to methylmercury via grain, there is no reason to postulate that ingestion of methylmercury in a
fish matrix would result in decreased toxicity compared to ingestion in a non-fish matrix. Indeed, the
study in cats addressed this directly found no difference in toxicity or tissue levels when methylmercury
was administered as contaminated fish or added to a non-fish meal (Charbonneau et al., 1974).
5-2
-------
Table 5-1
Interactions of Mercury with Other Compounds
Compounds
Diethylmaleate and
inorganic mercury
Ethanol and
methylmercury
Ethanol and elemental
mercury
Ethanol and inorganic
mercury
Thiol compounds
[e.g.,W-
acetylpenicillamine,
penicillamine, 2-
mercapto propanol
(BAL)] and inorganic
mercury
Selenium and mercury
(simultaneous exposure)
Tellurium and elemental
or inorganic mercury
Effects Observed
Increased renal toxicity
Potentiated toxicity
Increased mortality,
increased severity of
neurotoxicity, renal
toxicity
Decreased time to onset of
neurotoxicity
No data on toxicity
Decreased mercury
absorption
Increased mercury levels
in liver and in fetus
No data on toxicity
Increased mercury
exhalation
Protection from renal
toxicity
Increased survival
Decreased or delayed
renal, developmental
toxicity
Decreased toxicity (effect
unspecified)
Retention in body
increased
Proposed Underlying Mechanism(s)
Diethylmaleate causes depletion of nonprotein
sulfhydryls
Unknown; increased mercury concentrations were
observed in brain and kidneys, but changes in
mercury content were insufficient to fully explain
the potentiated toxicity
Inhibition of oxidation of metallic mercury to
mercuric mercury by catalase
The effect on toxicity can not be predicted, due to
the opposing effects.
Elemental mercury was exhaled, suggesting that
ethanol increased the activity of an unidentified
enzyme that reduces mercuric mercury to
elemental mercury.
Because elemental mercury, but not mercuric
mercury, can cross the blood brain barrier and the
placenta, toxicity to the brain and the developing
fetus may be increased
Competition for protein binding sites; subsequent
increases in urinary excretion of mercury
Mercuric mercury and selenium form a complex
with a high molecular weight protein
Methylmercury forms a bismethylmercury
selenide complex
Potential mechanisms for protection:
-redistribution from sensitive targets
-competition of selenium for mercury
binding sites associated with toxicity
-increased selenium available for selenium-
dependent glutathione peroxidase
(prevention of oxi dative damage)
Complexation of tellurium with mercury, by
analogy to the chemically-related selenium
References
Girardi and Elias
1991
Rumbeihaetal.
1992
Tamashiro etal.
1986
Turner etal. 1981
Nielsen-Kudsk
1965
Magos and Webb
1979
Khayat and
Dencker 1982,
1984b
Dunn etal. 1981b
Magos and Webb
1979
Parizek and
Ostadolva 1967
Satoh et al. 1985
Naganuma and
Imura 1981
Mengel and Karlog
1980
Cuvin-Aralar and
Furness 1991
Imura and
Naganuma 1991
Nylander and
Weiner 1991
Magos and Webb
1979
Khayat and
Dencker 1984a
5-3
-------
Table 5-1 (continued)
Interactions of Mercury with Other Compounds
Compounds
Potassium dichromate
and inorganic mercury
Zinc pre-treatment and
inorganic mercury
Zinc-deficiency and
inorganic mercury
Atrazine and
methylmercury
Vitamin C deficiency
and methylmercury
Vitamin E and
methylmercury
Potassium dichromate
and mercuric chloride
Effects Observed
Decreased renal function
(measured as inhibition of
p-aminohippurate
transport)
Some protection from
nephrotoxicity of
inorganic mercury
Exacerbation of renal
toxicity
Early onset of
neurotoxicity
Increased severity of
neurological damage
Increased survival and
decreased toxicity
Synergistic inhibition of
renal transport
Proposed Underlying Mechanism(s)
Unknown; both chemicals are toxic to the renal
proximal tubule
Zinc pretreatment induces metallothionein binding
in kidneys
Mercury binds preferentially to metallothionein, so
that less mercury is available to cause oxidative
damage in the proximal tubules
Zinc-deficiency and mercury both independently
increase renal oxidative stress
Together, the protective mechanisms of the kidney
are overwhelmed and oxidative damage is
compounded
Atrazine causes depletion of nonprotein
sulfhydryls
Antioxidant properties of Vitamin C and
protection against oxidative damage caused by
mercury
Protection is possibly related to antioxidant
properties of Vitamin E affording protection
against oxidative damage caused by mercury
Mercuric chloride and potassium dichromate are
both toxic to renal proximal tubule
References
Baggett and Berndt
1984
Zalups and
Cherian 1992
Fukinoetal. 1992
Meydani and
Hathcock 1984
Yamini and Sleight
1984
Welsh 1979
Baggett and Berndt
1984
5-4
-------
6. HAZARD IDENTIFICATION AND DOSE-RESPONSE ASSESSMENT
6.1 Background
Risk assessments done by U.S. EPA follow the paradigm established by the National Academy
of Sciences (NRC 1983). This entails a series of interconnected steps including hazard identification,
dose response assessment, exposure assessment and risk characterization. Two processes, hazard
identification and dose response are the focus of this chapter. Volume IV of this Report presents the
assessment of exposure to mercury emissions in the atmosphere, and Volume VII covers the risk
characterization.
Hazard identification poses the following questions: is the agent in question likely to pose a
hazard to human health; and what types of adverse effects could be expected as a consequence of the
exposure to the agent. Dose-response assessment uses available human, experimental animal and in vitro
data to estimate the exposure level or dose which is expected to produce and adverse health effect. In
accomplishing the aims of risk assessment U.S. EPA applies published Guidelines for Risk Assessment.
6.1.1 Hazard Identification
U.S. EPA has published Guidelines for hazard identification in three areas: developmental
effects, germ cell mutagenicity, and carcinogenic effects. Guidelines for assessment of reproductive
effects were finalized while this Report to Congress was in the process of review; these have not been
applied to the Mercury Study. The specific categorizations for each of those endpoints described in
published guidelines are discussed below. For general, systemic noncancer effects, there is no structured
process resulting in a categorization; instead, the hazard identification step is included in the
dose-response assessment process, wherein a critical effect is selected.
6.1.1.1 Developmental Effects
Guidelines for hazard identification in the area of developmental effects were developed by U.S.
EPA in 1986 and subsequently revised (U.S. EPA 1989, 1991). The Guidelines direct that data from all
available relevant studies be considered, whether the studies indicate a potential hazard or not. Preferred
data are from human studies, when available, and animal studies. The revised guidelines do not use an
alphanumeric scheme such as that given in the carcinogenicity guidelines. Instead two broad categories
are used to characterize the health-related data base: Sufficient Evidence and Insufficient Evidence. The
Guidelines define Sufficient Human Evidence as follows:
"...data from epidemiologic studies (e.g., case control and cohort) that provide
convincing evidence for the scientific community to judge that a causal relationship is or
is not supported. A case series in conjunction with strong supporting evidence may also
be used".
Sufficient Experimental Animal Evidence/Limited Human Data is described in the following way:
"The minimum evidence necessary to judge that a potential hazard exists generally
would be data demonstrating an adverse developmental effect in a single, appropriate,
well-conducted study in a single experimental animal species. The minimum evidence
6-1
-------
needed to judge that a potential hazard does not exist would include data from
appropriate, well-conducted laboratory animal studies (at least two) which evaluated a
variety of the potential manifestations of developmental toxicity, and showed no
developmental effects at doses that were minimally toxic to the adult."
6.1.1.2 Germ Cell Mutagenicity
The U.S. EPA (1986) has published Guidelines for classification of potential hazard of
mutagenic effects in human germ cells. Evidence from human and animal in vivo and in vitro systems is
considered in the judgement as to which of eight numerical classes of concern most clearly defines the
data on an environmental agent. In general, the hierarchy of preference for data type is the following:
• Data on germ cells are preferred to data on somatic cells;
• In vivo tests are preferred to in vitro;
• Data from tests in eukaryotes are preferred to data from prokaryotes.
The weight-of-evidence categories are these, presented in order of decreasing strength of evidence for
human germ cell mutagenicity.
1. Positive data derived from human germ cell mutagenicity studies.
2. Valid positive results from studies on heritable mutational events (any kind) in
mammalian germ cells.
3. Valid positive results from mammalian germ cell chromosome aberration studies that do
not include an intergeneration test.
4. Sufficient evidence for a chemical's interaction with mammalian germ cells, together
with valid positive mutagenicity test results from two assays systems, at least one of
which is mammalian. The positive results may be both for gene mutations or both for
chromosome aberrations; if one is for gene mutations and the other for chromosome
aberrations, both must be from mammalian systems.
5. Suggestive evidence for a chemical's interaction with mammalian germ cells, together
with valid positive mutagenicity evidence from two assay systems as described under 4.
6. Positive mutagenicity test results of less strength than defined under 4, combined with
suggestive evidence for a chemical's interaction with mammalian germ cells.
7. Non-mutagenic. Although definitive proof of non-mutagenicity is not possible, a
chemical could be classified operationally as a non-mutagen for human germ cells, if it
gives valid negative results for all endpoints of concern.
8. Not classifiable based on inadequate evidence bearing on either mutagenicity or
chemical interaction with mammalian germ cells.
6-2
-------
These categories are intended as guidance in assessing a level of concern for an agent's
likelihood to be a germ cell mutagen. The three forms of mercury are discussed in Section 6.2.2 in terms
of level of concern rather than an assigned numerical category.
6.1.1.3 Carcinogenic effects
U.S. EPA categorizes the carcinogenic potential of a chemical, based on the overall weight-of-
evidence, according to the following scheme.
Group A: Human Carcinogen. Sufficient evidence exists from epidemiology studies
to support a causal association between exposure to the chemical and human cancer.
Group B: Probable Human Carcinogen. There is sufficient evidence of
carcinogenicity in animals with limited (Group Bl) or inadequate (Group B2) evidence
in humans.
Group C: Possible Human Carcinogen. There is limited evidence of carcinogenicity
in animals in the absence of human data.
Group D: Not Classified as to Human Carcinogenicity. There is inadequate human
and animal evidence of carcinogenicity or no data are available.
Group E: Evidence of Noncarcinogenicitv for Humans. There is no evidence of
carcinogenicity in at least two adequate animal tests in different species or in both
adequate epidemiologic and animal studies.
For specific guidance as to the use of human, animal and supporting data in the above
categorization for cancer, consult the Risk Assessment Guidelines of 1986 (U.S. EPA 1987a).
U.S. EPA has been in the process of revising its Guidelines for cancer risk assessment. The
revised Guidelines will implement the use of narrative categorization. The new guidelines also
encourage greater use of mechanistic data, including information which can be gained from genetic
toxicology. Data which elucidate the mode of action of an agent will also have a direct impact on the
dose response assessment for carcinogenicity. In the past a default procedure for dose response
assessment was most often followed; that of linear low dose extrapolation using an upper bound on the
low dose term of a linearized multistage mathematical model. The revised Guidelines dictate that the
type of low dose extrapolation to be used, if any, be guided by information on the carcinogen's mode of
action. Evidence of genetic toxicity has now become key in making decisions about dose response
assessment.
While the Mercury Study Report to Congress was in preparation, revised Carcinogen Risk
Assessment Guidelines were published in the Federal Register. As the approval process was not final, it
was necessary to apply the existing Guideline's alphanumeric categories; however, an expanded narrative
was done, and the weight of evidence judgement followed closely the revised format for expanded
consideration of mechanistic data.
An application of the proposed revisions to the Carcinogenic Risk Assessment Guidelines was
presented at the 1996 meeting of the Society for Risk Analysis (Schoeny 1996). An abstract of that
presentation is given at the end of section 6.
6-3
-------
6.1.2 Dose-response Assessment
6.1.2.1 Systemic Noncancer Effects
In the quantification of systemic noncarcinogenic effects, an oral reference dose (RfD), an
inhalation reference concentration (RfC) or both may be calculated. The oral RfD and inhalation RfC are
estimates (with uncertainty spanning perhaps an order magnitude) of a daily exposure to the human
population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious
health effects during a lifetime. The RfD and RfC are derived from a no-observed-adverse-effect level
(NOAEL), or lowest-observed-adverse-effect level (LOAEL), identified from a subchronic or chronic
study and divided by an uncertainty factor(s) times a modifying factor. The RfD or RfC is calculated as
follows:
RfD - (NOAEL or LOAEL) . _ mglkg_day
[Uncertainty Factor (s) x Modifying Factor]
(NOAELHFC or LOAELHFC]
RfC = ^ ^ = — mg/m3
[Uncertainty Factor (s) x Modifying Factor]
The methodologies used to derive the RfD or inhalation RfC require the adjustment of NOAEL
and LOAEL values (whether from experimental animal or human studies) to lifetime exposure
conditions (i.e., 24 hours per day for a lifetime of 70 years). Inhalation RfC methods further require
conversion by dosimetric adjustment or the use of a physiologically-based pharmacokinetic model from
NOAELs and LOAELs observed in laboratory animal experiments to human equivalent concentrations
(HEC). Different default adjustments are made based on whether the observed toxicity is in the upper or
lower respiratory tract or at remote sites, and a NOAEL^q or LOAEL g^o i§ derived for use in the
equation above (U.S. EPA 1990).
Selection of the uncertainty factor (UF) to be employed in the calculation of the RfD/RfC is
based upon professional judgment which considers the entire data base of toxicologic effects for the
chemical. In order to ensure that UFs are selected and applied in a consistent manner, the U.S. EPA
(1994) employs a modification to the guidelines proposed by the National Academy of Sciences (NAS
1977, 1980), as shown in the box on the next page.
As noted in the box, the standard UF for extrapolating from animals to humans has been reduced
to three for the derivation of inhalation RfCs. A factor of three was chosen because, assuming the range
of the UF is distributed log normally, the reduction of a standard 10-fold UF by half (i.e. 1005) results in
three. Other UFs can be reduced to three if the situation warrants, based on the scientific judgement of
the U.S. EPA RfD/RfC Work Group (an Agency peer review group). Considerations in the selection of
UFs and/or a modifying factor include, but are not limited to, pharmacokinetics/pharmacodynamics,
concomitant exposures, relevance of the laboratory animal models, species sensitivity, severity of the
effect, potential for recovery, slope and shape of the dose-response curve, exposure uncertainties, quality
of the critical study, and data gaps.
6-4
-------
Uncertainty Factors Used in RfD/RfC Calculations
Standard Uncertainty Factors (UFs)
Use a 10-fold factor when extrapolating from valid experimental results from studies using prolonged
exposure to average healthy humans. This factor is intended to account for the variation in sensitivity
among the members of the human population. [10J
Use an additional 10-fold factor when extrapolating from valid results of long-term studies on
experimental animals when results of studies of human exposure are not available or are inadequate. This
factor is intended to account for the uncertainty in extrapolating animal data to risks for humans. [10 J A
3-fold uncertainty factor is used for extrapolating from inhalation studies on experimental animals to
humans for the derivation of an inhalation RfC. This difference is because dosimetric adjustments reduce
the uncertainty associated with extrapolation between experimental animals and humans.
Use an additional 10-fold factor when extrapolating from less than chronic results on experimental
animals when there are no useful long-term human data. This factor is intended to account for the
uncertainty in extrapolating from less than chronic NOAELs to chronic NOAELs. [10 ^
Use an additional 10-fold factor when deriving an RfD from a LOAEL instead of a NOAEL. This factor
is intended to account for the uncertainty in extrapolating from LOAELs to NOAELs. [10 J
Modifying Factor (MF)
Use professional judgment to determine another uncertainty factor (MF) that is greater than zero and less
than or equal to 10. The magnitude of the MF depends upon the professional assessment of scientific
uncertainties of the study and database not explicitly treated above, e.g., the completeness of the overall
data base and the number of species tested. The default value for the MF is 1.
From the RfD, a Drinking Water Equivalent Level (DWEL) can be calculated. The DWEL
represents a medium-specific (i.e., drinking water) lifetime exposure at which adverse, noncarcinogenic
health effects are not anticipated to occur. The DWEL assumes 100% exposure from drinking water.
The DWEL provides the noncarcinogenic health effects basis for establishing a drinking water standard.
For ingestion data, the DWEL is derived as follows:
DWEL = (RfD) x (Body weight in kg) = _ mg/L
Drinking Water Volume in L/day
where:
Body weight = assumed to be 70 kg for an adult
Drinking water volume = assumed to be 2 L/day for an adult
6.1.2.2 Developmental Effects
For agents considered to have sufficient evidence for developmental toxicity it is appropriate to
consider calculation of a quantitative dose-response estimate. In general, a threshold is assumed for the
dose-response curve for agents producing developmental toxicity. This is "based on the known capacity
6-5
-------
of the developing organism to compensate for or to repair a certain amount of damage at the cellular
level. In addition, because of the multipotency of cells at certain stages of development, multiple insults
at the molecular or cellular level may be required to produce an effect on the whole organism" (U.S. EPA
1991).
Due to the paucity of human data, dose-response assessment of developmental toxicity is most
often done using animal data. The assessment includes the identification of dose levels associated with
observed developmental effects as well as those doses which apparently produce no adverse effects. The
critical effect is ascertained from the available data. A critical effect is defined as the most sensitive
developmental effect from the most appropriate and/or sensitive mammalian species; LOAEL and
NOAEL determinations are then made. The NOAEL is defined as "the highest dose at which there is no
statistically or biologically significant increase in the frequency of an adverse effect in any of the
possible manifestations of developmental toxicity when compared with the appropriate control group in a
data base characterized as having sufficient evidence for use in risk assessment" (U.S. EPA 1991). The
LOAEL is defined in the following manner: "The LOAEL is the lowest dose at which there is a
statistically or biologically significant increase in the frequency of adverse developmental effects when
compared with the appropriate control group in a data base characterized as having sufficient evidence".
Because of the limitations associated with the use of the NOAEL/LOAEL approach, U.S. EPA is
investigating the use of alternative methods employing more data in a dose-response assessment. One
such approach is the estimation of a benchmark dose (BMD). This approach is based on the use of a
mathematical model to derive an estimate of an incidence level (e.g., 1%, 5%, 10%, etc). This is done by
applying a model to data in the observed range, selecting an incidence level at or near the observed range
(typically 10%), and then determining an upper confidence limit on the modeled curve. The value of the
upper limit, for a 10% incidence, is then used to derive the BMD, which is the lower confidence limit on
dose for that incidence level.
The last step in dose-response assessment is the calculation of a reference dose or reference
concentration for developmental toxicity (RfDDT or RfCDT). This is done by applying appropriate
uncertainty factors to the LOAEL, NOAEL, or BMD. Uncertainty factors generally include the
following:
• 10 for interspecies variation (animal to human)
• 10 for intraspecies variation
Additional factors may be applied to account for other areas of uncertainty, such as identification
of a LOAEL in the absence of a NOAEL. In this case, the factor may be as much as 10 fold, depending
on the sensitivity of the endpoints evaluated in the data base. An uncertainty factor is generally not used
to account for duration of exposure when calculating the RfDDT. If developmental toxicity is the critical
effect for the chronic RfD, an additional uncertainty factor (for study duration) may be used. Modifying
factors may be used to deal with the degree of confidence in the data base for the agent being evaluated.
For a discussion of application of uncertainty factors to BMDs, see U.S. EPA (1991).
6.1.2.3 Germ Cell Mutagenicity
According to U.S. EPA (1986), a dose-response assessment of an agent's potential for human
germ cell mutagenicity can presently be done using only data from in vivo heritable germ cell tests. This
will remain the case until such time as other assays are demonstrated to have an equivalent predictability
for human effects. The usable tests are, thus, limited to the following: morphological specific locus and
6-6
-------
biochemical specific locus assays; and heritable translocation tests. Data from such assays are generated
from exposures much higher than those expected for humans as a consequence of environmental
exposure. Estimation of extent of human risk is done by extrapolating the observed mutation frequency
or phenotypic effects downward to the expected human exposure range. Available data and mechanistic
considerations are used in the choice of the dose-response model and extrapolation procedure.
6.1.2.4 Carcinogenic Effects
Mathematical models can be used, if data are sufficient, to calculate the estimated excess cancer
risk associated with either the ingestion or inhalation of the contaminant if toxicologic evidence leads to
the classification of the contaminant as one of the following: A, Known Human Carcinogen; B, Probable
Human Carcinogen; or C, Possible Human Carcinogen. The data used in these estimates usually come
from lifetime exposure studies using animals. In order to estimate the potential cancer risk for humans
from animal data, animal doses must be converted to equivalent human doses. This conversion includes
correction for noncontinuous exposure for less-than-lifetime exposure studies and for differences in size.
The factor to compensate for the size difference should be determined from appropriate experimental
data. In the absence of such data, a default value should be used, such as the cube root of the ratio of the
animal and human body weights. A default assumption is that the average adult human body weight is
70 kg, that the average water consumption of an adult human is 2 L of water per day, and that the
average adult breathes 20 m3 of air per day.
For contaminants with a carcinogenic potential, chemical levels are correlated with a
carcinogenic risk estimate by employing a cancer potency (unit risk) value together with the assumption
for lifetime exposure. The cancer unit risk has generally been derived by assuming low dose linearity
and applying a mathematical model such as a linearized multistage model with a 95% upper confidence
limit. Cancer risk estimates have also been calculated using other models such as the one-hit, Weibull,
logit and probit. There is little basis in the current understanding of the biologic mechanisms involved in
cancer to suggest that any one of these models is able to predict risk more accurately than any other.
Because each model is based upon differing assumptions, the estimates derived for each model can differ
by several orders of magnitude.
The scientific data base used to calculate and support the setting of cancer risk rate levels has an
inherent uncertainty that is due to the systematic and random errors in scientific measurement. In most
cases, only studies using experimental animals have been performed. Thus, there is uncertainty when the
data are extrapolated to humans. When developing cancer risk rate levels, several other areas of
uncertainty exist, such as the incomplete knowledge concerning the health effects of contaminants in
environmental media, the impact of the experimental animal's age, sex and species, the nature of the
target organ system(s) examined and the actual rate of exposure of the internal targets in experimental
animals or humans. Dose-response data usually are available only for high levels of exposure and not for
the lower levels of exposure closer to where a standard may be set. When there is exposure to more than
one contaminant, additional uncertainty results from a lack of information about possible interactive
effects.
6.2 Hazard Identification for Mercury
Because there are no U.S. EPA guidelines for hazard identification of systemic noncancer
effects, this section does not include a discussion of systemic noncancer effects.
6-7
-------
6.2.1 Developmental Effects
6.2.1.1 Elemental Mercury
Data for developmental effects of elemental mercury are detailed in Section 3.1.3.11 (Tables 3-
25, 3-26 and 3-27). Human studies are inconclusive. The study by Mishinova et al. (1980) provided
insufficient experimental detail to permit evaluation of an exposure-response relationship. Sikorski et al.
(1987) found an increase in reproductive failure among 57 dental professionals by comparison to
controls. This reproductive failure (described as spontaneous abortions, stillbirths or congenital
malformations) was significantly correlated with exposure level. Maternal toxic signs were not reported.
The study was limited by the small population and the lack of description of the control group. These
findings were not reproduced in Ericson and Kallen's 1989 study of 8157 infants born to dental
professionals in Sweden. When compared to the general population, there was no increase in
malformations, abortions or stillbirths. Exposure data were limited in this study.
There are four animal studies evaluating potential developmental effects associated with
exposure to elemental mercury. In Baranski and Szymczyk (1973), female rats (strain not specified)
were exposed to 2.5 mg/m3 mercury vapor for 6 to 8 weeks before fertilization or for 3 weeks prior to
mating and on days 7-20 of gestation. In the first experiment, mortality among pups was increased in
the exposed group, and there were changes in pup organ weights (decreased kidney and liver weight and
increased ovary weight). In the second exposed group, mean number of live pups was decreased;
mortality among pups was 100% by day 6 postpartum. There were signs of frank toxicity in the dams
including spasms, tremors and death. Information is taken from an English translation of this Polish
paper.
Steffek et al. (1987) is reported in abstract. Rats (strain not specified) were exposed to 0.1, 0.5
or 1.0 mg/m3 mercury for either the entire gestation period or for days 10-15. No effects on resorption
or gross abnormality were seen in the low-dose group. Exposure to the mid and high doses for days
10-15 resulted in increased numbers of resorption (5/41 and 7/71, respectively; denominators are
presumed to be numbers of litters ~ not specified in text); exposure for the entire gestational period
resulted in gross defects in 2/115 fetuses in the low dose and increased resorption (19/38) in the high
dose. Maternal and fetal weight was decreased in the group exposed to 1.0 mg/m3 for the entire gestation
period. No statistical analyses were reported in the abstract.
Two studies in rats focused on behavioral changes consequent to inhalation of elemental mercury
during development. In the first, Danielsson et al. (1993) exposed pregnant Sprague-Dawley rats to
mercury vapor at 1.8 mg/m3 for either 1 or 3 hours on gestation days 11-14 and 17-20. There were no
signs of toxicity in the dams and offspring of treated animals were no different from controls on the
following measures: body weight; clinical signs; pinna unfolding; surface righting reflex development;
tooth eruption; and results of a negative geotaxis test at days 7, 8 or 9 post partum. Male rats exposed in
utero were significantly hypoactive by comparison to controls at 3 months and hyperactive at 14 months.
Exposed males were impaired in a test of habituation to novel environments and showed decreased
ability to learn a maze. They were not different from controls in a circular swim test administered at 15
months of age. Females were tested only in the spontaneous motor activity tests; treated females were no
different from controls on this measure.
These results were similar to those reported by Fredriksson et al. (1992). In this instance rats
were exposed postnatally on days 11-17 of age to 0.05 mg Hg/m3 for either 1 or 3 hours/day. High-dose
rats showed increased activity (rearing) at 2 months but had decreased activity by comparison to controls
6-8
-------
at 4 months. Low-dose rats were no different from controls at 2 months; at 4 months this group showed
increased total activity and decreased rearing. In the spatial learning test administered at 6 months low-
dose rats showed increases in time to complete the task. High-dose animals were observed to have
increases in both time to complete the task and in numbers of errors. Data were not reported on gender
differences in behavior as a result of exposure to mercury vapor.
Both of these studies involved exposure during critical developmental periods, one pre-natal and
one post-natal prior to sexual maturity. Both showed differences from controls (by ANOVA) on one of
four major manifestations of developmental effects listed in the Guidelines for Developmental Toxicity
Risk Assessment (U.S. EPA 1991); namely, functional deficits, in this case in locomotion and learning.
In the Danielsson et al. (1993) paper, these deficits were observed in male offspring in the absence of
maternal toxicity, which according to the Guidelines raises the level of concern. The studies suggest that
the observed effects are not reversible. Latency to reach a platform in the circular swim maze was
significant in the high-dose group at 15 months but not at 7 months, and total activity was decreased in
the low-dose group and increased in the high-dose group at 14 months.
The Guidelines specify that for a judgement of Sufficient Experimental Animal
Evidence/Limited Human Data the minimum data set is the following:
" The minimum evidence necessary to judge that a potential hazard exists generally
would be data demonstrating an adverse developmental effect in a single, appropriate,
well conducted study in a single experimental animal species."
As the data set for elemental mercury consists of two appropriate studies albeit with minimal
group sizes and two incompletely reported studies suggestive of effect, the judgement of Sufficient
Experimental Animal Evidence/Limited Human Data is the most appropriate.
6.2.1.2 Inorganic Mercury
Data on developmental effects of mercuric chloride are found in Section 3.2.3.8 (Tables 3-46 and
3-47). There is one study in mice of developmental effects of inhaled mercuric chloride and none in
humans. Selypes et al. (1984) reported increases in delayed ossification and dead or resorbed fetuses as
a consequence of exposure of CFLP/N mice to 0.17 and 1.6 mg Hg/m3 as mercuric chloride in an aerosol
for 4 hours on days 9-12 of gestation. There were no statistical analyses, reporting of blood mercury
levels or evaluation of maternal toxicity.
Developmental effects following oral exposure to methylmercury are reported in five oral studies
in rats and hamsters (Table 3-47). McAnulty et al. (1982) is reported in abstract. Oral (not further
specified) mercuric chloride was administered on days 6-15 of gestation at doses of 6, 9, 12, or 18 mg
Hg/kg-day. This resulted in decreased fetal and placental weights in fetuses in the 6 mg/kg-day group
and malformation at the highest dose. The authors concluded that inorganic mercury was a
developmental toxicant only at doses which were maternally toxic.
Rizzo and Furst (1972) treated Long Evans rats with 0 or 2 mg Hg/kg-day as mercuric oxide on
gestation days 5, 12 or 19. Effects noted were growth retardation and inhibition of eye formation in the
group treated on day 5. No statistical analyses were reported, nor were blood mercury levels given.
Pritchard et al. (1982a) reported in abstract results of treating rats (strain not specified) with
mercuric chloride at 3.0, 6.0 or 12.0 mg Hg/kg-day for about 32 days (gestation day 15 until 25 days
6-9
-------
postpartum). Effects included decreased pup weight and weight gain with a LOAEL of 6.0. In another
experiment reported in an abstract, Pritchard et al. (1982b) exposed female rats to mercuric chloride
doses of up to 18 mg Hg/kg-day before mating and during gestation. High implantation loss was
observed with exposure to 9 mg Hg/kg-day and higher. Embryonic and fetal development was reported
to be unaffected with doses up to 9 mg Hg/kg-day. The abstracts presented insufficient details, and there
was no reporting of statistical analyses.
Gale (1974) gavaged female hamsters (10/group) on gestation day 8 with mercuric acetate at the
following doses: 2.5, 5.0, 16.0, 22.0, 32.0, 47.0, or 63.0 mg Hg/kg-day. There were 3 control animals.
A variety of malformations and growth effects were noted in animals treated with 16 mg/kg-day or
higher. The authors also treated hamsters via other routes. Their evaluation of efficacy in production of
fetal effects was i.p. > i.v. > s.c. > oral. Maternal toxicity included weight loss, diarrhea, slight tremor,
somnolence, tubular necrosis and hepatocellular necrosis (dose levels not specified). There was
insufficient detail reported for determination of aNOAEL for dams.
In addition to studies of oral or inhalation administration of inorganic mercury there are several
studies which indicate that mercury salts cause developmental toxicity when delivered i.p., s.c. or i.v.
routes (Bernard et al. 1992; Gale and Perm 1971; Gale 1974, 1981; Kajiwara and Inouye 1986, 1992;
Kavlock et al. 1993). In Gale (1981), wherein exposure was of six strains of hamster to mercuric acetate,
s.c., there was no description of maternal toxicity. In Kavlock et al. (1993) (s.c., rats, mercuric acetate)
fetal effects were noted at doses above the lowest observed maternally toxic dose. Kajiwara and Inouye
(1986) reported their opinion that in mice injected i.v. with mercuric chloride, fetal toxicity was related
to maternal toxicity. In their 1992 study, there was no determination whether implantation loss in
mercuric chloride exposed dams was due to fetal toxicity or to maternal uterine dysfunction. The effects
reported by Bernard et al. (1992) can be better characterized as a transitory nephrotic effect rather than a
developmental deficit.
Each of these studies is limited in its usefulness for assessment of the risk of inorganic mercury
to cause human developmental toxicity. The data base as a whole suggests an effect of inorganic
mercury at doses as low as 2 mg Hg/kg-day. The data, however, are considered insufficient for risk
assessment based on any single study or on the database as a whole (Insufficient Evidence, in the
language of the Guidelines).
6.2.1.3 Methylmercury
Data for developmental effects of methylmercury are presented in Section 3.3.3.8 (Tables 3-66,
3-67 and 3-68); studies are primarily by the oral route and none by the inhalation route. Human studies
of developmental effects include evaluation of children born to mothers exposed to contaminated grain in
Iraq (Amin-Zaki et al. 1976; Marsh et al. 1981, 1987) and contaminated fish in Japan (Harada 1978).
Effects noted in the Iraqi children included delays in speech and motor development, mental retardation,
reflex abnormalities and seizures. Infants born to mothers ingesting fish from the contaminated
Minamata Bay in Japan appeared normal at birth. Within several months, however, the following effects
were noted: mental retardation, retention of primitive reflexes, cerebellar symptoms, dysarthria,
hyperkinesia, hypersalivation, strabismus and pyramidal symptoms. Histologic examination of brain
tissues of infants from both populations showed a number of signs of pathology. Kjellstrom et al.
(1989), in a study of a population in New Zealand, has observed an inverse correlation between IQ in
children and hair mercury levels in their mothers. In a group of Cree Indians in Quebec, maternal hair
mercury level was correlated with abnormal muscle tone in male children (McKeown-Eyssen et al.
1983).
6-10
-------
Numerous animal studies have demonstrated a variety of developmental effects occurring in rats,
mice and monkeys exposed orally to methylmercury and are presented in Chapter 3 (Table 3-68).
Developmental effects have been observed in offspring of rats of three strains treated orally with
methylmercury. Developmental effects have also been seen in two strains of mice as well as in guinea
pigs, hamsters and monkeys.
In rodents exposed in utero, decreased fetal weight and increased fetal malformations and deaths
have been reported (Fuyuta et al. 1978, 1979; Inouye and Kajiwara 1988a; Inouye and Murakami 1975;
Khera and Tabacova 1973; Nolen et al. 1972; Reuhl et al. 1981; Yasuda et al. 1985).
Methylmercury exposure during gestation as well as during the lactation period produces
neurodevelopmental effects (structural and functional alterations) in the exposed pups. Structural effects
include lesions in the brain mantle, corpus callosum, caudate putamen, and cerebellum. In guinea pigs,
early gestational exposures (weeks 3-5 of pregnancy) resulted primarily in developmental disturbances
of the brain (smaller brains, dilated lateral ventricles, and reduced size of caudate putamen), whereas
later gestational exposures (>week 6 of pregnancy) resulted in widespread neuronal degeneration (Inouye
and Kajiwara 1988b). Functional changes include abnormal tail position during walking, flexion,
hindlimb crossing, decreased locomotor activity, increased passiveness and startle-response, impaired
maze performance, operant behavior, swimming behavior, tactile-kinesthetic function, visual recognition
memory, and temporal discrimination (Bornhausen et al. 1980; Buelke-Sam et al. 1985; Burbacher et al.
1990; Eisner 1991; Geyer et al. 1985; Gunderson et al. 1988; Hughes and Annau 1976; Inouye et al.
1985; Musch et al. 1978; Olson and Boush 1975; Rice 1992; Rice and Gilbert 1990; Stoltenburg-
Didinger and Markwort 1990; Suter and Schon 1986).
While there are limitations to some of these studies (e.g., lack of information on BML, small
study size), the totality of the data base supports a judgment of Sufficient Human and Animal Data for
Developmental Toxicity of methylmercury.
6.2.2 Germ Cell Mutagenicity
6.2.2.1 Elemental Mercury
Data for genotoxicity of elemental mercury are described in Section 3.1.3.13 (Table 3-30).
Results for an association of somatic cell chromosomal effects with occupational exposure to elemental
mercury are variable. Popescu et al. (1979) and Verschaeve et al. (1976) reported increased incidence of
aberrations or aneuploidy. Most recently Barregard et al. (1991) showed a significant correlation
between cumulative exposure to elemental mercury and micronuclei induction in T-lymphocytes.
Negative results were reported by Verschaeve et al. (1979) and Mabille et al. (1984). No studies of
mutagenic effect are reported.
Elemental mercury once absorbed is widely distributed throughout the body; there are no data,
however, on elemental mercury in gonadal tissue. Based on both positive and negative findings for
somatic cell chromosomal aberrations in workers, elemental mercury is placed in a group of low
confidence for potential as a human germ cell mutagen.
6-11
-------
6.2.2.2 Inorganic Mercury
Data for genotoxic effects of inorganic mercury are described in Section 3.2.3.10 (Table
3-50). There are no data on inorganic mercury from human germ cell mutagenicity studies or from
studies on heritable mutational events in animals. Anwar and Gabal (1991) reported a statistically
significant increase by comparison to age-matched controls in both chromosomal aberrations and
micronuclei in lymphocytes of workers exposed to mercury fulminate. There was a correlation between
frequency of aberrations and exposure duration. Elemental mercury has been shown to be clastogenic
both in vivo and in vitro. Results of tests for mutagenicity have been variable; generally test results in
prokaryotes are negative for mutagenicity (but may be positive for DNA damage), and results in
eukaryotes are positive. Suter (1975) observed a small, but statistically significant increase in non-viable
implants when female mice were administered mercuric chloride intraperitoneally; the authors were not
certain whether this was a true dominant lethal effect or was attributable to maternal toxicity.
Chromosome aberrations were observed in somatic cells in occupationally exposed humans
(Anwar and Gabal 1991), in somatic cells of mice exposed by gavage (Ghosh et al. 1991), and in Chinese
Hamster Ovary cells treated in vitro (NTP 1993; Howard et al. 1991). Sex-linked recessive mutations
were not observed in Drosophila (NTP 1993), and positive results in a dominant lethal test were
compromised by maternal toxicity (Suter 1975). There are other data for DNA damage and limited data
for gene mutation. Inorganic mercury is less well-distributed in the body than is elemental mercury; it
does not readily pass blood-brain or placental barriers. In one reported study (Jagiello and Lin 1973),
mice treated intraperitoneally were not shown to have an increased incidence of aneuploidy in
spermatogonia. Watanabe et al. (1982), however, showed that while hamsters injected s.c with mercuric
chloride had no increase in aberrations in metaphase II oocytes, there was detectable mercuric chloride in
ovaries and some inhibition of ovulation.
The totality of available data indicates a moderate weight of evidence for germ cell
mutagenicity: sex-linked recessive and dominant lethal results were compromised, but there are positive
results for chromosomal aberrations in multiple systems (including in vivo exposure) and evidence that
mercuric chloride can reach female gonadal tissue.
6.2.2.3 Methylmercury
Summaries of data for genotoxicity of methylmercury are presented in Section 3.3.3.10 (Tables
3-70, 3-7 land 3-72).
Methylmercury appears to be clastogenic but not a potent mutagen. Methylmercury is widely
distributed in the body, breaching both blood-brain and placental barriers in humans. There are data
indicating that methylmercury administered i.p. reaches germ cells and may produce adverse effects.
Suter (1975) observed a slight reduction in both numbers of implantations and viable embryos in (SEC x
C57Bl)Fj females which had been mated to treated males. This was not noted in (101 x C3H)Fj mice.
When Syrian hamsters were treated intraperitoneally with methylmercury, aneuploidy but not
chromosomal aberrations was seen in oocytes. Sex-linked recessive lethal mutations were increased in
Drosophila melanogaster given dietary methylmercury. Watanabe et al. (1982) noted some decrease in
ovulation in hamsters treated s.c. with methylmercury, further indication that methylmercury is
distributed to female gonadal tissue.
Studies have reported increased incidence of chromosome aberrations (Skerfving et al. 1970,
1974) or SCE (Wulf et al. 1968) in lymphocytes of humans ingesting mercury-contaminated fish or
6-12
-------
meat. Chromosome aberrations have been reported in cats treated in vivo and in cultured human
lymphocytes in vitro. Evidence of DNA damage has been shown in a number of in vitro systems.
As there are data for mammalian germ cell chromosome aberration and limited data from a
heritable mutation study, methylmercury is placed in a group of high concern for potential human germ
cell mutagenicity. All that keeps methylmercury from the highest level of concern is lack of positive
results in a heritable mutation assay.
6.2.3 Carcinogenic Effects
This section presents the critical carcinogenicity studies evaluated by the U.S. EPA for the
weight-of-evidence classification of elemental, inorganic (mercuric chloride) and organic
(methylmercury) forms of mercury. These studies are discussed more completely in Chapter 3 and
summarized in Tables 3-1, 3-31, 3-32, 3-33, 3-52, and 3-53.
6.2.3.1 Elemental Mercury
Human data regarding the carcinogenicity of inhalation of elemental mercury are insufficient to
determine whether such exposures may result in increased cancer incidence. Several studies report
statistically significant increases in lung cancer mortality among groups of exposed workers (Amandus
and Costello 1991; Barregard et al. 1990; Buiatti et al. 1985; Ellingsen et al. 1992). The interpretation of
these studies is limited by small sample sizes, probable exposure to other known lung carcinogens,
failure to consider confounders such as smoking and failure to observe correlations between estimated
exposure and the cancer incidence. A study of dental professionals found a significant increase in the
incidence of glioblastomas (Ahlbom et al. 1986). It is not known whether exposure to mercury, X-rays,
or other potential carcinogens in the workplace contributed to the effects observed. No increase in
cancer mortality was observed among workers exposed to mercury vapor in a nuclear weapons facility
(Cragle et al. 1984), but this study was also limited by the small sample size. No studies were identified
that examined cancer incidence in animals exposed chronically to elemental mercury vapor. These
studies are presented in greater detail in Section 3.1.2.
The overall findings from cytogenetic monitoring studies of workers occupationally exposed to
mercury by inhalation provide very limited evidence of genotoxic effects. Popescu et al. (1979)
compared four men exposed to elemental mercury vapor with an unexposed group and found an
increased number of chromosomal aberrations. Verschaeve et al. (1979) found an increased incidence of
aneuploidy after exposure to low concentrations.
In summary, human epidemiological studies failed to show a correlation between exposure to
elemental mercury vapor and increased cancer incidence, but the studies are limited by confounding
factors. Only one study in animals is reported (Druckrey et al. 1957); tumors were found only at contact
sites, and the study is incompletely reported as to controls and statistics. Animal data are, thus, also
inadequate. Findings from assays for genotoxicity are limited and provide no convincing evidence that
mercury exposure has an effect on the number or structure of chromosomes in human somatic cells. The
most appropriate category is, thus, Group D, not classifiable as to human carcinogenicity.
This classification was reviewed by the Carcinogen Risk Assessment Verification Endeavor
(CRAVE), an Agency Peer Review Work Group. The classification was accepted as appropriate on
March 3, 1994.
6-13
-------
6.2.3.2 Inorganic Mercury
There are no data available on the carcinogenic effects of inorganic mercury (mercuric chloride)
in humans. In animals, there is equivocal evidence of carcinogenicity in rats and mice. In rats gavaged
with mercuric chloride for two years (NTP 1993), survival was significantly reduced in males (17% and
8% survival in low and high-dose males versus 43% survival in controls), indicating that the maximally
tolerated dose (MTD) was exceeded. There was an increased incidence of forestomach squamous cell
papillomas (0/50, 3/50, 12/50 in control, low, and high-dose males, respectively; 0/50, 0/49 and 2/50 in
control, low and high-dose females, respectively). Papillary hyperplasia of the forestomach was also
significantly elevated in both male dose groups and in high-dose females. In addition, the incidence of
thyroid follicular cell carcinomas in treated males (1/50, 2/50 and 6/50 in control, low- and high-dose
males, respectively) showed a significantly positive trend. There were, however, no increases in thyroid
hyperplasia of adenomas; it is not clear that the increase in thyroid carcinomas is a treatment-related
effect. The NTP also considered the forestomach tumors to be of limited relevance to humans; there was
no evidence that these contact site tumors progressed to malignancy.
In a companion study in mice (NTP 1993), there was a significantly increasing trend for renal
tubular cell tumors (adenomas and adenoma carcinomas). No dose groups were statistically significantly
different from the control by pair-wise comparison, although the incidence in the high-dose group was
elevated. There was a significant increase in severe nephropathy in treated animals. The NTP studies
and two nonpositive bioassays are summarized in Section 3.2.2.
In summary, there are no data in humans linking mercuric chloride with carcinogenic effects.
Data in animals are limited. Focal hyperplasia and squamous cell papillomas of the forestomach as well
as thyroid follicular adenomas and carcinomas were observed in male rats gavaged with mercuric
chloride. In the same study, evidence for increased incidence of squamous cell forestomach papillomas
in female rats and renal adenomas and carcinomas in male mice were considered equivocal. All
increased tumor incidences were observed at what were considered high doses (in excess of the MTD).
In this context, the relevance of the thyroid tumor to human health evaluation has been questioned; these
tumors are considered to be secondary to the hyperplastic response. Results from in vitro and in vivo
tests for genotoxicity have been mixed with no clear indication of a strong somatic cell genotoxic effect
of mercuric chloride exposure.
Based on the absence of human data and limited data for carcinogenicity in animals, mercuric
chloride is classified as Group C, possible human carcinogen. This classification was reviewed by
CRAVE on March 3, 1994 and found to be appropriate.
6.2.3.3 Methylmercury
The available human data are inconclusive regarding the carcinogenicity of methylmercury in
humans exposed by the oral route. A study of leukemia patients from a rural area in Poland showed a
significantly higher mercury content in hair in the leukemia patients than in healthy unrelated patients or
healthy relatives (Janicki et al. 1987). The population studied was small, and the study did not adjust for
other leukemia risk factors. In addition, two studies of larger populations exposed to methylmercury
during the Minamata incident failed to show increases in leukemia or total cancer incidence (Tamashiro
et al. 1984, 1986). Although one of these studies showed a significant increase in liver cancer incidence,
factors other than mercury exposure were likely contributors to the increase. These epidemiological
studies are presented in greater detail in Section 3.3.2.1.
6-14
-------
Animal studies show some evidence of carcinogenicity in two strains of mice, but studies in rats
have not shown similar results. Male ICR mice given methylmercuric chloride in the diet for up to two
years had significantly increased incidences of renal epithelial adenomas and/or adenocarcinomas
(Hirano et al. 1986; Mitsumori et al. 1981). Similarly, male B6C3F1 mice given methylmercuric
chloride in the diet for up to two years had significantly increased incidences of renal epithelial
carcinomas and adenomas (Mitsumori et al. 1990). In contrast, Sprague-Dawley rats administered
methylmercury in the diet for up to 130 weeks exhibited no increase in tumor incidence (Mitsumori et al.
1983, 1984). Although the dose was lower in the rats than in the mice, a maximally tolerated dose was
achieved in the rat study as evidenced by an approximately 20-30% decrease in body weight gain and by
significant increases in renal and neuronal toxicity in both male and female rats at the highest dose
tested. Other studies also failed to show increases in tumor incidence after chronic exposure to
methylmercury (Schroeder and Mitchener 1975; Verschuuren et al. 1976), but these studies were limited
by small sample sizes, failure to achieve a maximally tolerated dose and/or incomplete histopathological
examinations. These studies are presented more completely in Section 3.3.2.2.
In summary, data for carcinogenicity from human studies are considered inadequate. Three
studies that examined the relationship between methylmercury exposure in humans and increased
incidence of cancer were limited by poor study design or incomplete description of methodology or
results. Data from animal studies are considered to provide limited evidence of carcinogenicity. Male
ICR and B6C3F1 mice exposed to methylmercuric chloride in the diet were observed with increased
incidence of renal adenomas, adenocarcinomas and carcinomas. Tumors were observed at a single site,
in a single species and sex. Renal epithelial cell hyperplasia and tumors were observed only in the
presence of profound nephrotoxicity; tumors were suggested to be consequent to reparative changes in
the affected organs. Although genotoxicity test data suggest that methylmercury is clastogenic, there are
also negative tests.
The limited data in animals above support a categorization of Group C, possible human
carcinogen. The CRAVE Work Group accepted this weight-of-evidence judgment as appropriate at its
March 3, 1994 meeting.
6.2.3.4 Application of proposed revision of the Guidelines for Carcinogen Risk Assessment
Data described in the above three sections were re-evaluated using criteria described in the
proposed revisions to the Guidelines for Carcinogen Risk. Among the changes in the revised guidelines
are emphasis on use of data describing the mode of action of the putative carcinogen, both in weight of
evidence judgements and in decisions as to the most appropriate type of low dose extrapolation. The
revised Guidelines encourage consideration of relevance to human health risk of route and magnitude of
exposure in animal bioassays. Both of these considerations were part of a re-evaluation of the data for
carcinogenicity of the three forms of mercury. Results of a presentation (Schoeny, 1996) on the
application of the revised Guidelines are summarized below.
6-15
-------
Elemental mercury
The likelihood of elemental mercury to be a human carcinogen cannot be determined; data in
humans and animal bioassays are inadequate. Epidemiologic studies, though confounded showed no
correlation between exposure to elemental mercury vapor and carcinogenicity. Animal data were not
positive, but the published study was considered inadequate. The study was done by a route
(intramuscular injection) not relevant to human exposure (inhalation). Genetic toxicity data are limited
and equivocal.
Inorganic (mercuric chloride)
The data for inorganic mercury indicate that it is not likely to be a human carcinogen under
conditions of exposure generally encountered in the environment. There are no data on carcinogenicity in
humans. Findings in animals included squamous cell papillomas of the forestomach and thyroid
follicular cell adenomas and carcinomas in gavaged rats, and renal adenomas and adenocarcinomas in
male mice. All increased tumor incidences were at high doses (in excess of the MTD). Genetic toxicity
data gave a mix of positive and negative response for chromosomal breakage and were equivocal for
somatic cell point mutations. The mode of action for forestomach tumors appears to be a high dose
effect related to irritation and cytotoxicity. The mode of action for thyroid tumors is not clear; there was
no treatment-related increase in incidence of hyperplasia. There was high mortality in rats from renal
toxicity. Renal toxicity in male mice was less severe; the mode of action of inorganic mercury in
producing renal neoplasms is not clear. Human exposure (other than occupational or accidental
poisoning) is likely to be to low levels of inorganic mercury in water or food plants.
Methylmercury
Methylmercury is not likely to be a human carcinogen under conditions of exposure generally
encountered in the environment. Data in humans were inadequate; interpretation is limited by
inappropriate study design and incomplete descriptions of methodology. Dietary exposure in two strains
of mice resulted in increased renal adenomas and adenocarcinomas. Tumors were observed only in dose
groups experiencing profound nephrotoxicity. Studies in rats exposed to a MTD showed no increased
tumor incidence. Several studies show that methylmercury can cause chromosomal damage in somatic
cells. While evidence is good for chromosomal effects, it does not appear that methylmercury is a point
mutagen. The mode of action in renal tumor induction is likely to be related to reparative changes in the
tissues. Human exposure is likely to be from consumption of contaminated foods especially fish. It is
expected that exposure, even in groups consuming large amounts offish from contaminated sources, will
be to levels far below those likely to cause the tissue damage associated with tumor formation in
animals.
6.3 Dose-Response Assessment For Mercury
6.3.1 Systemic Noncancer Effects
6.3.1.1 Oral Reference Doses (RfDs)
Elemental mercury
Metallic mercury is only slowly absorbed by the gastrointestinal tract (-0.01%) and because of
this is thought to be of no toxicological consequence (Klaassen et al. 1986) when ingested. Further
discussion of an RfD for this form of mercury is not presented.
6-16
-------
Inorganic mercury (mercuric chloride)
An RfD for inorganic mercury of 3xlO~4 mg/kg-day has been verified by the RfD/RfC Work
Group. The critical effect serving as the basis for the RfD is kidney toxicity due to an auto-immune
disease caused by the accumulation of IgG antibodies in the glomerular region of the kidneys.
On October 26 and 27 of 1987, a panel of mercury experts met at a Peer Review Workshop
convened by U.S. EPA for the purpose of reviewing outstanding issues concerning the health effects and
risk assessment of inorganic mercury (U.S. EPA 1987). The panel participants are listed in Appendix C.
Five consensus conclusions and recommendations were agreed to as a result of this workshop; these are
presented in Table 6-1. The RfD was determined using data on autoimmune glomerulonephritis
observed in rats. Based on three studies using the Brown-Norway rat, a DWEL value was determined
using studies described below. The Brown-Norway rat is very sensitive to this mercuric mercury-
induced autoimmune effects, although this effect has also been demonstrated in other strains of rats and
other species of experimental animals (Andres 1984; Bernaudin et al. 1981; Hultman and Enestrom
1992). The Brown-Norway rat is believed to be a good surrogate for the study of mercury-induced
kidney damage in sensitive humans (U.S. EPA 1987b). The glomerulonephritis is characterized by
deposition of anti-glomerular basement membrane antibodies (IgG) in renal glomeruli and after
prolonged exposure is often accompanied by proteinuria and, in some cases, nephrosis (Druet et al.
1978).
LOAEL values were identified from three individual studies. In Druet et al. (1978), Brown-
Norway rats were exposed to mercuric chloride via subcutaneous injection, 3 times/week, for 8 weeks.
The dose levels administered were 0, 0.1, 0.25, 0.5, 1.0 and 2.0 mg Hg/kg, and there were
6-20 animals/group. An additional group of animals received 0.05 mg Hg/kg for 12 weeks. (The
number of animals/sex was not stated.) Druet and colleagues measured antibody formation (using a
fluoresceinated sheep anti-rat IgG antiserum) and urinary protein levels. Proteinuria occurred at doses
> 0.1 mg/kg (LOAEL); the proteinuria was considered a highly deleterious effect, as it frequently led to
hypoalbuminemia and even death. A LOAEL for lifetime exposure was calculated to be 0.226 mg/kg-
day, using the following conversion:
0.05 mg/kg x 3 days/7 days x 0.739 [HgCl2 - Hg2+] x 100% absorption/7%
= 0.226 mg Hg/kg-day
In a 60-day study conducted by Bernaudin et al. (1981), Brown-Norway rats (5/group) were
force-fed 0 or 3 mg/kg/week mercuric chloride. At the end of the 60 days, there were no classic
histological abnormalities in the kidneys of treated animals. Using immunofluorescence, however, IgG
deposition was evident in all of the treated rats, and weak proteinuria was noted in 3/5 dosed animals. A
lifetime LOAEL was calculated to be 0.317 mg Hg/kg-day. Dose conversion was done in the following
manner:
3 mg/kg x 1 day/7 days x 0.739 [HgCl2 - Hg2+]
= 0.317 mg Hg/kg-day
6-17
-------
Table 6-1
Consensus Decisions of Peer Review Panel
The most sensitive adverse effect for mercury risk assessment is formation of mercuric
mercury-induced autoimmune glomerulonephritis. The production and deposition of IgG
antibodies to the glomerular basement membrane can be considered the first step in the
formation of this mercuric mercury-induced autoimmune glomerulonephritis.
The Brown-Norway rat should be used for mercury risk assessment. The Brown-Norway rat
is a good test species for the study of Hg2+-induced autoimmune glomerulonephritis. The
Brown-Norway rat is not unique in this regard (i.e., this effect has also been observed in
rabbits).
The Brown-Norway rat is a good surrogate for the study of mercury-induced kidney damage
in sensitive humans. For this reason, the uncertainty factor (for interspecies variability) used
to calculate criteria and health advisories (based on risk assessments using the Brown Norway
rat) should be reduced by 10-fold.
Hg2+ absorption values of 7% from the oral route and 100% from the subcutaneous route
should be used to calculate criteria and health advisories.
A DWEL of 0.010 mg/L was recommended based on the weight of evidence from the studies
using Brown Norway rats and limited tissue data.
Similar results were obtained by Andres (1984). Five Brown-Norway rats were exposed to
3 mg/kg mercuric chloride via gavage 2 times/week for 60 days. In this same study, Lewis rats (n=2)
were also exposed using the same dosing regimen. After 60 days, the kidneys of all treated animals
appeared normal histologically, and no proteinuria was reported in any treated animals; IgG deposition in
the renal glomeruli was demonstrated using immunofluorescence in Brown-Norway rats. No antibody
deposition was noted in the Lewis rats. The lifetime LOAEL was determined to be 0.633 mg Hg/kg-day.
Dose conversion was done in the following manner:
3 mg/kg x 2 days/7 days x 0.739 [HgCl2 - Hg2+]
= 0.633 mg Hg/kg-day
As the result of intensive review of these and other studies, as well as the discussions of the
panel of mercury experts convened for this purpose, a recommended DWEL of 0.01 mg/L was derived
from the LOAELs above, and, subsequently, the oral RfD value was back-calculated:
RfD = DWEL x 2 Uday
70 kg bw
RfD = 0.010 mg/L x 2L/day/70 kg bw
= 0.0003 mg/kg bw/day
6-18
-------
The RfD for inorganic mercury was reviewed by the RfD/RfC Work Group which reached
consensus for verification on November 16, 1988. The Work Group agreed to application of an
uncertainty factor of 1000 to the LOAELs above (which ranged from 0.23 to 0.63 mg Hg/kg-day). The
uncertainty factor was composed of a 10-fold each for subchronic to chronic and LOAEL to NOAEL
extrapolation, and an additional 10-fold factor for both animal to human and sensitive populations. The
resulting RfD of 3xlO"4 mg/kg-day was given high confidence based on the weight of the evidence from
the studies using Brown Norway rats and the entirety of the data base.
A literature search for the years 1988 to 1994 has been conducted and recently reviewed
(September 1994). The NTP (1993) study was among those considered. A rat NOAEL of 0.23 mg
Hg/kg administered dose has been identified for renal effects for the 6-month portion of the study. A
description of the NTP gavage study has been included in the summary information for IRIS. U.S. EPA
concluded that no change in the RfD for inorganic mercury is needed at this time.
Methylmercury
U.S. EPA has on two occasions published RfDs for methylmercury which have represented the
Agency consensus for that time. These are described in the sections below. The original RfD of 0.3
(jg/kg/day was determined in 1985. The current RfD of 0.1 (jg/kg/day was established as Agency
consensus in 1995. At the time of the generation of the Mercury Study Report to Congress, it became
apparent that considerable new data on the health effects of methylmercury in humans were emerging.
Among these are large studies offish or fish and marine mammal consuming populations in the
Seychelles and Faroes Islands. Smaller scale studies are in progress which describe effects in
populations around the U.S. Great Lakes. In addition, there are new evaluations, including novel
statistical approaches and application of physiologically-based pharmacokinetic (PBPK) models, to
published work described in section 3.3.1.1 of this volume.
As much of this new data has either not yet been published or have not yet been subject to
rigorous review, it was decided that it was premature for U.S. EPA to make a change in the 1995
methylmercury RfD at this time. This decision was approved by the Science Advisory Board (SAB), a
public advisory group providing extramural scientific information and advice to the Administrator and
other officials of the Environmental Protection Agency. The SAB is structured to provide balanced,
expert assessment of scientific matters relating to problems facing the Agency. Their report makes the
following statement.
"In general, from the standpoint of looking at human health effects and the uncertainties, the
draft report is a very good document and an important step forward in terms of bringing the relevant
information together into one place for the first time. The current RfD, based on the Iraqi and New
Zealand data, should be retained at least until the on-going Faeroe and Seychelles Islands studies have
progressed much further and been subjected to the same scrutiny as has the Iraqi data."
The SAB report continues:
"Investigators conducting two new major prospective longitudinal studies-one in the Seychelles
Islands the other in the Faeroe Islands—have recently begun to publish findings in the literature and are
expected to continue releasing their findings during the next 2-3 years. These studies have advantages
over those cited in the previous paragraph in that they have much larger samples sizes, a larger number
of developmental endpoints, potentially more sensitive developmental endpoints, and control a more
extensive set of potential confounding influences. On the other hand, the studies have some limitations
in terms of low exposures (to PCBs in the Faeroes) and ethnically homogenous societies. Since only a
6-19
-------
small portion of these new data sets have been published to date and because questions have been raised
about the sensitivity and appropriateness of the several statistical procedures used in the analyses, the
Subcommittee concluded that it would be premature to include any data from these studies in this report
until they are subjected to appropriate peer review. Because these data are so much more
comprehensive and relevant to contemporary regulatory issues than the data heretofore available,
once there has been adequate opportunity for peer review and debate within the scientific
community, the RfD may need to be reassessed in terms of the most sensitive endpoints from these
new studies."
An inter-agency process, with external involvement will be undertaken for the purpose of
reviewing these new data, their evaluations, and the evaluations of existing data. An outcome of this
process will be an assessment by U.S. EPA of its RfD for methylmercury to determine if a change is
warranted.
Former RfD
A hazard identification and dose-response assessment was proposed for methylmercury in 1980
(U.S. EPA 1980) and later verified by the RfD/RfC Work Group on December 2, 1985. This assessment
was subsequently included on U.S. EPA's Integrated Risk Information System (IRIS). The critical
effects were multiple central nervous system (CNS) effects including ataxia and paresthesia in
populations of humans exposed to methylmercury through consumption of contaminated grain
(summarized by Clarkson et al. 1975, Nordberg and Strangert 1976 and WHO 1976); see study
descriptions in Section 3.3.
The RfD for methylmercury was determined to be 3xlO"4 mg/kg-day, based on a LOAEL of
0.003 mg/kg-day (corresponding to 200 (jg/L blood concentration) and an uncertainty factor of 10 used
to adjust the LOAEL to what is expected to be a NOAEL. An additional uncertainty factor of 10 for
sensitive individuals for chronic exposure was not deemed necessary at the time of the RfD's
verification, as the adverse effects were seen in what was regarded as a sensitive group of individuals,
namely adults who consumed methylmercury-contaminated grain.
Medium confidence was ascribed to the choice of study, data base and RfD. The blood levels
associated with the LOAEL were well supported by more recent data, but neither the chosen studies nor
supporting data base described a NOAEL. Medium confidence indicates that new data may change the
assessment of the RfD.
Since the time of verification, several submissions to IRIS have questioned the value of this RfD,
and, specifically, whether or not this RfD is protective against developmental effects. Subsequent to the
RfD verification, the effects in Iraqi children of in utero exposure to methylmercury were reported by
Marsh et al. (1987). Discussion of the methylmercury RfD by the RfD/RfC Work Group was reported in
1992 and 1994. Consensus for verification of the RfD described below was reached in January of 1995.
6-20
-------
Current U.S. EPA RfD
Determination of critical effect
Marsh et al. (1987) was chosen as the most appropriate study for determination of an RfD
protective of a putative sensitive subpopulation; namely infants born to mothers exposed to
methylmercury during gestation. This paper describes neurologic abnormalities observed in progeny of
women who consumed bread prepared from methylmercury-treated seed grain while pregnant (See
Chapter 3 for study description). Among the signs noted in the infants exposed during fetal development
were cerebral palsy, altered muscle tone and deep tendon reflexes as well as delayed developmental
milestones (i.e., walking by 18 months and talking by 24 months). Each child in the study was examined
by two neurologists who scored observed effects on a scale for severity ranging from 0 to 11. The data
collected by Marsh et al. (1987) summarize clinical neurologic signs of 81 mother and child pairs. From
x-ray fluorescent spectrometric analysis of selected regions of maternal scalp hair, concentrations
ranging from 1 to 674 ppm mercury were determined, then correlated with clinical signs observed in the
affected members of the mother-child pairs. Among the exposed population there were affected and
unaffected individuals throughout the exposure range.
Method employed for determination of critical dose
In order to quantitate an average daily mercury ingestion rate for the mothers, hair
concentrations were determined for periods during gestation when actual methylmercury exposure had
occurred. This procedure is possible since hair grows an average rate of 1 cm/month (Al-Shahristani et
al. 1976) and since Iraqi women wear their hair very long; appropriate samples were, thus, available for
the period of gestation when exposure occurred.
A number of laboratory studies support a correlation between hair concentrations and concurrent
blood concentrations. Some variation in the ratio exists; a ratio of 250:1 (|ig mercury/mg in hairing
mercury/ml of blood) was used to derive the RfD critical dose. A more complete discussion for the
choice of this ratio is provided in the next section.
The hair concentration at a hypothetical NOAEL for developmental effects was determined by
application of a benchmark dose approach (see subsequent section for discussion of methods and data
used). The analysis used the combined incidence of all neurological effects in children exposed in utero
as reported in the Marsh et al. (1987) study. A Weibull model for extra risk was used to determine the
benchmark dose of 11 ppm mercury in maternal hair (11 mg/kg hair). This was converted to 44 //g/L
blood using the above 250:1 ratio.
11 mg/kg hair / 250 =44 //g/L blood
To obtain a daily dietary intake value of methylmercury corresponding to a specific blood concentration,
factors of absorption rate, elimination rate constant, total blood volume and percentage of total mercury
that is present in circulating blood were taken into account. Calculation was by use of the following
equation based on the assumptions that steady state conditions exist and that first-order kinetics for
mercury are being followed.
d f.g/day - CxbxV
A x f
6-21
-------
Where:
d = daily dietary intake (expressed as ug of methylmercury)
C = concentration in blood (expressed as 44 ug/liter)
b = elimination constant (expressed as 0.014 days"1)
V = volume of blood in the body (expressed as 5 liters)
A = absorption factor (expressed as a unitless decimal fraction of 0.95)
f = fraction of daily intake taken up by blood (unitless, 0.05)
Solving for d gives the daily dietary intake of mercury which results in a blood mercury concentration of
44 //g/L. To convert this to daily ingested dose (/ug/kg-day) a body weight of 60 kg was assumed and
included in the equation denominator.
, c x b x V
a =
A x f x bw
d = 44 fj.g/L x 0.014 days 1 x 5L
0.95 x 0.05 x 60 kg
d = 1.1 /^g/kg-day
The dose d (1.1 /ug/kg-day) is the total daily quantity of methylmercury that is ingested by a 60 kg
individual to maintain a blood concentration of 44 /ug/L or a hair concentration of 11 ppm.
The rationales for use of specific values for equation parameters follow.
Hair to blood concentration ratio. The hairblood concentration ratio for total mercury is
frequently cited as 250:1 expressed as /ug mercury/g hair to /ug mercury/ml of blood. Ratios reported in
the literature range from 140 to 416, a difference of about a factor of 3. Table 6-2 provides the results of
12 recent studies in which hair to blood ratios were calculated for a variety of human populations.
Differences in the location of hair sampled (head versus chest and distance from scalp) may contribute to
the differences observed. Variability in the hair-blood relationship for mercury concentration can also be
attributed to the fact that unsegmented hair analysis gives a time-weighted average of mercury exposure,
while analysis of mercury in blood reflects a much shorter period average of exposure. As much as a 3-
fold seasonal variation in mercury levels was observed in average hair levels for a group of individuals
with moderate to high fish consumption rates, with yearly highs occurring in the fall and early winter
(Phelps et al. 1980; Suzuki et al. 1992). The relatively high ratio reported by Tsubaki (Table 6-2) may
have reflected the fact that mercury levels were declining at the time of sampling so that the hair levels
reflect earlier, higher blood levels. Cernichiari et al. (1995a) reported a maternal hairblood ratio of
416:1 for residents of the Seychelles Islands. The authors remarked that while this ratio was high,
statistical uncertainties do not permit a judgement as to whether it is truly outside the range reported in
WHO (1990) recapitulated in Table 6-2. Phelps (1980) obtained multiple blood samples and sequentially
analyzed lengths of hair from individuals. Both hair and blood samples were taken for 339 individuals in
Northwestern Ontario. After reviewing the various reports for converting hair concentrations to blood
concentrations, the analysis in the Phelps (1980) paper was selected by the Agency RfD/RfC Work
Group because of the large sample size and the attention to sampling and analysis that was made. The
6-22
-------
ratio Phelps observed between the total mercury concentration in hair taken close to the scalp and
simultaneous blood sampling for this group was 296:1. To estimate the actual ratio the authors assumed
that blood and hair samples were taken following complete cessation of methylmercury intake. They
also assumed ahalf-life of methylmercury in blood of 52 days and a lag of 4 weeks for appearance of the
relevant level in hair at the scalp. Phelps also determined that 94% of the mercury in hair is
methylmercury. Based on these assumptions, they calculated that if the actual hairblood ratio were
200:1, they would have observed a ratio of 290. Based on these and other considerations, Phelps states
that the actual ratio is "probably higher than 200, but less than the observed value of 296." As the
authors point out, one-third of the study population was sampled during the rising phase of seasonal
variation ( and two-thirds or more in the falling phase). Phelps et al. (1980) had assumed that all were
sampled in the falling phase. This fact would tend to result in a lower observed ratio; therefore, the
actual average is likely to be greater than 200. It was concluded by U.S. EPA that a midpoint value of
250 is acceptable for the purpose of estimating average blood levels in the Iraqi population.
Fraction of mercury in diet that is absorbed (A). After administration of radiolabeled
methylmercuric nitrate in water to 3 healthy volunteers, uptake was reported to be >95%. (Aberg et al.
1969). This value is supported by experiments in human volunteers conducted by Miettinen et al.
(1971). These researchers incubated fish liver homogenate with radiolabeled methylmercury nitrate to
produce methylmercury proteinate. The proteinate was then fed to fish for a week; the fish were killed,
cooked and fed to volunteers after confirmation of methylmercury concentration. Mean uptake exceeded
94%. Based upon these experimental results an absorption factor of 0.95 was used in these calculations.
Fraction of the absorbed dose that is found in the blood ffl. There are three reports of the
fraction of absorbed methylmercury dose distributed to blood volume in humans. Kershaw et al., (1980)
reported an average fraction of 0.059 of absorbed dose in total blood volume, based on a study of 5 adult
male subjects who ingested methylmercury-contaminated tuna. In a group of 9 male and 6 female
volunteers who had received 203Hg-methylmercury in fish, approximately 10% of the total mercury body
burden was present in one liter of blood in the first few days after exposure; this dropped to
approximately 5% over the first 100 days (Miettinen et al. 1971) In another study, an average value of
1.14% for the percentage of absorbed dose in one kg of blood was derived from data on subj ects who
consumed a known amount of methylmercury in fish over a 3-month period (Sherlock et al. 1984).
Average daily intake in the study ranged from 43 to 233 (jg/day, and there was a dose-related effect on
percentage of absorbed dose that ranged from 1.03%to 1.26% in one liter of blood. Each of these values
was multiplied by 5 to yield the total amount in the blood compartment, as there are approximately 5
liters of blood in an adult human body (0.01 x 5 = 0.05). The value 0.05 has
6-23
-------
Table 6-2
Available Data on HainBlood Ratio (total Hg)
Reference
Sumarietal., 19691
Soriaetal., 1992
Tejning, 19671
Skerfving, 1974
Haxtonetal., 1979
Tsubaki, 1971b2
Birkeetal., 19722
Den Tonkelaar et al., 1974
Kershaw et al., 1980
Phelpsetal., 1980
Sherlock et al., 1982
Cernichiari et al., 1995
Tsubaki, 1971a'
Hair to
Blood
Ratio
140
218
230
230
250
260
2803
280
292"
296
367
416
370
Number
of
Subjects
50
16
51
60
173
45
12
47
5
339
98
740
-25
Hg Range in
Whole Blood
Gug/L)
5-270
2.4-9.1
4-110
44-550
0.4-26
2-800
4-650
1-40.5
-
1-60
1.1-42.3
0.5-26.7
-
Hair Samples
Hg Range in
Hair (ppm)
1-57
0.15-20
1-30
1-142
0.1-11.3
20-325
1-180
O.5-13.2
-
1-150
0.2-21
-
Length
(mm)
-
-
-
5
20
-
5
-
5
10
24
10
"longer tuft"1
Distance
to
Scalp
-
at scalp
axillary
at scalp
-
-
at scalp
-
at scalp
at scalp
-
at scalp
-
1 As cited in Berglund et al. 1971
2 As cited in WHO, 1976
3 Ratio of methylmercury in hair to methylmercury in blood
4 Based on repeated measurements at different time points (3-8 ratios per individual) of the ratio of 5 mm hair segments to
corresponding 2-week average blood levels (assuming hair growth of 1.1 cm/month).
"--" = Not reported
been used for this parameter in the past by other groups; e.g.., Berglund et al. (1971) and WHO (1990).
A value of 0.05 was used for "f' in the above equation.
Elimination constant (b). Several studies reported clearance half-times for methylmercury from
blood or hair in the range of 35-189 days (Miettinen 1972; Kershaw et al. 1980; Al-Shahristani et al.
1974; Sherlock et al. 1984). Two of these studies included the Iraqi population exposed during the
1971-1972 incident. A value reported in Cox et al. (1989) was derived from the study group which
included the mothers of the infants upon which this risk assessment is based. The average elimination
constant of the 4 studies is 0.014; the average of individual values reported for 20 volunteers ingesting
from 42 to 233 (ig mercury/day in fish for 3 months (Sherlock et al. 1982) is also 0.014. A value of
0.014 days"1 was, thus, used for term "b" in the above equation.
Volume of blood in the body (V). That blood volume is 7% of body weight has been determined
by various experimental methods. There is an increase of 20% to 30% (to about 8.5 to 9%) during
6-24
-------
pregnancy (Best 1961). Specific data for the body weight of Iraqi women were not found. Assuming an
average body weight of 5 8 kg and a blood volume of 9% during pregnancy, a blood volume of 5.22 liters
was derived. In the equation on page 6-19, term "V" was taken to be 5 liters.
Body weight. While the critical endpoint for the RfD is developmental effects in offspring, the
critical dose is calculated using parameters specific to the mothers who ingested the mercury
contaminated grain. Data on body weights of the subjects were not available. A default value of 60 kg
(rounded from 58) for an adult female was used.
Grouping of data
Data used in the U.S. EPA benchmark dose calculation were excerpted from the publication,
Seafood Safety (NAS 1991). The tables of incidence of various clinical effects in children that were
provided in this document readily lent themselves to the benchmark dose modeling approach. The
continuous data for the Iraqi population that were reported in Marsh et al. (1987) were placed in five
dose groups, and incidence rates were provided for delayed onset of walking, delayed onset of talking,
mental symptoms, seizures, neurological scores above 3, and neurological scores above 4 for affected
children. Neurologic scores were determined by clinical evaluation for cranial nerve signs, speech,
involuntary movement, limb tone strength, deep tendon reflexes, plantar responses, coordination,
dexterity, primitive reflexes, sensation, posture, and ability to sit, stand and run. Table 6-3 shows the
input data for the modeling procedure for effects found in children. Incidence data for each of the
adverse effects in children were taken directly from Table 6-11, Seafood Safety (NAS, 1991). The
effects of late walking, late talking, and neurologic scores greater than 3 were also combined for
additional analysis. Table 6-4 shows the incidence data for each of the effects observed in adults as
grouped in Table 6-13 of the Seafood Safety document.
Adjustments for background incidence
As an adjustment for background rates of effects, the benchmark dose estimates for
methylmercury were calculated to estimate the dose associated with "extra risk." Another choice would
have been to calculate based on "additional risk." Additional risk (AR) is defined as the added incidence
of observing an effect above the background rate relative to the entire population of interest, AR = [P(d)-
P(0)]/l. In the additional risk calculation, the background rate is subtracted off, but still applied to the
entire population, including those exhibiting the background effect, thus in a sense "double counting" for
background effects. It can be seen that extra risk (ER) is always mathematically greater than or equal to
additional risk, ER = [P(d)-P(0)]/[l-P(0)], and is thus a more conservative measure of risk (whenever the
background rate is not equal to zero). Conceptually, extra risk is the added incidence of observing an
effect above the background rate relative to the proportion of the population of interest that is not
expected
6-25
-------
Table 6-3
Incidence of Effects in Iraqi Children By Exposure Group2
Effect
Late Walking
Late Talking
Mental Symptoms
Seizures
Neurological Scores >3
Neurological Scores >4
All Endpoints
N
1.37
0
2
1
0
3
0
4
27
10
2
1
0
0
1
1
3
14
52.53
2
3
1
1
4
9
6
13
163.38
3
4
3
2
3
2
8
12
436.60
12
11
4
4
9
6
14
15
' From Table 6-11 of Seafood Safety; Dose is geometric mean in ppm maternal hair.
Table 6-4
Incidence of Effects in Iraqi Adults by Exposure Group"
Effect
Paresthesia
Ataxia
Visual Changes
Dysarthria
Hearing Defects
Deaths
N
50
2
1
0
1
0
0
21
350
i
0
0
i
0
0
19
750
8
2
4
1
1
0
19
1500
10
8
9
4
0
0
17
2500
20
15
14
6
3
0
25
3500
14
17
10
13
6
3
17
4500
7
7
6
6
5
2
7
' From Table 6-13 of Seafood Safety; Dose is geometric mean in ppb blood.
to exhibit such an effect. Extra risk is then more easily interpreted than additional risk, because it applies
the additional risk only to the proportion of the population that is not represented by the background rate.
6-26
-------
Extra risk has been traditionally used in U.S. EPA's cancer risk assessments (Anderson et al. 1983) and is
discussed in detail in a report on the benchmark dose by U.S. EPA's Risk Assessment Forum (U.S. EPA 1995).
Derivation of a benchmark dose
Benchmark dose estimates were made by calculating the 95% lower confidence limits on doses
corresponding to the 1%, 5% and 10% extra risk levels using a quantal Weibull model (K.S.Crump
Division of ICF Kaiser International). The Weibull model was chosen for the benchmark dose
calculations for the methylmercury data as recent research suggests it may be the best model for
developmental toxicity data (Faustman et al., 1994). The form of the quantal Weibull that was used is
the following:
P(d) = AO + (1-/\0)(1-exp[-/\7 * dA2],
where d = dose,
AO = background rate = 0.12468,
Al = slope = 9.47 x 10'3, and
A2 = shape parameter = 1.000.
For each endpoint and for the combined endpoints, the incidence of response was regressed on
the dose. A Chi-squared test of goodness-of-fit was used to test the null hypothesis (Ho) that the
predicted incidence was equal to the observed incidence, so that Ho would be rejected for p-values less
than 0.05.
Results for individual effects and all effects combined for children exposed in utero are given in
Table 6-5; results for adults are given for comparison in Table 6-6. For calculation of the lower bound
on the 10% risk level, AO = 0.12468, Al = 9.470230 x 10'3, A2 = 1.00000. The RfD/RfC Work Group
chose the benchmark (lower bound on the dose for 10% risk) based on modeling of all effects in
children. Recent research (Allen et al. 1994a, b) suggests that the 10% level for the benchmark dose
roughly correlates with a NOAEL for developmental toxicity data. Note that this conclusion was based
on controlled animal studies and on calculation of additional risk. Both the polynomial and Weibull
models place a lower 95% confidence limit on the dose corresponding to a 10% risk level at 11 ppm hair
concentration for methylmercury. The benchmark dose rounded to 11 ppm was used in the calculation of
theRfD.
Dose groupings other than those used in Seafood Safety were also done and benchmark doses run
as above for comparison. Both density-based grouping and uniform concentration intervals were used.
The local density of observations relative to the mercury level in hair was analyzed using a
density estimation algorithm (smooth function in S-PLUS for Windows, Ver. 3.1; S-PLUS Guide to
Statistical and Mathematical Analysis). The function estimates a probability density for the distribution
of a variable by calculating a locally-weighted density of the observations. That is, the function
estimates the probability that an observation will be near a specific value based on how the actual values
are clustered. In this case, the function was used to estimate the probability density for an observation in
the neighborhood of any given maternal hair mercury concentration. The density plot is shown in Figure
6-1. The peaks represent relatively greater numbers of data points than the troughs in the vicinity of the
associated hair mercury concentrations.
6-27
-------
Table 6-5
Methylmercury Benchmark Dose Estimates (ppm hair)
Maximum Likelihood Estimates and 95% Lower Confidence Limits from Weibull Model
Incidence of Effects in Children (Marsh et al. 1987)
Effect
Late Walking
Late Talking
Mental Symptoms
Seizures
Neuro Score >3
Neuro Score >4
All Endpoints
0.01
MLE
3.3
4.7
12.0
11.8
5.6
8.1
1.6
95% CL
2.1
2.4
6.4
6.7
3.3
4.6
1.1
0.05
MLE
16.7
22.1
61.0
60.4
28.8
41.4
8.3
95% CL
10.9
12.3
32.8
34.3
17.0
23.7
5.4
0.10
MLE
34.3
43.8
125.4
124.2
59.1
84.9
17.1
95% CL
22.4
25.3
67.5
70.5
34.9
48.7
11.1
G-O-F3
P-Value
0.16
0.79
0.63
0.86
0.58
0.48
0.94
' Goodness-of-fit p-value for testing the null hypothesis, Ho: Predicted Incidence = Observed Incidence.
Table 6-6
Methylmercury Benchmark Dose Estimates (ppb blood)
Maximum Likelihood Estimates and 95% Lower Confidence Limits from Weibull Model
Incidence of Effects in Adults (Bakir et al. 1987)
Effect
Paresthesia
Ataxia
Visual Changes
Dysarthria
Hearing Defects
Deaths
0.01
MLE
45.3
330.4
64.1
728.9
1462.9
2226.3
95% CL
14.3
140.8
25.4
235.9
535.2
1106.8
0.05
MLE
169.0
652.9
249.1
1265.4
2137.5
3007.2
95% CL
73.2
369.7
129.9
621.8
1202.8
2167.3
0.10
MLE
302.2
882.0
453.6
1614.4
2527.1
3434.2
95% CL
150.5
564.7
266.9
949.8
1705.6
2797.0
G-O-F3
P-Value
0.36
0.22
0.26
0.41
0.53
0.83
* Goodness-of-fit p-value for testing the null hypothesis, Ho: Predicted Incidence = Observed Incidence.
6-28
-------
Figure 6-1
Density of Data Points Relative to Mercury
Concentration in Hair for Iraqi Cohort Data
5 10 50 100
ppm Hg in hair
500
The density distribution is characterized by four distinct peaks. Exposure dose groups were
defined as trough-to-trough intervals with the peak values taken as the nominal value for each interval.
The nominal dose-group value, concentration ranges, and incidence of combined developmental effects
are given in Table 6-7. A benchmark dose was calculated from the incidence of all effects as grouped in
Table 6-7. The lower 95% confidence interval on the benchmark dose for the 10% response is 13 ppm,
compared to the 11 ppm value used as the basis for the RfD.
Another alternative dose grouping approach was to divide the entire exposure range into four
equal log-dose intervals. The geometric midpoint of each interval was taken as the nominal value for the
interval. The nominal dose-group value, concentration ranges, and incidence of combined
developmental effects are given in Table 6-8. The benchmark calculated as the lower bound on the 10%
incidence for all effects is 10.3 ppm, compared to the 11 ppm used for the RfD.
Table 6-7
Density-Based Dose Groupings
Nominal Dose (ppm)
1.18
10.6
78.8
381
Dose Range (ppm)
1 -4
5-28
29- 156
157 - 674
Incidence
5/27
3/16
10/17
18/21
6-29
-------
Table 6-8
Uniform Dose Groupings
Nominal Dose (ppm)
2.25
11.5
58.6
299
Dose Range (ppm)
1 -5
6-25
26- 132
133 - 674
Incidence
5/28
3/14
9/17
19/22
Two other analyses were done. In the first, data on males and females were grouped as
published in Seafood Safety, and the Weibull model was as for the data in Table 6-9. The lower 95%
confidence interval on the benchmark dose for the 10% response for males only was lOppm and for
females only, 1 Ippm. The last analysis consisted of fitting all data (for males and females ) on all
endpoints in Table 6-9 without grouping . When the model was restricted such that the Weibull power
would drop below 1, the lower 95% bound on the dose for the 10% response was 1 Ippm. For an
unrestricted model the benchmark dose was. Table 6-9 lists all equivalent benchmarks (lower 95%
bounds on a 10% effect level for all measured endpoints) calculated by U.S.EPA on the data of Marsh et
al. (1987).
Table 6-9
Benchmark Dosed Calculated on Data from Marsh et al. (1987)
Data Grouping
Grouping in Seafood Safety
Density-based grouping
Uniform concentration intervals
Males only (Seafood Safety groups)
Females only (Seafood Safety groups)
Individual data points (restricted model)
Benchmark Dose (ppm maternal hair)
11
13
10
10
10
11
Uncertainty and modifying factors
A composite uncertainty factor of 10 was used. This uncertainty factor was applied for
variability in the human population, in particular the wide variation in biological half-life of
methylmercury and the variation that occurs in the hair to blood ratio for mercury. In addition, the factor
accounts for lack of a two-generation reproductive study and lack of data for possible chronic
manifestations of the adult effects (e.g., pare sthesia that was observed during gestation). The default
value of one was used for the modifying factor.
6-30
-------
Calculation of the oral RfD for methylmercury
In this instance the RfD was calculated using the following equation:
Benchmark Dose
RfD =
UF x MF
1.1 ug/kg-day
10
= 1 x 10 4 mglkg-day
Confidence in the oral RfD for methylmercury
The principal study (Marsh et al. 1987) is a detailed report of human exposures with quantitation
of methylmercury by analysis of specimens from affected mother-child pairs. A strength of this study is
that the quantitative data are from the affected population and quantitation is based upon biological
specimens obtained from affected individuals. A threshold was not easily defined; extended application
of modeling techniques were needed to define the lower end of the dose-response curve. This may
indicate high variability of response to methylmercury in the human mother-child pairs or
misclassification of assigning pairs to the cohort. Confidence in the supporting data base and confidence
in the RfD were considered medium by the RfD/RfC Work Group.
An analysis of uncertainty in an RfD based on the Iraqi data is found in Appendix D of this
volume. Discussions of areas of uncertainty can also be found in Volume VII, Risk Characterization.
Choice of Benchmark or NOAEL as the basis for the RfD
Estimates of threshold levels for neurotoxicity have been performed by WHO (1990) using data
from the Niigata episode and the Iraqi poisoning. In the exposures in Japan, hair levels associated with
thresholds for neurotoxicity were estimated by WHO to be approximately 100 ppm. Estimates of
threshold levels associated with paresthesia in the Iraqi episode indicate that the threshold level for
paresthesia is approximately 25 to 40 mg (total body burden). This corresponds to blood levels of
approximately 250 to 400 //g/L and hair levels of approximately 50 //g/g. Thresholds (total body
burden) estimated by Bakir et al. (1973) for other neurotoxic signs were 55 mg for ataxia, 90 mg for
dysarthria (difficulty with speech), 170 mg for deafness, and 200 mg for death.
A number of additional studies of human populations generally support the dose range of the
benchmark dose level for perinatal effects. The designs for these studies as well as summaries of results
are given in section 3.3.1.1. A few of the studies have data suitable for calculation of benchmark doses
as was done for the results of Marsh et al. (1987). Table 6-10 compares one published benchmark dose
(on the New Zealand population) as well as several others calculated for the Mercury Study Report to
Congress. Table 6-12 provides a compilation of NOAELs and LOAELs determined from inspection of
human data sets as well as other published comparable measures.
A recent analysis of the Kjellstrom (Kjellstrom et al. 1986a, b,1989) data was published by
Gearhart et al. (1995). In this analysis the authors used a PBPK model which incorporated a fetal
6-31
-------
compartment. They calculated a benchmark dose on all 28 tests included in the initial study design by
Kjellstrom; this was done assuming values of 1 and 5% for background deficiency in test scores. The
range of benchmark doses calculated was 10 to 31 ppm maternal hair mercury. The authors' preferred
benchmark was 17 ppm, for an estimated background incidence of 5% and the lower bound on the 10%
risk level (Table 6-10).
Table 6-10
Summary of Benchmark Doses (BMD) Estimated for Methylmercury
Study site
Iraq
Iraq
Seychelles
New Zealand
Duration of
Exposure
short term
(~ 3 mo)
short term
(~ 3 mo)
long term
long term
na
81
81
789
237
Endpoint
all developmental
effects reported in
Marsh etal.(1987)
all effects except
late walking, late
talking
DDST, abnormal
plus questionable
scores
all measures in 28
tests
Benchmark
(ppm maternal
hair mercury)
11
15
16
17
Reference13
Marsh etal( 1987);
this document
Marsh etal( 1987);
this document
Meyers etal. (1995);
this document
Kjellstrom et al.
(1986a,b, 1989);
Gearhart etal. (1995)
a n=number of subjects in analysis
b first reference is to source of data; second is to source of BMD calculation.
Data have recently been published from the pilot (or cross sectional) study and testing of
children from up to age 29 months in the main or prospective study in the Seychelles (Myers et al.
1995a,b,c,d; Davidson et al. 1995). The range of maternal hair mercury levels for the pilot study was 0.6
to 36.4 ppm; the range for the main study was 0.5 to 26.7 ppm. The Among tests administered to the 60
month old children in the cross-sectional study was the Denver Developmental Screen Test (DDST)
(Myers et al. 1995b) which was included largely to provide a point of comparison with other population
studies (e.g. Kjellstrom et al. 1986a, b, 1989). The frequency of abnormal plus questionable scores was
analyzed by multiple logistic regression which showed an association for increased frequency of non-
normal scores with increasing maternal hair mercury. Results for the abnormal plus questionable scores
(Table 6-11) were used to calculate a benchmark dose as a lower 95% limit on a 10% effect level. The
resulting estimate is 16 ppm maternal hair. The study authors have commented that the main study,
which was done under more controlled circumstances, did not show any relationship between DDST-R
results and maternal hair mercury.
6-32
-------
Table 6-11
Results of Revised Denver Developmental Screening Test (DDST)
Administered to Seychellois Children in Cross-sectional Study"
Result
normal
abnormal
questionable
abnormal plus
questionable
Maternal Hair Mercury, ppm
0-3
114
(92.7)
0
9
9
(7.3)
>3-6
209
(92.5)
1
16
17
(7.5)
>6-9
169
(91.4)
1
15
16
(8.7)
>9-12
107
(93.0)
0
8
8
(7.0)
>12
122
(87.1)
1
17
18
(12.9)
total
721
(91.4)
3
65
68
(8/6)
a Data from Myers et al. (1995b)
In a recent publication, Gearhart et al. (1995) proposed a RfD in the range of 0.8 to 2.5 (jg/kg-
day based on their analysis of effects in a population of children in New Zealand. This population was
assumed to be exposed in utero to methylmercury as a consequence of high fish consumption by their
mothers. Gearhart et al. (1995) estimated that a maternal intake of methylmercury in the range of 0.8 to
2.5 (ig/kg-day corresponded to a NOAEL for developmental effects. These results support the U.S. EPA
estimate of maternal intake of 1 (ig/kg-day methylmercury corresponding to a benchmark dose of 11 ppm
mercury in hair for developmental effects, prior to applying an uncertainty factor. The primary area of
disagreement between Gearhart et al. (1995) and the U.S. EPA is in the use of an uncertainty factor. The
authors felt that no uncertainty or modifying factors were needed as the NOAEL was calculated on
effects in a sensitive subpopulation. U.S. EPA applied a 10-fold uncertainty factor to account for
interindividual variation in the human population (particularly in hair to blood mercury ratio) and for
lack of certain types of data.
The NOAEL was implicitly defined by Gearhart et al. (1995) as the lower 95% confidence
interval on the dose associated with a 10% change in the test scores from the New Zealand study
(Kjellstrom et al. 1989). That is, the benchmark dose calculation was performed on continuous variables
by contrast to the binary variables from the Iraqi study (Marsh et al. 1987) used by the U.S. EPA. The
qualitative equivalence of the two kinds of benchmark doses has not been established; both presumably
represent a minimum risk level as does a NOAEL.
Using a hockey stick parametric dose response analysis of the data on delayed walking in the
Iraqi children, Cox et al. (1989) concluded that the "best statistical estimate" of the threshold for health
effects was 7.3 ppm mercury in hair with a 95% range of uncertainty between 0 and 14. A more recent
analysis of the same data (Cox et al. 1995) focuses on the importance of four data points termed
influential for the estimation of the population threshold. The newer analysis indicates that when a
background response rate of 4% is assumed, the threshold estimate is 9 ppm maternal hair (Table 6-12).
The authors indicate that dose-response analyses based on the "late walking" endpoint are unreliable
because it relies on four influential observations in the data set from Marsh et al. (1978). The data points
in question are the only responders below 150 ppm (mercury in hair). In particular, Cox et al. (1995)
state that the four observations are isolated from the remainder of the responders and would be expected
to have considerable influence on threshold estimate. This conclusion is based on a visual interpretation
of a plot of the data (Figure 2 in Cox et al. 1995). Based on visual inspection of the same figure, an
argument could
6-33
-------
Table 6-12
Estimates of No Observed Adverse Effect Levels (NOAELs) and
Lowest Observed Adverse Effects Levels (LOAELs) from Human Studies
Study Site
Iraq
Iraq
Iraq
Iraq
Iraq
New
Zealand
Canada
(Cree
population)
Amazon
Niigata
Exposure
in utero
and post
partum
/short term
in utero
/short term
in utero
/short term
in utero
/short term
in utero
/short term
in utero
/long term
in utero
/long term
adult/long
term
adult
women/
long term
n
84
81
81
81
81
237
247
29
430
Endpoint
All developmental
All developmental
delayed walking
delayed walking
delayed walking
IQ tests (WISC-R,
TOLD)
deep tendon reflex
visual
discrimination
Minimata disease
Estimate
type3
NOAEL
NOAEL
LOAEL
best
estimate
of
threshold
best
estimate
of
threshold
threshold
LOAEL
LOAEL
threshold
Estimate
inppm
mercury
in hair
13
7-10
14
7.3
9
5-15
2-15
20
25
Reference13
Marsh et al.
(1981); Marsh et
al. (1981)
Marsh et al.
(1987); this
document
Marsh et al.
(1987); this
document
Marsh et al.
(1987); Cox etal.
(1989).
Marsh et al.
(1987); Cox etal.
(1995)
Kjellstrom et al.
(1986a,b, 1989);
Lipfert (1994)
McKeowin -
Eyssen(1983);
Lipfert (1994).
Lebel etal. (1995);
this document
Kinjo etal. (1995);
this document
6-34
-------
Table 6-12 (continued)
Estimates of No Observed Adverse Effect Levels (NOAELs) and
Lowest Observed Adverse Effects Levels (LOAELs) from Human Studies
Study Site
Iraq, Cree,
New
Zealand
Seychelles
(pilot, 66
month old
children)
Seychelles
(main, 29
month old
children)
Seychelles
(main, 29
month old
children)
Faroes
Exposure
in utero
/short and
long term
in utero
/long term
in utero
/long term
in utero
/long term
in utero
/long term
n
217
736
736
917
Endpoint
multiple effects
Preschool
Language Scale
(auditory
comprehension)
BSID
Activity level,
boys
Neuropsychologica
1 tests
Estimate
type3
LOAEL
LOAEL
NOAEL
LOAEL
LOAEL
Estimate
inppm
mercury
in hair
10-20
12-36
12-26.7
12-26.7
<10
Reference13
Marsh et al.
(1981), McKeowin
-Eyssen (1983),
Kjellstrom et al
(1986a, b, 1989);
Hoover etal 1997.
Myers et al.
(1995a);this
document
Davidson et al.
1995; this
document
Davidson et al.
1995; this
document
Grandjean et al.
(1997); Grandjean
etal. (1997)
a Threshold and best estimate defined by author of estimate.
b First reference is to source of data; second is to source of analysis.
be made that the separation is not that marked considering the first eight responders. No quantitative
sensitivity analysis was performed to investigate the effect of removing one or more of these data points.
Cox et al. (1995) point out that if the four points are assumed to represent background, then the threshold
for late walking would be greater than 100 ppm. It would seem unlikely, however, that these
observations represent background given that no responses were observed in the 37 individuals with
lower levels of exposure. It should be noted that the U.S. EPA benchmark dose was done on incidence
of all effects, rather than on late walking only.
Crump et al. (1995) reanalyzed data from the Iraqi methylmercury poisoning episode presented
in Marsh et al. (1987). In their analysis, Crump et al. (1995) reported that the statistical upper limit of
the threshold could be as high as 255 ppm. Furthermore, the maximum likelihood estimate of the
threshold using a different parametric model was presented as virtually zero. These and other analyses
demonstrated that threshold estimates based on parametric models exhibit high statistical variability and
model dependency, and are highly sensitive to the precise definition of an abnormal response.
6-35
-------
Using a statistical analysis for trend that does not require grouping of the data, Crump et al.
(1995) demonstrated that the association between health effects and methylmercury concentrations in
hair is statistically significant at mercury concentrations in excess of about 80 ppm. In addition, Crump
et al. calculated benchmark doses by applying dose-response models to each of the three endpoints: late
walking, late talking and neurological score. Unlike the benchmark calculations made by the U.S. EPA
(1994), these analyses did not involve grouping of the data into discrete dose groups, nor did they require
dichotomizing continuous responses like age first walked into "late walking" or "no walking." Their
calculation of the 95% lower bounds on the hair concentration corresponding to an additional risk of
10% ranged from 54 ppm to 274 ppm mercury in hair. Crump et al. (1995) concluded that the trend
analyses and benchmark analyses provided a sounder basis for determining RfDs than the type of hockey
stick analysis presented by Cox et al. (1989). They felt that the acute nature of the exposures, as well as
other difficulties with the Iraqi data, present limitations in the use of these data for a chronic RfD for
methylmercury.
The Cox et al. (1995) and Crump et al. (1995) analyses deal primarily with one endpoint, late
walking. This appears to be the most sensitive of the endpoints described in Marsh et al. (1978). Both
Cox et al. and Crump et al., as well as the U.S. EPA analysis in Appendix D of Volume V, show
considerable uncertainty in thresholds estimated from the data on late walking. The peculiar nature of the
uncertainty, in this case, makes it difficult to distinguish between 7 ppm maternal hair mercury and 114
ppm as a best (maximum likelihood) estimate for the threshold. Cox et al. (1995) attribute this bimodal
uncertainty to four influential observations between 14 ppm and 60 ppm isolated from the remainder of
the responders beginning at 154 ppm; they do not present arguments, other than visual, for censoring
these data. Crump et al. (1995) show that changing the definition of late walking from greater than 18
months to 18 months or greater eliminates the bimodal uncertainty with a best threshold estimate of 230
ppm. The implication in both analyses is that the background incidence of late walking, as reported in
other studies, is not consistent with the lower thresholds. While this is true, the use of historical controls
for this analysis may not be appropriate, given the relatively large number of observations at low
exposure levels in the Iraqi cohort; 33% of the observations were at hair mercury concentrations
considered to be background levels (3 ppm or less).
Late walking, as assessed in the exposed Iraqi population (Marsh et al. 1978), is almost certainly
a valid indicator of methylmercury toxicity but may be unreliable as the sole basis for detailed dose-
response analysis. The primary reason for this may be the uncertainty in maternal recall for both birth
date and date of first walking. The uncertainty in this particular case could be quite large, given the lack
of recorded information. The primary impact of this kind of uncertainty would be on the response
classification of individuals at the upper bound of normal (18 months for first walking) and at the lower
bound of abnormal. The lowest abnormal first walking time presented in Marsh et al. (1978) was 20
months. The impact of assuming uncertainty in the classification of the observations in these two groups
is large given the large number of observations in the two groups (19 data points at 18 months and 8 data
points at 20 months). The analysis in Appendix D to Volume V of the Mercury Study Report to
Congress shows that thresholds estimated for late walking are unstable when classification uncertainty is
considered. The same kind of subjective uncertainty is applicable to the late talking endpoint, as well.
The thresholds for late talking, however, are much more stable, statistically, as there are fewer
observations that are near the normal/abnormal threshold value of 24 months.
McKeown-Eyssen et al. (1983) observed a positive association between abnormal tendon
reflexes in boys and increasing maternal hair mercury. This was a study of 234 Cree children between
the ages of 12 to 30 months residing in northern Quebec communities. Average maternal hair mercury
for boys and girls was 6 ppm; the maximum was 24 ppm and 6% of the population had hair mercury
levels in excess of 20ppm. A LOAEL of between 2-15 ppm maternal hair mercury was calculated by
Lipfert et al. (1996). These authors felt that the McKeown-Eyssen et al. (1983) study provides only
marginal support for a LOAEL above 10 ppm; they concluded that part of the significance of the adverse
effects rests on the in (in their opinion) inappropriate inclusion of two observations of increased muscle
6-36
-------
tone in the 2-15 ppm group. In a later paper, Hoover et al. (1997) reported that a LOAEL range of 10-20
ppm from this study was warranted.
Marsh et al. (1995) have published results of a study conducted between 1981 and 1984 in
residents of coastal communities of Peru. The prospective study was of 131 child-mother pairs; testing
for potential effects of fetal methylmercury exposure was patterned after the study of children exposed in
utero in Iraq. Peak maternal hair methylmercury ranged between 1.2 to 30 ppm with a geometric mean
of 8.3 ppm. These authors showed no effects of methylmercury on measures similar to those performed
on the Iraqi children (including time of first walking and talking). A NOAEL (in the absence of a
LOAEL) from this study would be 30 ppm maternal hair mercury (Table 6-12).
Lebel et al. (1996) published a study of 29 young adults (less than 35 years of age) who resided
in villages on the Tapajos River, about 200 Km downstream of Amazonian gold-mining sites. Hair
mercury ranged from 5.6 ppm to 38.4 ppm; methymercury constituted between 72.2% and 93.3% of the
total. The authors found that decreases in several measures of visual acuity (color discrimination loss,
contrast sensitivity and visual field reduction) were related to increased hair mercury. The authors note
that constriction of the visual field has been reported in other instances of mercury intoxication.
Inspection of the data presented as charts indicates a LOAEL of about 20ppm.
Kinjo et al.(1995) have published an analysis of the relationship between hair mercury
concentration and the incidence of Minamata disease in Niigata, Japan. Hair samples were collected in
1965 and analyzed by a colorimetric procedure. The population used in the statistical analyses consisted
of 147 males and 430 females. The authors felt that the colorimetric method used to determine hair
mercury was not reliable at levels below 20 ppm. Data on individuals with hair mercury less than 20
ppm were excluded from the analysis; these data, however were included in Figure 3 of the paper.
Minamata disease was defined by criteria established for receipt of compensation by the Japanese
government medical committee (Tamashiro et al 1985). Both hockey stick and logit models were
applied to obtain a range of threshold values of 24.7 to 49.3 for females; inspection of the graphed data
in Figure 3 indicates a threshold for Minamata disease in females between 20 and 30 ppm. For males the
threshold estimates are between 43 and 48 ppm.
A pilot study (essentially a cross-sectional study) of developmental effects in a seafood
consuming population in the Seychelles Islands focused on all children born between February 1989 and
February 1990. A total of 804 mother-infant pairs were observed to have maternal hair mercury in the
range of 0.6 to 36.4 ppm with a median of 6.6 ppm (Myers et al. 1995b). Children were tested once
between the ages of 5 and 109 weeks of age. An association was found for in utero mercury exposure
and DDST-R abnormal plus questionable scores combined (Myers et al 1995b). A benchmark dose was
derived on this data set as had been done on the Kjellstrom et al. (1986a, b, 1989) data. The BMD (95%
lower bound on the 10% effect level) is 16 ppm maternal hair mercury (Table 6-10).
A subset of the pilot cohort (217 children) was evaluated at 66 months using the McCarthy
Scales of Children's Abilities, the Preschool Language Scale, and tests from the Woodcock-Johnson
Tests of Achievement that were appropriate to the children's age (Myers et al. 1995a). The median
maternal hair mercury for this group was 7.11 ppm. Mercury exposure (measured as maternal hair
mercury) was negatively associated with four endpoints: the McCarthy General Cognitive Index and
Perceptual Performance subscale; and the Preschool Language Scale Total Language and Auditory
Comprehension subscale. When statistically determined outliers and points considered to be influential
were removed from the analyses, statistical significance of the association remained only for auditory
comprehension. If a decrease in the scores on the auditory comprehension test is considered an adverse
effect, a LOAEL from this study would be in the range of 12-36 ppm maternal hair mercury.
The main study was designed to be prospective; children were evaluated at 6.5, 19, 29 and 66
months of age (data on the 66 month old children have not yet been published) (Shamlaye et al, 1995) .
6-37
-------
In the group evaluated at 6.5 months, median maternal hair mercury was 5.9 ppm with a range of 0.5 ppm
to 26.7 ppm. No association with maternal hair mercury was found for any endpoint tested. (Myers et al.
1995c). Evaluations at 19 and 26 months were done on groups of 738 and 736 individuals, respectively
(Davidson et al. 1995). Median maternal hair mercury was 5.9 ppm, and the range was 0.5 to 26.7 ppm.
Children were evaluated with the BSID at 19 months of age. No effects of mercury exposure were seen
on outcome of tests administered at 19 months. At 29 months, children were administered the BSID as
well as the Bayley Infant Behavior Record. At 29 months there was an association between mercury
exposure and decreased activity level in male children only. If the decrease in activity level is considered
an adverse effect, a LOAEL for the study would be in the range of 12-26.7 ppm maternal hair mercury
(Table 6-12). If this study is considered non-positive a NOAEL would be in the range of 12-26.7 ppm
maternal hair mercury.
The overall conclusion of the studies published to date is that it is yet unclear whether an
association exists between low level mercury exposure and neurologic deficits in children. The study
shows a close correlation between maternal hair mercury and neonatal levels of mercury in brain tissue
(Cernichiari et al. 1995). The authors cautioned in several papers that subtle neurologic and
neurobehavioral effects are more likely to be detected in older rather than younger children. The overall
conclusion of the authors is that their results require careful interpretation, and that an association
between relatively low level mercury exposure in utero and neurologic deficits has not been
demonstrated.
In 1986 a large study was initiated in the Faroe Islands on neurologic developmental effects of
methylmercury and PCB exposure in utero (Grandjean et al. 1997). Subjects were a group of 917
children born between 1986 and 1987 and examined at about 7 years of age. Mercury was measured in
maternal hair and cord blood, and a subset of cords was evaluated for PCBs. The median maternal hair
mercury concentration was 4.5 ppm, and 13% were greater than 10 ppm (Grandjean et al 1992a). The
geometric mean cord blood mercury concentration for the group of children who completed the
evaluation was 22.8 ppm. Significant negative associations were seen for several neuropsychological
tests. Even after inclusion of covariates with uncertain influence on these tests, multiple regression
analysis indicated that 9/20 measures showed mercury related decrements (p< 0.05, one tailed).
Application of a Peters-Belson adjustment resulted in significant mercury associations for 11/20
measures. PCB determinations were done on a total of 436 cords, and PCB exposure was included as a
covariate in the regression analyses. This had an effect only on the regression for the Boston Naming
Test. After exclusion of children with maternal hair mercury concentrations above 10 ppm, these
associations remained almost unchanged. The authors concluded that in utero exposure to
methylmercury at levels below 10 ppm maternal hair mercury affects several domains of cerebral
function; in particular, attention, language and memory (Grandjean et al. 1997). Table 6.12 thus lists a
LOAEL for this study as less than 10 ppm maternal hair.
With the exception of the estimates for threshold for Minimata disease, the NOAEL, BMD and
threshold estimates (nine altogether) fall in a narrow range of 5-26.7 ppm maternal hair mercury. This
is a large degree of overlap with the 1 Ippm BMD and provides a great deal of support for this estimate
based on the Iraqi data.
Chronic rodent (e.g. Bornhausen et al. 1980) and nonhuman primate studies (e.g. Burbacher et al.
1984; Gunderson et al. 1986; Rice et al. 1989a,b) provide data to support estimated NOAELs and
LOAELs for developmental end points. The endpoints measured in these animal studies are relevant to
the types of toxicity which have been reported in children and they have been induced by dosing
protocols that are relevant to human exposures. Experiments in nonhuman primates have identified
adverse effects of methylmercury exposure in these areas: sensory (visual, somatosensory, auditory),
cognitive (learning under concurrent schedules, recognition effaces), social play, and
schedule-controlled operant behavior. The sensory, cognitive, and motor deficits appear reliable over a
consistent range of doses in nonhuman primates exposed to methylmercury during development.
6-38
-------
Table 6-13 lists NOAELs and LOAELs from animal studies. As the RfD for methylmercury is
based on effects in children who were exposed in utero, it is particularly useful to consider
developmental effects in animals exposed to methylmercury. NOAELs and LOAELs from selected
developmental studies of methylmercury in animals can be found in Table 6-14; a more complete
compilation is in section 3.3.3.8.
6-39
-------
Table 6-13
Estimates of NOAELs and LOAELs from Animal Studies
Species/
Strain/
No. per
Sex per
Group
Rat/Wistar/
10 F
Rat/Wistar/
50 F, 50 M
Swiss
origin
Mouse
M
Cat/Breed
NS/15-16
both sexes
Cat/Breed
NS/8-10
NS
Monkey/
Macaca
fasciculari
5/1-2 both
sexes
Monkey/
Macaca
artoides,
Macaca
nemestrina
12 both
sexes
Exposure
Duration
0-12 or
12-20 d,
Ix/d
up to 26
mo
28 wk
11 mo
2yr
7d/wk
36-132 d
Ix/d
90-270 d
1 x/wk
NOAEL,
LOAEL
(mg/kg-day)
NOAEL =2;
LOAEL=4
NOAEL =
0.05;
LOAEL=0.25
LOAEL = 1.9
LOAEL =
0.015
NOAEL =
0.020; LOAEL
= 0.046
NOAEL =
0.02; LOAEL
=, 0.03
NOAEL= 0.4;
LOAEL = 0.5
Effects
Hindlimb crossing after 0-12 days
Ruffled fur, loss of balance, hindlimb
crossing, paralysis after 6 mo (males
more affected); demyelination of dorsal
nerve roots and damage in teased
sciatic nerves at 0.25
Ataxia; degenerative changes of
Purkinje cells; granule cell loss in
cerebellum
Degeneration of cerebellum and
cerebral cortex; necrosis of dorsal root
ganglia of kittens fed mercury-
contaminated tuna
Impaired hopping reaction; decreased
pain sensitivity; degeneration of dorsal
root ganglia
Atrophy of neurons in calcarine cortex;
focal degeneration in sural nerves
Tremor; visual impairment
Reference
Inouye and
Murakami
1975
Munro et al.
1980
MacDonald
and
Harbison
1977
Chang et al.
1974
Charbonneau
etal. 1976
Sato and
Ikuta 1975
Evans et al.
1977
6-40
-------
Table 6-13 (continued)
Estimates of NOAELs and LOAELs from Animal Studies
Species/
Strain/
No. per
Sex per
Group
Monkey/
Macaca
fasciculari
5/5
exposed, 2
control
(sexNS)
Monkey/
Macaca
fasciculari
5/7-8 F
Monkey/
Macaca
fasciculari
5/4 M, 1 F
exposed, 1
M, 2F
controls
Exposure
Duration
~4 yr
7d/wk
Ix/d
~3yr
Ix/d
6.5-7 yr
7d/wk
Ix/d
NOAEL,
LOAEL
(mg/kg-day)
LOAEL= 0.05
LOAEL = 0.04
LOAEL = 0.05
Effects
Spatial visual impairment
Slight tremor; motor incoordination;
blindness ; time to onset was 177-395 d
Six years after end of dosing (follow-up
study to Rice and Gilbert 1982):
decreased fine motor performance;
diminished touch and pinprick
sensitivity; impaired high frequency
hearing (p<0.05)
Reference
Rice and
Gilbert 1982
Burbacher et
al. 1988
Rice 1989b;
Rice and
Gilbert 1992
6-41
-------
Table 6-14
NOAELs and LOAELs for Developmental Toxicity of Methylmercury in Animals
Species/
Strain/
No. per Sex per
Group
Rat/Holtzman
/5F
Rat/Charles
River CD/
20 F
Rat/Long-
Evans/4
exposed, 6
control
Rat/Wistar-
Neuherberg/
No. F. NS
Rat/Wistar/10
F
Rat/Sprague-
Dawley/
15-19 F
Exposure
Duration
during
gestation,
during
lactation,
or
postnatal
days 21-
30
47 d
prior to
and
during
gestation
Once
Gd7
4d
Gd6-9
4d
Ix/d
Gd6-9
4d
Ix/d
Gd6-9
NOAEL, LOAEL
(mg/kg-day)
LOAEL=2.5
NOAEL=0.7,
LOAEL=1.4
LOAEL=4
LOAEL=0.04
NOAEL=0.004,
LOAEL=0.008
NOAEL=1.6,
LOAEL=4.8
Effects
Decreased visual evoked potential
latencies for peaks in 30-day old
pups exposed during gestation,
during lactation, or during postnatal
days 21-30.
Ultrastructural changes, dose-
related decrease in biochemical
activity in mitochondria of fetal
hepatocytes
Increased P1-N1 amplitudes and
decreased P2 and N2 latencies of
cortically visual evoked potential
Impaired ability to perform operant
conditioning procedures (number
of responses on lever required in
specified period of time) (
Reduction in behavioral
performance in offspring of treated
mice following operant
conditioning
Delayed vaginal patency, delayed
surface righting, retarded
swimming development, lower
activity, impaired complex water
maze performance. Increased
mortality of pups at 1-21 days of
age
Reference
Zenick 1976
Fowler and
Woods 1977
Dyer et al.
1978
Musch et al.
1978
Bornhausen et
al. 1980
Vorhees 1985
6-42
-------
Table 6-14 (continued)
NOAELs and LOAELs from Developmental Toxicity of Methylmercury in Animals
Species/
Strain/
No. per Sex per
Group
Rat/Wistar/
38 M, 38 F
Rat/HAN-
Wistar/10 F
Rat/Wistar/ 16
F
Mouse/SvSl/
No. FNS
Monkey/
Macaca
fascicularis/9
F exposed, 8
F control
Monkey/
Macaca
fascicularisl
12 F exposed,
13 F control
Exposure
Duration
during
gestation
and
lactation
13 days
prior to
mating
until
post-
natal day
21
2wk
prior to
mating
through
weaning
Once
Gd 7 or 9
(i.p.)
approx.
1-3 yr 1
x/d prior
to mating
through
gestation
approx. 4
mo to 2
yr Ix/d
prior to
mating
through
gestation
NOAEL, LOAEL
(mg/kg-day)
NOAEL=0.2,
LOAEL=0.6
LOAEL=0.2
LOAEL=0.08
LOAEL=0.16
LOAEL=0.04
LOAEL=0.04
Effects
Increase in response latency in
male (p<0.05) and female pups
(p<0.01) and in passiveness
(p<0.05) in visual discrimination
reversal task
Delayed sexual maturity (vaginal
opening and testes descent)
Impaired tactile-kinesthetic
function (p<0.05)
Impaired swimming ability and
open-field behavior (p<0.05) in 30-
day old pups.
Impaired visual recognition
memory (data pooled from both
groups of infants of exposed
mothers) compared to unexposed
controls; test performed at 50-60
days of age.
Decrease in social play behavior
and concomitant increase in
nonsocial passive behavior
compared to unexposed controls;
tests performed at 2 weeks to
8 months of age.
Reference
Schreiner et al.
1986
Suter and
Schon 1986
Eisner 1991
Spyker et al.
1972
Gunderson et
al. 1988
Gunderson et
al. 1988
6-43
-------
Table 6-14 (continued)
NOAELs and LOAELs from Developmental Toxicity of Methylmercury in Animals
Species/
Strain/
No. per Sex per
Group
Monkey/
Macaca
fascicularis/5
mothers
Monkey/
Macaca
fascicularis/4
M, IF
exposed, 1 M,
2 F controls
Monkey/
Macaca
fascicularisl
23 F
Monkey/Saim
iri sciureus/3
F
Exposure
Duration
4-4.5 yr
Ix/din
utero and
postnatal
iy
6.5-7 yr
7d/wk
Ix/d
unspecifi
ed period
prior to
mating
through
gestation
week 1 1
or 14.5
until
parturitio
n
NOAEL, LOAEL
(mg/kg-day)
LOAEL=0.01
LOAEL=0.05
NOAEL=0.08
LOAEL=0.7 to
0.9 ppm
methylmercury
in maternal
blood
Effects
Spatial visual impairment
Six years after end of dosing
(follow-up study to Rice and
Gilbert 1982); decreased fine motor
performance; diminished touch and
pinprick sensitivity; impaired high
frequency hearing (p<0.05).
No effect on spatial memory of
adult offspring of animals treated
with methylmercury hydroxide
(data pooled from 24 animals, all
treated groups).
Monkeys exposed in utero tested
(on learned lever pulling activity)
at ages 5-6 yr. decreased
sensitivity to degrees in
reinforcement; change in
reinforcement degree resulted in
either no behavior change or slow
change by comparison to controls.
Reference
Rice and
Gilbert 1990
Rice 1989b;
Rice and
Gilbert 1992
Gilbert et al.
1993
Newland et al.
1994
At least three long term studies of non-human primates exposed in utero or as infants have been
undertaken. Description of the study populations (adapted from Rice 1996) is given in Table 6-15.
Postnatal, pre-pubescent exposure of the primates approximates the exposures experienced by
human children in both the short term (e.g. Iraq) and long term, steady state (e.g. Seychelles) scenarios
used in risk assessment. According to Guidelines for Risk Assessment established by the U.S. EPA
(1991), exposures of organisms until the time of sexual maturation should be considered in the
assessment of developmental toxicants.
Exposure of monkeys to methylmercury during development has produced effects on sensory
systems; visual, auditory and somatosensory changes have been observed. These changes were
6-44
-------
apparently permanent. There were inter-individual variations in the degree and type of impairment.
These observations have implications for human studies in that they emphasize the necessity of assessing
the function integrity of multiple sensory systems.
Rice (1996) noted that delayed neurotoxic effects were observed in monkeys six years after
exposure was stopped. This was expressed as clumsiness and decreased ability to grasp a variety of
objects when the animals were 13 years old. The author pointed out the relevance of this finding to the
potential for neurodegenerative disease development in aging humans exposed to methylmercury earlier
in life. There is increasing evidence, from Minamata and other exposures, that degenerative processes
associated with aging are exacerbated by mercury exposure (Igata 1993; Schantz et al. 1996).
Monkeys exposed to methylmercury in utero showed variable effects on cognitive processes
(Rice 1996). These animals were delayed in development of object permanence (ability to conceptualize
the existence of a hidden object) and recognition memory (recognition effaces). These abilities,
however, did not appear to be impaired in adult animals. The methylmercury effect appears to have been
a developmental delay rather than a permanent impairment. The emotional and social sequelae of a
similar reversible developmental delay in humans is not known. Newland et al. (1994) demonstrated a
persistent learning impairment in a group of three squirrel monkeys (Saimiri sciureus) exposed to
methylmercury in utero. They concluded that the toxic effect led to insensitivity to changes in the
consequences of behaviors.
6-45
-------
Table 6-15
Studies in Non-Human Primates
Investigators /Exposure schedule
Rice and co-workers
birth- 7 yrs
in utero - 4yrs
Gunderson, Burbacher et al.
in utero
Newland et al
in utero
dose
Mg/kg/d
0
50
0
10
25
50
0
50
0
n
5
5
5
1
2
5
12
12
3
3
Total blood mercury in ppm
peak at 200
days
<0.1
1.2
mothers
<0.1
0.3
0.7
1.4
mothers
1.1
mothers
0.7-0.9
birth
<0.1
0.5
0.9
2.7
birth
1.6
birth
Steady state
<0.1
0.7
Steady state
<0.1
0.2
0.4
0.7
The study of visual psychophysics provides information about function and dysfunction in the
visual system, which can be applied to studies in humans. For example data on contrast sensitivity
functions in non-human primates provide links between important features of visual function as
expressed in behavior and the neural mechanisms underlying vision. The learning impairments
observed in behavior under concurrent schedules not only raise concerns about cognitive effects of
methylmercury exposure but also point to behavioral mechanisms by which these effects occur. Results
suggesting methylmercury-related deficits in the visual recognition effaces are congruent with
well-established areas of neuroscience that show how higher-order functioning is accomplished in the
primate sensory (including visual) cortex. These results point to links between the integration of
complex visual information and higher order cognitive abilities.
The sensory and motor deficits observed in animals exposed to methylmercury indicate that the
exposed individual is missing the full complement of important capabilities. There are implications for
the adversity and long-term impact of some of the subtle changes in cognitive ability noted in children
exposed to mercury in utero. This research is leading to the recognition that forms of learning and
reading dysfunction in people can be traced to subtle alterations in sensory systems; these findings raise
concerns about deficits in functional domains not traditionally linked directly to sensory function. The
motor deficits are consistent with neural systems that are affected by methylmercury and, therefore,
indicate non-trivial impairment of the individual. Whether the cognitive endpoints are traceable to this
sensory loss remains to be determined, but some, such as the learning deficits under concurrent schedules
or alterations in fixed-interval schedule performance may be independent of such loss.
6-46
-------
The rodent studies support conclusions drawn from non-human primates, although the results are
not always as consistent. Rats have most often been used, and the endpoints identified to date have
usually been less specific than those examined in the primate literature. Deficits in schedule-controlled
operant behavior (Bornhausen et al. 1980; Schreiner et al. 1986) and subtle characteristics of motor
function (Eisner 1991) have, however, been reported. The rodent studies generally find effects at doses
predictable from a consideration of the kinetics of methylmercury in these species and the sensitivity of
the procedures used. Data on cognitive function generally show only weak effects and at high dose.
Most often observed have been deficiencies in motor function. Sensory system function has not been as
extensively tested, although there have been reports of effects for in utero exposure on visual evoked
potentials (Zenick 1976; Dyer et al. 1978).
Rice (1996) used data from animal studies described above to derive RfDs for methylmercury.
She identified LOAELs (in the absence of aNOAEL) of 0.01 to 0.05 mg/kg/day from the studies of
effects in monkeys exposed in utero and/or post partum. Standard U.S.EPA methodology was followed
in the application of 10-fold uncertainty factors for the following: variability in human populations,
extrapolation from animal data and use of a LOAEL in the absence of a NOAEL. Dividing by the
uncertainty factor of 1000 results in a range of RfDs of 0.01 to 0.05 (jg/kg/day. If one uses the rat data, a
NOAEL of 0.005 mg/kg/day is identified and two 10 fold uncertainty factors are most appropriate (for
human variability and extrapolation from animal data). The resulting RfD would be 0.05 (ig/kg/day.
RfDs based on use of sensitive, but relevant, endpoints measured in animals are 5- to 10-fold lower than
those calculated from data in humans. This leads to a conclusion that the RfD based on observation of
clinical and other effects in Iraqi children is not unduly conservative.
Uncertainty in the dose conversion (maternal hair mercury to dietary intake)
It was assumed in the derivation of the RfD that there was no substantial difference in
pharmacokinetics of methylmercury as a consequence of the food medium in which it is presented to
humans. The recent SAB report (U.S.EPA 1997) makes the following statement.
There is no compelling evidence to suggest that the toxicokinetics of methylmercury (MeHg)
ingested in grain (as in the Iraqi poisoning episode) is different from that resulting form ingestion offish
(the typical exposure route of humans). The best evidence for this is a study in which cats were fed
contaminated fish, control fish, or MeHg in a non-fish diet (Charbonneau et al., 1974). No differences
were observed in degree of MeHg neurotoxicity, latency to toxicity (ataxia), tissue levels or distribution
ofHg.
Gearhart et al. (1995) applied a PBPK model to their benchmark dose calculated in ppm maternal
hair mercury from the Kjellstrom et al. (1986a, b, 1989) data. Details of the application of this model
were not provided in the publication. A dose-conversion factor of about 0.05 [ppm mercury in
hair/((jg/kg-day)] can be estimated from Figure 5c in Gearhart et al. (1995). Applying this factor to the
benchmarks of 17 to 50 ppm mercury in hair yields the reported intake range of 0.85 to 25 (ig/kg-day.
Although this dose-conversion factor is about half of that estimated by the U.S. EPA, the two estimates
are consistent when duration of exposure is considered. The Gearhart et al. (1995) pharmacokinetic
model predicts that equilibrium is not reached until about 400 days. The dose conversion of 0.05
corresponds to hair mercury levels at equilibrium and is the appropriate factor to apply to the New
Zealand hair concentrations, which arose from longer-term exposure. In contrast, the Iraqi exposure was
only for a few months (Marsh et al. 1978). A dose conversion of about 0.1, which is virtually the same
as that used by the U.S. EPA, can be estimated for a 3-month exposure from Figure 4 in Gearhart et al.
(1995) and from Sherlock et al. (1984) assuming a hairblood mercury concentration ratio of 250:1.
In their 1994 Toxicological Profile, ATSDRused the analysis reported by Cox et al. (1989), (see
discussion below) of the Iraqi developmental data in the derivation of an intermediate MRL (minimal
6-47
-------
risk level). Using delayed onset of walking as the critical effect, a LOAEL of 14 ppm mercury in hair
was determined. A dose conversion from ppm hair to daily intake to maintain blood mercury levels in
pregnant women was done in a very similar manner to that employed by U.S. EPA. Values for
parameters in the equation on page 6-18 were consistent between the two Agencies with one exception;
namely the use of a blood volume of 4.1L by ATSDR compared to 5L by U.S. EPA. The methylmercury
intake level calculated by ATSDR to maintain a hair level of 14 ppm is 1.2 //g/kg-day compared to 1.1
/ug/kg-day to maintain a hair level of 11 ppm (used by U.S. EPA).
ATSDR (1997) has recently released for public comment an updated Toxicological Profile with
a revised Minimal Risk Level (MRL) for methylmercury. In their dose conversion, they used the
following values: hairblood ratio = 250; body weight = 60 kg; blood volume = 4.2 L; elimination
constant = 0.014; absorbed dose found in the blood = 0.05; absorption factor = 95%.
The state of New Jersey currently uses an RfD of 0.7xlO"4mg/kg-day (described in Stern 1993)
compared to the U.S. EPA's RfD of IxlO"4 mg/kg-day. The critical effect chosen was developmental
endpoints in the Iraqi children exposed in utero including delayed onset of walking. A recent discussion
of this RfD was presented in the context of the external peer review of the Mercury Study Report to
Congress. Stern described the LOAEL as the mercury hair level equivalent to a mercury blood level of
44 //g/L. To determine the intake level, the equation on page 6-18 was used but with different values for
two parameters; namely, b and f.
= 0.70 /ug/kg-day
C x b x V
d =
A x f x bw
= 44 uglL x 0.013 days x 5L
0.95 x 0.077 x 60 kg
Choice of the value of 0.077 for f, fraction of daily intake taken up by blood was based on a
paper by Smith et al. (1994) which was not published at the time of the RfD/RfC work group discussions
and was not considered by them in choosing the parameter values. Smith et al. (1994) (described briefly
in chapter 2 of this volume) presents a study of methylmercury excretion kinetics based on measurement
of i.v. administered methylmercury (1.7-7.4 (jg) in blood, urine and feces of 7 male volunteers. The
authors claim that data from this study are superior to those from previous studies in accounting for the
portion of the labeled mercury which is metabolized to inorganic mercury. Based on the linear
extrapolation of the plot of blood concentration of methylmercury versus time, the authors calculated that
approximately 7.5% of the methylmercury remained in the blood following rapid equilibration among
tissue compartments. Based on fitting the experimental data to a five compartment pharmacokinetic
model, they calculate that 7.7% (geometric mean) of the methylmercury is found in the blood. It should
be noted that the values for this parameter among the seven subjects ranged from 6.5-9.5%.
Smith et al. (1994) taking conversion to inorganic mercury into account, reported an overall
estimate (geometric mean) of the half-life in blood (methylmercury-specific as per discussion in previous
paragraph) of 45 days (0.015 days"1). The half-lives (elimination constants) for the 7 subjects ranged
from 35 days (0.020 days"1) to 53 days (0.013 days"1). Stern (1993) notes the half-life reported by Cox et
al. (1989) was 48 days with a range of about 18-37. This corresponds to a value forb of 0.0144 day"1.
The mean value is not reported by Cox, but a Monte Carlo simulation of the data estimated a mean of
about 47 days. The most frequently reported value (mode), however, was 55 days corresponding to a
6-48
-------
value for b of 0.013 day"1. Ultimately the value of b = 0.013 day"1 was chosen by Stern as the most
"typical" value.
Uncertainty factors
The current RfD was derived by application of an uncertainty factor of 10. This was intended to
cover three areas of uncertainty: lack of data from a two-generation reproductive assay; variability in the
human population, in particular the wide variation in biological half-life of methylmercury and the
variation that occurs in the hair to blood ratio for mercury; and lack of data on long term sequelae of
developmental effects. There was no factor applied for sensitive subpopulations as data were obtained
from exposure to a sensitive human group; namely, the developing fetus.
The interpretation by some risk assessors is that the effects noted in the Iraqi population exposed
to contaminated grain are not being seen at similar doses of methylmercury delivered in utero via
contaminated seafood. One assessment by a scientist at FDA is that the U.S. EPA RfD of l.OxlO"4
mg/kg-day for methylmercury is somewhat conservative and is certainly protective; a suggestion was
made that the uncertainty factor could be decreased to 3, resulting in a RfD of 3.OxlO"4mg/kg-day.
Stern (1997) did an analysis of the degree of uncertainty associated with interindividual
variability applied to the dose conversion from maternal hair mercury to ingested daily dose of
methylmercury. His conclusion was that a calculated ingested dose intended to be inclusive of 95-99%
of women 18-40 years old would be 0.1 to 0.3 (ig/kg/day (by contrast to EPA's calculated 1 (ig/kg/day).
His recommended uncertainty factor of three (for lack of reproductive data and data on long-term
sequelae) would result in a RfD of 0.01 to 0.03 (ig/kg/day.
ATSDR has recently released for public comment an updated Toxicological Profile with a
revised Minimal Risk Level (MRL) for methylmercury. This report chooses as a NOAEL the median
maternal hair mercury reported by Davidson et al. (1995) for the 29 month old Seychellois children
tested with the BSID and Bayley Infant Behavior record. The Toxicological Profile characterizes the
reported decrease in the male children's activity level as not adverse and chooses use of a midpoint of all
measured maternal hair levels rather than the highest measure or median of the top quartile. An
uncertainty factor was not used to account for human variability.
The recent review of the Mercury Study Report to Congress by the SAB recommended that EPA
consider information suggesting that the uncertainty factor applied to the Iraqi data be increased. Their
arguments are as follows.
For example, the Faeroe Islands data (also animal data) indicates that the fetal exposure may be
greater than maternal exposure. In this study fetal cord blood mercury levels averaged 80.2 ppb while
maternal blood levels were only 38.1 ppb. Animal data supports that the fetus may act as a sink for
mercury. The report extensively reviews blood mercury kinetics but has little to say about fetal brain
mercury levels. Although the data are slight there are indications from recent monkey studies that the
brain mercury half-life is very long (Vahter et al. 1995). The RfD for MeHg is based on results from an
acute exposure study while most MeHg exposure is thought to be long term. This may be an additional
reason to increase the uncertainty factor. There are also indications of age related changes where MeHg
may accelerate neurodegeneration associated with aging from human data (Igata, 1993) and animal data
(Rice 1989a; 1989b)). In evaluating neurotoxic effects from low exposures such as with methylmercury,
it must be remembered that few individuals may actually demonstrate clinical signs of disease but many
individuals may suffer subtle changes which can produce total population effects.
Risks among subpopulations
6-49
-------
The recent review by the SAB of the Mercury Study Report to Congress discusses at some length
the likelihood that there are differential responses among human subgroups. That discussion is excerpted
below.
Effect modification occurs when, at equal doses of a toxicant, adverse outcomes are observed in
some members of a population but not others. The affected individuals may possess one or more
distinctive characteristics such as age (stage of development), gender, social class, or certain premorbid
health factors (e.g., diabetes, liver disease, pulmonary dysfunction) or genetic predisposing factors that
are not well represented in unaffected members of the population.
Most studies of effects of environmental agents on child development have treated potential
effect modifiers as covariates or confounders in multiple linear regression models without interaction
terms models, or as matching variables in comparing so-called "exposed" and "unexposed" groups.
Interactions are infrequently explored or are dismissed.
The animal literature, however, presents numerous examples of neurotoxicity enhancement or
buffering as a result of species, strain, drug, and physical and social environmental interactions. An
example from the methylmercury literature is the phenomenon of male susceptibility. Studies have
shown that the relative risk of perinatal morbidity and mortality is higher in human males (Abramowizc
and Barnett 1970; Naeye et al, 1971), including the risk of poor reproductive outcomes and postnatal
development due to fetal exposure to industrial pollutants (Scragg et al. 1977; McKeown-Eyssen et al.
1983). Males also have a higher rate of cognitive developmental disability in the general population
(Gross and Wilson 1974) and display more profound intellectual deficits as a result of cortical lesions
(Bornstein and Matazarro 1984; Inglis and Lawson 1981). The McKeown-Eyssen et al. (1983) study
reported gender-related differences including dose-related deficits in sensorimotor behaviors assessed in
the BSID and increased prevalence of abnormal muscle tone and deep tendon. The latter effect was not
associated with methylmercury dose in females. In the Iraqi poisoning more severe neurological effects
were observed in male children (Marsh et al. 1987). Animal experiments have also observed sex
differences in neurodevelopmental vulnerability. For example, in Sageretal. (1984) a single low dose
of methylmercury administered to neonatal mice resulted in mitotic arrest in cells of the granule layer of
the cerebellum only in males.
There is evidence that lifestyle factors such as the quality of the home environment and nutrition
play a role in the expression of developmental neurotoxicity. The literature on connections between
social factors and methylmercury toxicity is sparse. Studies on lead, however, have found greater
neurocognitive deficits in exposed individuals from the lowest socio-economic groups (e.g., Bellinger et
al. 1989; Dietrich etal. 1987; Harvey et al. 1984; Lansdown et al. 1986).
The positive nutritional factors of a seafood diet may be a factor in the greater delay in the onset
of the Minamata as compared to the Iraqi outbreaks. Early results from the Faroe Islands studies have
shown a positive association between cord blood methylmercury concentrations and birth weight
(Grandjean et al. 1992; Grandjean etal. 1995). The authors attribute the finding to the benefits of n-3
polyunsaturated fatty acids in a high seafood diet and to the benefits of breast-feeding which, in itself,
can lead to higher methylmercury intake by the infant.
The protective effects of genetic and environmental factors may be expressed in the studies of
Seychellois children (Davidson et al. 1995). At 19 and 29, months scores on the Psychomotor
Development Index (PDI) were very negatively skewed. The mean PDI scores at 19 and 29 months were
1.7 and 1.3 standard deviations above the United States means of 100 +/- 16 points respectively. That is,
means for the PDI recorded for Seychelles children are above what would be classified as "accelerated
performance" on these scales. The "developmental health" of this population is also reflected in the
small number of subjects attaining abnormal scores on the DDST-R by comparison to samples in the
United States. Only 3 out of 737 individual examinations or 0.4% were rated as abnormal (i.e., below
6-50
-------
the 10th percentile for U.S. norms). Accelerated motor development has been noted in previous studies
of African cultures and is also observed in African-American infants under two years of age.
The conclusion of the SAB, however, was that "the data regarding effect modification in human
epidemiologic studies of mercury poisoning are currently too meager to base separate estimates of
human health risks or establish different RFD's for various subpopulations."
Other areas of uncertainty
Birth date uncertainty would have an impact on exposure uncertainty if correspondence of
exposure and gestation was estimated (Marsh et al. 1978) from birth date to any great extent. That is,
exposure may have occurred to a lesser extent (or not at all) than assumed during the critical period of
gestation. The result would be a lower exposure associated with the observation, depending on the width
of the critical time window during gestation and on the importance of duration of exposure in the
elicitation of the particular effect. If the exposure occurred after the critical period, any observation of an
effect would be attributed to causes other than methylmercury and be included in the background.
Several scientists have suggested that a developmental toxicity RfD is needed for
methylmercury. This may not be necessary, however, if the critical effect is developmental toxicity and
the uncertainty factors used to estimate the lifetime RfD do not involve an adjustment for less than
lifetime exposure nor lack of complete data base.
6-51
-------
6.3.1.2 Inhalation Reference Concentrations (RfCs)
Elemental mercury
The U.S. EPA has determined an RfC of SxlO'Vg/m3 for elemental mercury (U.S. EPA 1994).
The inhalation RfC is based on neurologic toxicity observed in several human occupational studies. The
observed neurologic changes included hand tremor, increases in memory disturbances and slight
subjective and objective evidence of autonomic dysfunction. Fawer et al. (1983) measured intention
tremor (tremors that occur at the initiation of voluntary movements) in workers exposed to a TWA
concentration of 0.026 mg/m3 over an average of 15.3 years. It was noted, however, that the tremors
may have resulted from intermittent exposures to concentrations higher than the TWA.
Piikivi and colleagues conducted several studies in chloralkali workers on electroencephalogram
(EEG) abnormalities (Piikivi and Toulonen 1989); subjective measures of memory disturbance and sleep
disorders and objective disturbances in psychological performance (Piikivi and Hanninen 1989); and
subjective and objective symptoms of autonomic dysfunction such as induced pulse rate variations and
blood pressure responses (Piikivi 1989). U.S. EPA extrapolated an occupational exposure level
associated with these neurological changes of 0.025-0.030 mg/m3 from blood levels, based on a
conversion factor calculated by Roels et al. (1987). The LOAEL (0.025 mg/m3 adjusted to 0.009 mg/m3
for continuous exposure of the general population) was divided by an uncertainty factor of 30 (10 to
protect sensitive individuals and for use of a LOAEL, and 3 for the lack of reproductive studies in the
database) to yield the RfC of 3xlO"4 mg/m3.
The RfC was, thus, calculated in the following way.
Rfc = LOAEL mg Hg/m3
UF
= 0.009 mg/m3
30
= 0.0003 mg/m3
This reference concentration was reviewed and verified by the RfD/RfC Work Group and was
verified on April 19, 1990. It was released under a special action by U.S. EPA (Jarabek 1992, personal
communication).
Confidence in the critical study, the data base, and, thus in the RfC were rated "medium" by the
Work Group. Factors which were positive for confidence in the critical study were the use of a sufficient
number of human subjects, inclusion of appropriate controls, sufficient exposure duration and that the
LOAEL can be corroborated in other studies. It was noted, however, that for all but one of the studies,
exposure had to be extrapolated from blood mercury levels. The lack of human or multispecies
reproductive or developmental studies precluded higher confidence in the data base.
Inorganic (mercuric) mercury
Developmental toxicity (skeletal abnormalities and retarded growth) in mice (Selypes et al.
1984) and autoimmune disease in Brown-Norway rats (Bernaudin et al. 1981) have also been observed
following inhalation exposures. Due to the limitations of these inhalation studies and the inadequacy of
the remaining toxicologic and pharmacokinetic data bases, the RfD/RfC Work Group determined the
derivation of an RfC is not possible. The posting of this determination on IRIS is proceeding concurrent
with the finalization of this Report to Congress.
6-52
-------
Methylmercury
No estimate of risk from inhalation of methylmercury has been done by U.S. EPA.
6.3.1.3 Estimation of Risk from Dermal Exposure
The dermal contribution of the different mercury species to the total systemic exposure of each
of these mercury species may be important for a full characterization of risk to the potentially exposed
human populations. Many of the necessary data needed for conducting a dermal risk assessment are
currently lacking or not well enough understood to assess systemic exposure and risks from dermal
exposure to mercury species.
For any of these mercury species to be a dermal health hazard they must be absorbed across the
skin (epidermis and dermis) and be systemically distributed to the affected critical organ systems
(kidneys or CNS) via the circulatory system. The percutaneous absorption for each mercury species is
dependent on skin-specific factors (e.g., skin thickness, hydration, age, condition, circulation, and
temperature) and compound-specific factors (e.g., lipophilicity, polarity, chemical structure, volatility,
and solubility), which are involved in determining the rate and amount of absorption by the cutaneous
route. Currently there are few known or agreed upon percutaneous absorption rates available for any of
the mercury species of interest. Some data on percutaneous absorption (Kp = Permeability coefficient)
for the mercuric forms of mercury in aqueous media have been reported in U.S. EPA (1992).
The media (aqueous, vapors or soil) where the mercury species are found must also be
considered in dermal risk assessments. Each medium has its own set of factors that impact the specific
percutaneous absorption rates for each of the mercury species. For example, mercury compounds found
in aqueous media are dependent on factors such as solubility in water and increased hydration of the
skin. Mercury compounds associated with soil must be assessed for binding to the particular soil type of
concern, the adhesion of the soil of concern to skin, the desorption of the mercury compound from the
soil, and the absorption of the mercury compound across the skin and into the circulatory system. Other
aspects that must be considered with a dermal assessment are binding or sequestration of the mercury
species at the site of exposure or closely nearby, and the metabolism of the mercury species in the skin
that may result in oxidation/reduction of the mercury species to other valence states (thereby, potentially
resulting in different critical effects than from the originally absorbed compound).
At present, many of the necessary mercury species/media factors have not been fully ascertained
and as a result credible dermal risk assessments cannot be accomplished at this time. A more extensive
discussion of dermal exposure assessment for risk assessment can be found in Dermal Exposure
Assessment: Principles and Applications (EPA/600/8-91/01 IB, January 1992).
6-53
-------
6.3.2 Developmental Effects
6.3.2.1 Elemental Mercury
Elemental mercury was judged to have sufficient animal data for developmental toxicity. The
two studies which contribute most to the level of concern are Danielsson et al. (1993) and Fredriksson et
al. (1992). Both of these studies are limited as a basis for an RfDDT by the small numbers of animals
tested and by the very few dose groups. A further limitation is lack of data on gender differences. No
RfDDT for elemental mercury is available at this time.
6.3.2.2 Inorganic Mercury
There are no data from human studies which are suitable for derivation of an RfDDT. Inspection
of available animal studies indicates that there are five reports of developmental effects of inorganic
mercury given orally. In all of these, exposure was by gavage to pregnant animals, and effects were
monitored in progeny. Three papers were reported as abstracts giving few experimental details.
Rizzo and Furst (1972) treated Long Evans rats (5/group) with a single gavage dose of 2 mg
Hg/kg as mercuric oxide on either day 5, 12, or 19 of gestation. Animals were sacrificed on day 20 or 21
of gestation. No effects of treatment on gestation day 12 or 19 were noted. According to the authors
treatment on day 5 resulted in a higher percentage of growth retardation and inhibition of eye formation,
but no statistical analyses were done.
In Gale (1974), pregnant Golden hamsters were administered 0, 2.5, 5, 16, 22, 32, 47, or 63 mg
Hg/kg-day mercuric acetate via gavage on gestation day 8. When the pregnant animals were sacrificed
on day 12 or 14, there was a significant increase in the incidence of abnormal fetuses including small,
retarded, or edematous (combined), and/or malformed fetuses. The NOAEL for developmental toxicity
was 5 mg mercuric chloride/kg or 2 mg Hg/kg. Maternal toxicity included dose-related weight loss,
diarrhea, slight tremor, somnolence, tubular necrosis and hepatocellular vacuolization, but insufficient
data were provided to allow determination of a LOAEL or NOAEL for maternal toxicity.
The advantage of the Gale (1974) study as a basis for quantitation of potential risk is that several
doses of inorganic mercury were tested; the spacing of the doses was adequate for identification of both a
LOAEL and NOAEL. It is not recommended, however, that the NOAEL in Gale (1974) serve as the
basis for an RfDDT. There were relatively few animals tested (decreasing the overall sensitivity of the
assay) and not all endpoints were thoroughly evaluated. The test compound was administered on only
one day of gestation, and there is some question as to the suitability of the golden hamster for
developmental assays. The data base for developmental effects, while generally supportive of the
LOAEL is not adequate to determine if the measured endpoints were the most sensitive for
developmental effects of inorganic mercury.
6.3.2.3 Methylmercury
Weight of evidence for developmental toxicity indicates that a developmental toxicity RfD is
appropriate for methylmercury. A separate RfDDT may not be necessary as the critical effect for the
lifetime RfD is developmental toxicity. The current RfD (IxlCT4 mg/kg-day) was based on
developmental endpoints in offspring of women exposed during pregnancy; it may be taken as protective
against developmental toxicity. For less than chronic exposures it should be noted that the RfDDT is not
intended as a lifetime exposure value.
6.3.3 Germ Cell Mutagenicitv
6-54
-------
Data do not support the generation of quantitative estimates for germ cell mutagenicity for any
form of mercury.
6.3.4 Carcinogenic Effects
6.3.4.1 Elemental Mercury
Elemental mercury is categorized as D, unable to classify as to human carcinogenicity. A
quantitative estimate for carcinogenic effect is, thus, inappropriate.
6.3.4.2 Inorganic Mercury
Quantification of the potential carcinogenic effects of mercuric chloride (classified as C, possible
human carcinogen) was not done. No increase in tumor incidence was observed in a carcinogenicity
study in which white Swiss mice were given 0.95 mg Hg/kg-day as mercuric chloride in drinking water
(Schroeder and Mitchener 1975). No statement regarding carcinogenicity was reported in a 2-year
feeding study in which rats were administered mercuric acetate in the diet at doses of 0, 0.02, 0.1, 0.4,
1.7 and 6.9 mg Hg/kg-day (Fitzhugh et al. 1950).
The incidence of squamous cell papillomas of the forestomach and thyroid follicular cell
carcinomas from NTP (1993) was evaluated. No slope factor was based on the forestomach tumors
because this type of tumor is probably the result of irritation of the forestomach, cell death and epithelial
proliferation. The carcinogenic mechanism may be specific to irritation at the high doses used in the
bioassay; use of these tumors as a basis for human health assessment of low doses of inorganic mercury
is inappropriate.
Regarding the thyroid carcinomas, a variety of drugs, chemicals, and physiological perturbations
result in the development of thyroid follicular tumors in rodents. For a number of chemicals, the
mechanism of tumor development appears to be a secondary effect of longstanding hypersecretion of
thyroid-stimulating hormone by the pituitary (Capen and Martin 1989; McClain 1989). In the absence of
such long-term stimulatory effects, induction of thyroid follicular cell cancer by such chemicals usually
does not occur (Hill 1989). Use of the incidence of thyroid tumors from NTP (1993) in low dose
extrapolation is, thus, questionable.
6.3.4.3 Methylmercury
Quantification of the potential carcinogenic effects of methylmercury (classified as C, possible
human carcinogen) was not done. No increased incidence of tumors was seen in rats exposed to doses of
up to 0.34 mg Hg/kg-day for 130 weeks (Mitsumori et al. 1983, 1984) or in cats exposed to a diet
containing up to 0.176 mg Hg/kg-day for 2 years (Charbonneau et al. 1976).
No slope factor was calculated for methylmercury based on the incidence of renal epithelial
tumors in male mice. The two studies by Mitsumori et al. (1981, 1990) were limited by high mortality in
the high-dose males, the only group to exhibit a statistically significant increase in tumor incidence. The
study by Hirano et al. (1986) was not limited by survival problems, but the tumors were observed in
conjunction with nephrotoxicity and appear to be a high-dose phenomenon that may not be linear at low
doses. The tumors appeared to originate from focal hyperplasia of the tubular epithelium induced as a
reparative change. The hyperplasia was not observed in tubular epithelium that was undergoing early
degenerative changes; thus, the tumors may not occur where degenerative changes do not occur. The
appropriateness of deriving a quantitative risk estimate using the assumption of linearity at low doses
based on data for which a threshold may exist is questionable.
6-55
-------
6.4 Risk Assessments Done By Other Groups
Quantitative estimates of hazards of oral exposure to methylmercury exposure have been
considered by the Food and Drug Administration, Agency for Toxic Substances and Disease Registry
(ATSDR), the Department of Energy and several State agencies. Several inhalation workplace exposure
limits are available in the United States and other countries.
6.4.1 Food and Drug Administration
In 1969, in response to the poisonings in Minamata Bay and Niigata, Japan, the U.S. FDA
proposed an administrative guideline of 0.5 ppm for mercury in fish and shellfish moving in interstate
commerce. This limit was converted to an action level in 1974 (Federal Register 39, 42738, December 6,
1974) and increased to 1.0 ppm in 1979 (Federal Register 44: 3990, January 19, 1979) in recognition that
exposure to mercury was less than originally considered. In 1984, the 1.0 ppm action level was
converted from a mercury standard to one based on methylmercury (Federal Register 49. November 19,
1984).
The action level takes into consideration the tolerable daily intake (TDI) for methylmercury, as
well as information on seafood consumption and associated exposure to methylmercury. The TDI is the
amount of methylmercury that can be consumed daily over a long period of time with a reasonable
certainty of no harm. U.S. FDA (and WHO) established a TDI based on a weekly tolerance of 0.3 mg of
total mercury per person, of which no more than 0.2 mg should be present as methylmercury. These
amounts are equivalent to 5 and 3.3 //g, respectively, per kilogram of body weight. Using the values of
methylmercury, this tolerable level would correspond to approximately 230 //g/week for a 70 kg person
or 33 //g/person/day. The TDI was calculated from data developed in part by Swedish studies of
Japanese individuals poisoned in the episode of Niigata which resulted from the consumption of
contaminated fish and shellfish and the consideration of other studies offish-eating populations.
Based on observations from the poisoning event later in Iraq, U.S. FDA has acknowledged that
the fetus may be more sensitive than adults to the effects of mercury (Federal Register 44: 3990, January
19, 179; Cordle and Tollefson, 1984, U.S. FDA Consumer, September, 1994). In recognition of these
concerns, U.S. FDA has provided advice to pregnant women and women of child-bearing age to limit
their consumption offish known to have high levels of mercury (U.S. FDA Consumer, 1994). U.S. FDA
believes, however, that given existing patterns offish consumption, few women (less than 1%) eating
such high mercury fish will experience slight reductions in the margin of safety. However, due to the
uncertainties associated with the Iraqi study, U.S. FDA has chosen not to use the Iraqi study as a basis
for revising its action level. Instead, the U.S. FDA has chosen to wait for findings of prospective studies
offish-eating populations in the Seychelles Islands and in the Faroes Islands.
6-56
-------
6.4.2 ATSDR
ATSDR has established Minimal Risk Levels (MRLs) for elemental, inorganic and
methylmercury (ATSDR 1994). Recently a revised Toxicological Profile has been released for public
comment (ATSDR 1997).
An acute inhalation MRL of 0.00002 mg/m3 has been derived for elemental mercury vapor based
on neurodevelopmental changes in rats. Specifically, the effects were changes in locomotor activity at 4
months of age and an increased time to complete a radial arm maze at 6 months of age following
exposure to 0.05 mg Hg/m3 for 1 hour during post-partum days 11-17 (Fredriksson et al. 1992). A
chronic inhalation MRL of 0.000014 mg/m3 was derived for elemental mercury vapor based on a
significant increased in the average velocity of naturally occurring tremors in occupational workers
(Fawer et al. 1983). The revised chronic MRL is calculated to be 0.0002 mg/m3 by application of an
uncertainty factor of 30 to a LOAEL of 0.026 mg/m3 for increased frequency of tremors in
occupationally exposed workers (Fawer et al. 1983).
Acute and intermediate oral MRLs were derived for inorganic mercury based on kidney effects
reported in the 1993 NTP study of mercuric chloride. The acute oral MRL was 0.007 mg Hg/kg-day
based on a 2-week study reporting aNOAEL of 0.93 mg Hg/kg-day for renal effects in rats (NTP 1993).
At higher doses, an increased incidence of tubular necrosis was observed. The intermediate oral MRL of
0.002 mg Hg/kg-day was established, based on a 6-month study reporting a NOAEL of 0.23 mg Hg/kg-
day for renal effects (increased absolute and relative kidney weights) (NTP 1993). There is no indication
that these values have been revised in the 1997 document.
An acute-intermediate oral MRL of 0.00012 mg Hg/kg-day was established in 1994 for
methylmercury. ATSDR derived their assessment from the Marsh et al. (1981) and Cox et al. (1989)
data; the MRL is based on the lowest observed peak of total mercury concentration in maternal hair
(0.0012 mg/kg-day equivalent to a LOAEL of 14 ppm mercury in maternal hair) during pregnancy
associated with a delayed onset of walking in offspring in Iraqi children. This assessment is discussed in
section 6.3.1.1 of this volume.
The 1997 Toxicological Profile calculates a chronic MRL for methylmercury of 0.5 (ig/kg/day.
This report chooses as a NOAEL the median maternal hair mercury of 5.9 ppm reported by Davidson et
al. (1995) for the 29 month old Seychellois children tested with the BSID and Bayley Infant Behavior
record. The Toxicological Profile characterizes the reported decrease in the male children's activity level
as not adverse and chooses use of a midpoint of all measured maternal hair levels rather than the highest
measure or median of the top quartile. Dose conversion was done as in the 1994 document to give an
estimated ingested dose of 0.5 (ig/kg/day. An uncertainty factor was not used to account for human
variability.
6.4.3 Department of Energy
Brookhaven Laboratories has prepared a report for Office of Clean Coal Technology, DOE. This
report describes a probabilistic-based assessment which considered the potential increased health risk for
paresthesia in adults. Their estimate is based upon a yearly emission rate of 180 kg/year from all fossil
fuel power plants in the United States. This estimate represents less than 1% of the existing global pool
of mercury that is introduced into the environment. Based upon the most sensitive adult sign of
paresthesia, the mercury emissions from power plants would result in an increased risk for paresthesia of
0.004-0.007% with an upper 95th percentile risk of 0.013-0.017% (Lipfert et al. 1994).
6-57
-------
6.4.4 National Institute of Environmental Health Sciences (NIEHS)
NIEHS, part of the National Institutes of Health, was required under section 301 of the CAA "to
conduct, and transmit to the Congress by November 15, 1993, a study to determine the threshold level of
mercury exposure below which adverse human health effects are not expected to occur." In section 112
(n)(l)(C), NIEHS was encouraged to evaluate the health effects threshold for mercury in the absence of
specifics as to species of mercury but to consider mercury in fish. As mercury in fish is primarily in the
form of methylmercury, the NIEHS limited their consideration to this species.
The report was completed in 1993 and delivered to Office of Management and Budget for
clearance. It describes dose- response assessments for methylmercury done by WHO, FDA and U.S.
EPA and presents all three estimates as recommended for tolerable mercury concentrations. The NIEHS
report also describes estimates offish consumption by the U.S. population.
6.4.5 Department of Labor
OSHA established a Permissible Exposure Limit (PEL), time-weighted average of 0.05 mg
Hg/m3 for mercury vapor, with a notation for skin exposure (U.S. Department of Labor 1989). A PEL as
a ceiling value of 0.1 mg Hg/m3, also with a notation for dermal exposure was set for aryl mercury and
inorganic mercury compounds.
NIOSH determined a Recommended Exposure Limit (REL), time-weighted average, of 0.05 mg
Hg/m3 for mercury and 0.1 mg Hg/m3 for aryl and inorganic mercury compounds (NIOSH 1973, 1988).
6.4.6 Various States
A number of states have released fish consumption advisories based upon their independent
analysis of the available scientific literature for methylmercury. Most active among these states are
Michigan, New Jersey, Maine, Idaho, and Oregon. Generally, there is a trend to move to more
conservative values based upon developmental neurotoxicity defined in the Marsh et al. (1981) and Cox
et al. (1989) papers. The methylmercury RfD of 0.7xlO"4 mg/kg-day used by the state of New Jersey is
discussed in section 6.3.1.1. Some states are waiting for more specific guidance from U.S. EPA.
6.4.7 World Health Organization
The International Programme on Chemical Safety (IPCS) of the World Health Organization
published a criteria document on mercury (WHO 1990). In that document, it was stated that" a daily
intake of 3 to 7 (ig Hg/kg body weight would cause adverse effects of the nervous system, manifested as
an approximately 5% increase in the incidence of paraesthesias". The IPCS expert group also concluded
that developmental effects in offspring (motor retardation or signs of CNS toxicity) could be detected as
increases over background incidence at maternal hair levels of 10-20 ppm mercury. These levels of
concern were based on evaluation of data including the human poisoning incident in Iraq described in
Chapter3.
6-58
-------
6.4.8 ACGIH
The ACGIH has established Threshold Limit values (TLV) as eight-hour time-weighted
averages. They include the following:
Aryl mercury compounds 0.1 mg Hg/m3
Mercury vapor 0.05 mg Hg/m3
Inorganic mercury 0.1 mg Hg/m3
No STEL is recommended at this time. The Biological Exposure Indices Committee has
recommended values for inorganic mercury in urine and blood of 35 pg/g creatinine and 15 (ig/L
respectively.
The ACGIH classified inorganic mercury including elemental mercury as follows: A4- Not
classifiable as a Human Carcinogen: There are inadequate data on which to classify the agent in terms of
its carcinogenicity in humans and/or animals.
6-59
-------
7.
ONGOING RESEARCH AND RESEARCH NEEDS
7.1 Ongoing Research
Table 7-1 lists ongoing research projects abstracted from the Federal Research in Progress Data
Base (FEDRIP, 1994).
Table 7-1
Ongoing Research
Investigator
Affiliation
Research Description
Sponsor
Human
T. Clarkson
P. Grandjean
W. Markesbery
M. Martin
R. Mitchell
G. Myers
T. Okabe
M. Owens
M. Rosenman
D. Savitz
University of Rochester, Rochester,
NY
Odense University, Odense,
Denmark
University of Kentucky, Lexington,
KY
University of Washington, Seattle,
WA
University of Kentucky, Lexington,
KY
University of Rochester, Rochester,
NY
Baylor College of Dentistry, Dallas,
TX
Science Applications International
Corp,
Falls Church, VA
Morehouse College, Atlanta, GA
University of North Carolina
Chapel Hill, Chapel Hill, NC
Dose-response relationships in humans
exposed to methylmercury and prenatal
and early postnatal body burdens of
methylmercury.
Neurotoxicity risk from exposure to
methylmercury from seafood
Role of mercury and dental amalgams in
Alzheimer's disease
Epidemiology of mercury in dentists
Amalgam restorations and the relative
risk of adverse pregnancy outcome
Child development following prenatal
methylmercury exposure via fish
Establish maximum levels of exposure
from amalgams for dental patients and
personnel
Potential and adverse effects associated
with dental amalgam
Effect of mercury in amalgam and urine
to cognitive functioning in children
Mercury and reproductive health in
women dentists
National Institute of
Environmental Health
Sciences (NIEHS)
NIEHS
National Institute on
Aging
National Institute of
Dental Research
National Institute of
Dental Research
NIEHS
National Institute of
Dental Research
National Institute of
Dental Research
National Institute of
General Medical
Sciences
National Institute of
Dental Research
Animal
P. Bigazzi
T. Burbacher
K. Mottet
K. Pollard
University of Connecticut,
Farmington, CT
University of Washington, Seattle,
WA
University of Washington, Seattle,
WA
University of California, San Diego,
CA
Mercury induced auto-immune disease in
rats
Developmental effects of methylmercury
in monkeys and rats
Long-term toxicity associated with
inorganic mercury and methylmercury
Animal model of systemic autoimmunity
induced by mercury
NIEHS
NIEHS
NIEHS
National Institute of
Arthritis and
Musculoskeletal and
Skin Diseases
7-1
-------
Table 7-1
Ongoing Research (continued)
Investigator
B. Weiss
Affiliation
University of Rochester, Rochester,
NY
Research Description
Neurotoxicity throughout the lifespan of
mice exposed prenatally to
methylmercury
Sponsor
NIEHS
Mechanistic
W. Atehison
D. Barfuss
T. Jensen
D. Lawrence
R. Noelle
K. Pollard
B. Raj anna
K. Ruehl
T. Sarafian
J. Stokes
R. Zalups
Michigan State University, East
Lansing, MI
Georgia State University, Atlanta,
GA
Herbert H. Lehman College, New
York, NY
Albany Medical College, Albany,
NY
Dartmouth Medical School,
Hanover, NH
Scripps Research Institute, San
Diego, CA
Selma University,
Selma, AL
Rutgers University,
New Brunswick, NJ
University of California, Los
Angeles, CA
Mount Desert Island Biological
Lab,
Salsbury Cove, ME
Mercer University School of
Medicine
Neurotoxic mechanism of chronic
methylmercury poisoning
Transport and toxicity of inorganic
mercury in the nephron
Effect on membrane structure and
organelle distribution
Effects of metals on the structure and
function of murine and human
lymphocytes
Effect of mercury on p-lymphocyte
function
Mechanisms of autoantibody response
induced by mercury which target the
nucleolus
Biomechanisms of heavy metal toxicity in
rats
Mechanism of methylmercury
neurotoxicity during development in mice
Effect of methylmercury on protein
phosphorylation in cerebellar granule cells
in brain
Effects of mercurials on transport
properties of the bladder
Cytotoxicity of mercuric chloride to
isolated rat proximal tubular cells
NIEHS
NIEHS
National Institute of
General Medical
Sciences
NIEHS
NIEHS
National Institute of
Allergy and Infectious
Diseases
National Institute of
General Medical
Sciences
NIEHS
NIEHS
NIEHS
NIEHS
Two of these ongoing studies deserve further discussion because they may fill critical data needs
for the development of a reference dose for methylmercury. The first is the Seychelles Islands Study led
by Dr. T.W. Clarkson from the University of Rochester. The objective of this study is to define the
extent of human health risks from prenatal exposure to methylmercury. Dose-response relationships in a
human population with dietary exposure to methylmercury at levels believed to be in the range of the
threshold for developmental toxicity are being studied. Both prenatal and early postnatal body burdens
of methylmercury will be examined as well as transport to the brain.
This study is testing the hypothesis, developed in previous studies of prenatal exposure in the
Iraq population, that subtle psychological and behavioral changes in prenatally exposed children can be
quantitatively related using dose-response models to the mother's methylmercury exposure during
pregnancy. In the Seychelles, a group of islands off the coast of Africa near Madagascar, a group of 779
infants who were prenatally exposed to methylmercury through maternal fish consumption is being
studied with annual administration of neurodevelopmental, psychological and educational testing of the
7-2
-------
children through 5.5 years of age. This population consumes a relatively large amount of marine fish and
marine mammals, both of which are likely to contain methylmercury. The study is testing the hypothesis
that methylmercury concentration in hair correlates with methylmercury in the brain by using human
autopsy data. Mechanisms of transport of methylmercury across the blood brain barrier also are being
studied to understand better the factors that limit the accuracy of hair mercury as a biological marker for
target tissue levels. Findings reported in recent publications are summarized in section 3.3.1.1.
The second study is the Faroe Islands Study led by Dr. P.A. Grandjean from Odense University
in Denmark. The purpose of this study is to determine whether a neurotoxic risk is present from
methylmercury exposure from seafood and, if so, the threshold for such effects. This study is examining
a cohort of 1,000 children in the Faroe Islands, located in the North Atlantic between Scotland and
Iceland. As is the case in the Seychelles, this population consumes a relatively large amount of seafood;
consumption includes marine fish and marine mammals. Intrauterine exposures were determined by
mercury analysis of umbilical cord blood and maternal hair collected at consecutive births during 21
months in 1986 and 1987. In 13 percent of the births, mercury levels were greater than 10 ppm in
maternal hair, and 25 percent of the cord blood samples had a mercury concentration above the
corresponding level of 40 (ig/L. No cases of gross methylmercury poisoning have been observed. The
persistence of mercury in the body is being assessed from mercury hair concentrations in the children at
one and six years of age, and dietary information is being collected. A detailed pediatric examination
and a test battery to identify possible subtle signs of neurobehavioral dysfunction are being conducted.
The test battery includes psychological tests and neurophysiological measurement of evoked potentials;
these methods are known from previous research to be particularly sensitive to the types of neurotoxicity
expected.
The Faroese population was chosen for this study because of the homogeneity and stability of the
population and the efficient coverage of the Danish health care system. The cohort includes 75% of all
births occurring during the sampling period. A high participation rate (about 80%) is expected at the 6-
year examination period. Alcohol use is minimal in Faroese women (75% were abstainers during
pregnancy), and 60% are nonsmokers. The lead exposure is low (median lead concentration in cord
blood was 1.7 (ig/100 mL). Exposure to polychlorinated biphenyls (PCB), however, may be a
confounder, and alcohol intake of the fathers may have been high. Due to the high seafood intake,
selenium exposure is increased, and its possible protective action against mercury toxicity is being
examined. Findings reported at recent scientific meetings are summarized in section 3.3.1.1.
7.2 Research Needs
In addition to the ongoing studies described above, further research is necessary for refinement
of the U.S. EPA's risk assessments for mercury and mercury compounds. In order to reduce uncertainties
in the current estimates of the oral reference doses (RfDs) and inhalation reference concentrations
(RfCs), longer-term studies with low-dose exposures are necessary. In particular, epidemiological
studies should emphasize comprehensive exposure data with respect to both dose and duration of
exposure. The current RfD and RfC values have been determined for the most sensitive toxicity endpoint
for each compound; that is, the neurological effects observed following exposure to elemental or
methylmercury, and the renal autoimmune glomerulonephritis following exposure to inorganic mercury.
For each of these compounds, experiments conducted at increasingly lower doses with more sensitive
measures of effect will improve understanding of the respective dose-response relationships at lower
exposure levels and the anticipated thresholds for the respective effects in humans. Similar information
from developmental toxicity studies would allow determination of RfDs for developmental toxicity
(RfDdt) for elemental and inorganic mercury. For inorganic mercury, furthermore, the many ongoing
studies in which mechanisms of action are being investigated will greatly assist in quantifying the risks
posed by these compounds.
7-3
-------
Well-conducted studies are also needed to clarify exposure levels at which toxic effects other
than those defined as "critical" could occur in humans. For all three forms of mercury, data are
inadequate, conflicting, or absent for the following: adverse reproductive effects (effects on function or
outcome, including multigeneration exposure); impairment of immune function; and genotoxic effects on
human somatic or germinal cells (elemental and inorganic mercury). Investigations that relate the toxic
effects to biomonitoring data will be invaluable in quantifying the risks posed by these mercury
compounds. In addition, work should focus on subpopulations that have elevated risk because they are
exposed to higher levels of mercury at home or in the workplace, because they are also simultaneously
exposed to other hazardous chemicals, or because they have an increased sensitivity to mercury toxicity.
Information on postnatal exposure without prenatal exposure is limited; therefore, analyzing the potential
risks associated with mercury exposure of young children is difficult.
There are data gaps in the carcinogenicity assessments for each of the mercury compounds. The
U.S. EPA's weight-of-evidence classification of elemental mercury (Group D) is based on studies in
workers who were also potentially exposed to other hazardous compounds including radioactive isotopes,
asbestos, or arsenic. There were no appropriate animal studies available for this compound.
Studies providing information on the mode of action of inorganic mercury and methylmercury in
producing tumors will be of particular use in defining the nature of the dose response relationship.
The assessment of both noncarcinogenic effects and carcinogenic effects will be improved by an
increased understanding of the toxicokinetics of these mercury compounds. In particular, quantitative
studies that compare the three forms of mercury across species and/or across routes of exposure are vital
for the extrapolation of animal data when assessing human risk. For elemental mercury there is a need
for quantitative assessment of the relationship between inhaled concentration and delivery to the brain or
fetus; in particular the rate of elemental to mercuric conversion mediated by catalase and the effect of
blood flow. Such assessment is needed for evaluation of the impact of mercury exposure from dental
amalgam.
Work has been done on development of physiologically-based pharmacokinetic models. While
one of these has developed a fetal submodel, data on fetal pharmacokinetics are generally lacking. The
toxicokinetics of mercury as a function of various developmental stages should be explored. Elemental
mercury and methylmercury appear to have the same site of action in adults; research is, therefore,
needed on the potential for neurotoxicity in newborns when the mother is exposed. This work should be
accompanied by pharmacokinetic studies and model development.
7-4
-------
8. REFERENCES
Aberg, B., L. Ekman, R. Falk, U. Greitz, G. Persson and J. Snihs. 1969. Metabolism of methyl mercury
(203Hg) compounds in man: Excretion and distribution. Arch. Environ. Health 19:478-484.
Abramowicz, M. and H.L. Barnett. 1970. Sex ratio of infant mortality: Trends of change. Amer JDis
Child. 119:314-315.
Adams, C., D. Ziegler and J. Lin. 1983. Mercury intoxication simulating amyotrophic lateral sclerosis.
J. Am. Med. Assoc. 250:642-643.
Afonso, J. and R. deAlvarez. 1960. Effects of mercury on human gestation. Am. J. Obstet. Gynecol.
80:145-154.
Agency for Toxic Substabnces and Disease Registry (ATSDR). 1994. Toxicological Profile for Mercury
(Update). U.S Department of Health and Human Services, Public Health Service. TP-93/10.
Agency for Toxic Substabnces and Disease Registry (ATSDR). 1997. Toxicological Profile for Mercury
~ Draft for Public Comment. U.S Department of Health and Human Services, Public Health Service.
Agrawal, R. and J.P.N. Chansouria. 1989. Chronic effects of mercuric chloride ingestion on rat
adrenocortical function. Bull. Environ. Contain. Toxicol. 43(3):481-484.
Ahlbom, A., S. Norell, Y. Rodvall and M. Nylander. 1986. Dentists, dental nurses, and brain tumours.
Br. Med. J. 292:662.
Airey, D. 1983. Mercury in hair due to environment and diet: a review. Environ. Health Perspect.
52:303-316.
Albers, J.W., L.R. Kallenbach, L.J. Fine, et al. 1988. Neurological abnormalities associated with remote
occupational elemental mercury exposure. Ann. Neurol. 24(5):651-659.
Alcser, K.H., K.A. Birx and L.J. Fine. 1989. Occupational mercury exposure and male reproductive
health. Am. J. Ind. Med. 15(5):517-529.
Ali, S.F., C.P. LeBel and S.C. Bondy. 1992. Reactive oxygen species formation as a biomarker of
methyl mercury and trimethyltin neurotoxicity. Neurotoxicology. 13(3):637-648.
Allen, B.C., R.J. Kavlock, C.A. Kimmel and E.M. Faustman. 1994a. Dose-response assessment for
developmental toxicity II. Comparison of generic benchmark dose estimates with no observed effect
levels. Fund. Appl. Toxicol. 23:487-495.
Allen, B.C., R.J. Kavlock, C.A. Kimmel and E.M. Faustman. 1994b. Dose-response assessment for
developmental toxicity III. Statistical models. Fund. Appl. Toxicol. 23:496-509.
Al-Mufti, A.W., J.F. Copplestone, G. Kazanitzis, et al. 1976. Epidemiology of organomercury
poisoning in Iraq: I. Incidence in a defined area and relationship to the eating of contaminated bread.
Bull. World Health Org. 53(suppl):23-36.
8-1
-------
Al-Saleem, T., Clinical Committee on Mercury Poisoning. 1976. Levels of mercury and pathologic
changes in patients with organomercury poisoning. Bull. WHO 53(suppl):99-104.
Al-Shahristani, H., and K.M. Shihab. 1974. Variation of biological half-life of methyl mercury in man.
Arch. Environ. Health. 28:342-344.
Amandus, H. and J. Costello. 1991. Silicosis and lung cancer in U.S. metal miners. Arch. Environ.
Health 46(2):82-89.
Amin-Zaki, L., S. Elhassani, M.A. Majeed, et al. 1974. Intra-uterine methyl mercury poisoning in Iraq.
Pediatrics. 84:587-595.
Amin-Zaki, L., S. Elhassani, M.A. Majeed, T.W. Clarkson, R.A. Doherty, M.R. Greenwood and T.
Giovanoli-Jakubczak. 1976. Perinatal methyl mercury poisoning in Iraq. Am. J. Dis. Child. 130:1070-
1076.
Amin-Zaki, L., M.A. Majeed, T.W. Clarkson and M.R. Greenwood. 1978. Methyl mercury poisoning in
Iraqi children: Clinical observations over two years. Br. Med. J. 11:613-616.
Anderson, O. 1983. Effects of coal combustion products and metal compounds on sister chromatid
exchange (SCE) in a macrophagelike cell line. Envrion. Health Perspect. 47:239-253.
Andres, P. 1984. IgA-IgG disease in the intestine of Brown Norway rats ingesting mercuric chloride.
Clin. Immunol. Immunopathol. 30:488-494
Anwar, W.A. and M.S. Gabal. 1991. Cytogenetic study in workers occupationally exposed to mercury
fulminate. Mutagenesis. 6(3): 189-192.
Arito, H. and M. Takahashi. 1991. Effect of methyl mercury on sleep patterns in the rat. In: Advances
in Mercury Toxicology, T. Suzuki, N. Imura, T.W. Clarkson, Ed. Plenum Press, New York, NY. p.
381-394.
Arlauskas, A., RS.U. Baker, A.M. Bonin, R.K. Tandon, P.T. Crisp and J. Ellis. 1985. Mutagenicity of
metal ions in bacteria. Environ. Res. 36:379-388.
Aronow, R., C. Cubbage, R. Wisner, et al. 1990. Mercury exposure from interior latex paint. MMWR.
39(8):125-126.
Aschner, M. and J.L. Aschner. 1990. Mercury neurotoxicity: Mechanisms of blood-brain barrier
transport. Neurosci. Biobehav. Rev. 14(2): 169-176.
Ashe, W., E. Largent, F. Dutra, et al. 1953. Behavior of mercury in the animal organism following
inhalation. Arch. Ind. Hyg. Occup. Med. 17:19-43.
Atchison, W.D. 1986. Extracellular calcium-dependent and -independent effects of methyl mercury on
spontaneous and potassium-evoked release of acetylcholine at the neuromuscular junction. J. Pharm.
Exp. Therap. 237(2):672-680.
8-2
-------
Atchison, W.D. and M.F. Hare. 1994. Mechanisms of methyl mercury-induced neurotoxicity. FASEB
J. 8:622-629.
Atchison, W.D., A.W. Clark and T. Narahashi. 1984. Presynaptic effects of methyl mercury at the
mammalian neuromuscular junction. In: Cellular and Molecular Neurotoxicology, T. Narahashi, Ed.
Raven Press, New York, NY. p. 23-43.
Atchison, W.D., U. Joshi and J.E. Thornburg. 1986. Irreversible suppression of calcium entry into nerve
terminals by methyl mercury. J. Pharmacol. Exp. Therap. 238(2):618-624.
Aten, J., A. Veninga, E. Deheer, et al. 1991. Susceptibility to the induction of either autoimmunity or
immunosuppression by mercuric-chloride is related to the major histocompatibility complex class-II
haplotype. Europ. J. Immunol. 21(3):611-616.
Baggett, J.M. and W.O. Berndt. 1984. Interaction of potassium dichromate with the nephrotoxins,
mercuric chloride and citrinin. Toxicology. 33:157-169.
Bakir, F., S.F. Damluji, L. Amin-Zaki, et al. 1973. Methyl mercury poisoning in Iraq. Science.
181:230-241.
Bakir, F., H. Rustam, S. Tikriti, et al. 1980. Clinical and epidemiological aspects of methyl mercury
poisoning. Postgrad. Med. J. 56:1-10.
Ballatori, N. and T. Clarkson. 1982. Developmental changes in the biliary excretion of methyl mercury
and glutathione. Science. 216(2):61-63.
Baranski, B. and I. Szymczyk. 1973. [Effects of mercury vapor upon reproductive functions of female
white rats.] Med. Pr. 24:248. (Polish)
Barber, R.E. 1978. Inorganic mercury intoxication reminiscent of amyotrophic lateral sclerosis. J.
Occup. Med. 20:667-669.
Barber, T. and G. Wallis. 1986. Correction of urinary mercury concentration by specific gravity,
osmolality, and creatinine. J. Occup. Med. 28(5):354-359.
Barregard, L., B. Hultberg, A. Schutz, et al. 1988. Enzymuria in workers exposed to inorganic mercury.
Int. Arch. Occup. Environ. Health. 61(l-2):65-69.
Barregard, L., G. Sallsten and B. Jarvholm. 1990. Mortality and cancer incidence in chloralkali workers
exposed to inorganic mercury. Br. J. Ind. Med. 47(2):99-104.
Barregard, L., B. Hogstedt, A. Schutz, et al. 1991. Effects of occupational exposure to mercury vapor on
lymphocyte micronuclei. Scand. J. Work Environ. Health. 17(4):263-268.
Barregard, L., G. Sallsten, A. Schutz, et al. 1992. Kinetics of mercury in blood and urine after brief
occupational exposure. Arch. Environ. Health. 47(3): 176-184.
8-3
-------
Barrett, J.N., D. Botz and D.B. Chang. 1974. Block of neuromuscular transmission by methyl mercury.
In: Early Detection of Occupational Hazards: Proceedings of Workshop, C. Xintra, B.L. Johnson and I.
de Groot, Ed. U.S. Department of Health, Education and Welfare, National Institute for Occupational
Safety and Health, Cincinnati, OH. 5:277-287.
Bellinger, D., A. Leviton, C. Waternaux, H. Needleman, and M. Rabinowitz. 1989. Low-level lead
exposure, social class, and infant development. Neurotoxicology and Teratology, 10: 497-503.
Bencko, V., V. Wagner, M. Wagnerova, et al. 1990. Immunological profiles in workers occupationally
exposed to inorganic mercury. J. Hyg. Epidemiol. Microbiol. Immunol. 34(1):9-15.
Berg, G. and B. Smith. 1983. Distribution and elimination of ingested mercuric oxide in mice. J. Am.
Coll. Toxicol. 2(4):307-317.
Berglund, A. 1990. Estimation by a 24-hour study of the daily dose of intra-oral mercury vapor inhaled
after release from dental amalgam. J. Dent. Res. 69(10):1646-1651.
Berglund, F., M. Berlin, G. Birke, R. Cederlof, U. von Euler, L. Friberg, B. Holmsteadt, E. Jonsson et al.
1971. Methyl mercury in fish: A toxicologic-epidemiologic evaluation of risks. Report from an expert
group. Nordisk Hygienisk Tidskrift, Stockholm. Suppl. 4:19-364.
Berlin, M. 1986. Mercury. In: Handbook on the Toxicology of Metals, 2nd ed., L. Friberg, G.R.
Nordberg, V.B. Vouk, Ed. Elsevier Press, New York, NY.
Berlin, M., J. Fazackerley and C. Nordbery. 1969a. The uptake of mercury in the brains of mammals
exposed to mercury vapor and to mercuric salts. Arch. Environ. Health. 18:719-729.
Berlin, M., G. Nordberg and F. Serenius. 1969b. On the site and mechanism of mercury vapor
resorption in the lung. Arch. Environ. Health. 18:42-50.
Berlin, M., C. Blomstrand, C.A. Grand et al. 1975. Tritiated methyl mercury in the brain of squirrel
monkeys. Arch. Environ. Health. 30:591-597.
Bernard, S. and P. Purdue. 1984. Metabolic models for methyl and inorganic mercury. Health Phys.
46(3):695-699.
Bernard, A.M., H.R. Roels, J. M. Foldart and R.L. Lauwerys. 1980. Assessment of renal function of
workers exposed to inorganic lead, cadmium or mercury vapor. J. Occup. Med. 22(11): 741-750.
Bernard, A.M., H.R. Roels, J.M. Foidart, et al. 1987. Search for anti-laminin antibodies in the serum of
workers exposed to cadmium, mercury vapor or lead. Arch. Occup. Environ. Health. 59:303-309.
Bernard, A.M., C. Collette and R. Lauwerys. 1992. Renal effects of in utero exposure to mercuric
chloride in rats. Arch. Toxicol. (Germany). 66(7):508-513.
Bernaudin, J.F., E. Druet, P. Druet, et al. 1981. Inhalation or ingestion of organic or inorganic
mercurials produces auto-immune disease in rats. Clin. Immunol. Immunopathol. 20:129-135.
8-4
-------
Best CH. 1961. The Physiological Basis of Medical Practice. Baltimore, p. 19 and 29.
Betti, C., T. Davini and R. Barale. 1992. Genotoxic activity of methyl mercury chloride and dimethyl
mercury in human lymphocytes. Mutat. Res. 281(4):255-260.
Bidstrup, P., J. Bonnell, D.G. Harvey, et al. 1951. Chronic mercury poisoning in men repairing direct
current meters. Lancet:856-861.
Birke, G., A. Johnels, L-O. Plantin, et al. 1972. Studies on humans exposed to methyl mercury through
fish consumption. Arch. Environ. Health. 25:77-91.
Blakley, B.R. 1984. Enhancement of urethane-induced adenoma formation in Swiss mice exposed to
methyl mercury. Can. J. Comp. Med. 48:299-302.
Blakley, B.R., C.S. Sisodia and T.K. Mukkur. 1980. The effect of methyl mercury, tetraethyl lead, and
sodium arsenate on the humoral immune response in mice. Toxicol. Appl. Pharmacol. 52:245-254.
Blondell, J.M. and S.M. Knott. 1993. Risk analysis for phenylmercuric acetate in indoor latex house
paint. ACS Symposium Series 522:307-317.
Bluhm, R.E., R.G. Bobbitt, L.W. Welch, et al. 1992a. Elemental mercury vapour toxicity, treatment, and
prognosis after acute, intensive exposure in chloralkali plant workers: Part I. History,
neuropsychological findings and chelator effects. Hum. Exp. Toxicol. 11(3):201-210.
Bluhm, R.E., J.A. Breyer, R.G. Bobbitt, et al. 1992b. Elemental mercury vapour toxicity, treatment, and
prognosis after acute, intensive exposure in chloralkali plant workers: Part II. Hyperchloraemia and
genitourinary symptoms. Hum. Exp. Toxicol. 11(3):211-215.
Bornhausen, M., M.R. Musch and H. Greim. 1980. Operant behavior performance changes in rats after
prenatal methyl mercury exposure. Toxicol. Appl. Pharmacol. 56:305-316.
Bornmann, G., G. Henke, H. Alfes, et al. 1970. Concerning the enteral absorption of metallic mercury.
Arch. Toxicol. 26:203-209.
Bornstein, R.,A., and J.D. Matazarro. 1984. Relationship of sex and the effects of unilateral lesions on
the Wechsler Scales: Further considerations. JNervMent
Boscolo, P., M. Carmignani, G. Giuliano and P. Preziosi. 1989. Peripheral catecholaminergic
mechanisms and baroreflex pathways are involved in vascular and cardiac effects of long-term exposure
to inorganic mercury in rats. In: Advances in Vascular Pathology, A. Strano and S. Novo, Ed. Elsevier
Science Publishers, Amsterdam, p. 1061-1066.
Brodsky, J.B., E.N. Cohen, C. Whitcher, et al. 1985. Occupational exposure to mercury in dentistry and
pregnancy outcome. J. Am. Dent. Assoc. 111(5):779-780.
Brune, D., G.R. Nordberg, O. Vesterberg et al. 1991. A review of normal concentration of mercury in
human blood. Sci. Total Environ. 100:235-282.
8-5
-------
Buchet, J., H. Roels, A. Bernard, et al. 1980. Assessment of renal function of workers exposed to
inorganic lead, cadmium, or mercury vapor. J. Occup. Med. 22:741-750.
Buelke-Sam, J., C.A. Kimmel, J. Adams, et al. 1985. Collaborative behavioral teratology study:
Results. Neurobehav. Toxicol. Teratol. 7:591-624.
Buiatti, E., D. Kriebel, M. Geddes, et al. 1985. A case control study of lung cancer in Florence, Italy: I.
Occupational risk factors. J. Epidemiol. Comm. Health. 39:244-250.
Bunn, W.B., C.M. McGill, T.E. Barber, J.W. Cromer and L.J. Goldwater. 1986. Mercury exposure in
chloralkali plants. Am. Ind. Hyg. Assoc. J. 47:249-254.
Burbacher, T.M., C. Monnett, K.S. Grant, et al. 1984. Methyl mercury exposure and reproductive
dysfunction in the nonhuman primate. Toxicol. Appl. Pharmacol. 75:18-24.
Burbacher, T.M., K. S. Grant and N.K. Mottet. 1986. Retarded object permanence development in
methyl mercury exposedMacaca fasicularis infants. Dev. Psychobiol. 22:771-776.
Burbacher, T.M., M.K. Mohamed and N.K. Mottett. 1988. Methyl mercury effects on reproduction and
offspring size at birth. Reprod. Toxicol. l(4):267-278.
Burbacher, T.M., G. Sackett, and N. Mottet. 1990. Methyl mercury effects on the social behavior of
Macaca fascicularis infants. Neurotoxicol. Teratol. 12(1):65-71.
Byrne, A.R. and L. Kosta. 1974. Simultaneous neutron-activation determination of selenium and
mercury in biological samples by volatilization. Talanta. 21:1083-1090.
Campbell,!. 1948. Acute mercurial poisoning by inhalation of metallic vapor in an infant. Can. Med.
Assoc. J. 58:72-75.
Cantoni, O. and M. Costa. 1983. Correlations of DNA strand breaks and their repair with cell survival
following acute exposure to mercury(II) and X-rays. Mol. Pharmacol. 24:84-89.
Cantoni, O., R.M. Evans and M. Costa. 1982. Similarity in the acute cytotoxic response of mammalian
cells to mercury (II) and X-rays: DNA damage and glutathione depletion. Biochem. Biophys. Res.
Commun. 108:614-619.
Cantoni, O., N.T. Christie, S.H. Robison, et al. 1984a. Characterization of DNA lesions produced by
HgCl2 in cell culture systems. Chem. Biol. Interact. 49:209-224.
Cantoni, O., N.T. Christie, A. Swann, et al. 1984b. Mechanism of HgCl2 cytotoxicity in cultured
mammalian cells. Mol. Pharmacol. 26:360-368.
Cappon, C.J. and J.C. Smith. 1982. Chemical form and distribution of mercury and selenium in edible
seafood. J. Anal. Toxicol. 6:10-21.
Cardenas, A., H. Roels, A.M. Bernard, et al. 1993. Markers of early renal changes induced by industrial
pollutants. I. Appl.ication to workers exposed to mercury vapour. Br. J. Ind. Med. 50(1): 17-27.
8-6
-------
Carmignani, M., P. Boscolo and P. Preziosi. 1989. Renal ultrastructural alterations and cardiovascular
functional changes in rats exposed to mercuric chloride. Arch. Toxicol. (Suppl. 13):353-356.
Carmignani, M., P. Boscolo, L. Artese, et al. 1992. Renal mechanisms in the cardiovascular effects of
chronic exposure to inorganic mercury in rats. Br. J. Ind. Med. 49(4):226-232.
Carrico, L.C. 1985. Mercury. In: Mineral Facts and Problems. Bulletin 675. U.S. Department of the
Interior, Bureau of Mines, Washington, DC. p. 499-508.
Cember, H. 1962. The influence of the size of the dose on the distribution and elimination of inorganic
mercury, Hg(NO3)2, in the rat. Am. Ind. Health. Assoc. J. 23:304-313.
Cernichiari, E., R. Brewer, G Meyeres, D. Marsh, L. Lapham, C. Cox, C. Shamlaye, M. Berlin, P.
Davidson, and T. Clarkson. 1995. Monitoring methylmercury during pregnancy: Maternal hair predicts
fetal brain exposure. NeuroToxicology 16:705-710.
Chaffin, D., B. Dinman, J. Miller, et al. 1973. An evaluation of the effects of chronic mercury exposure
on EMG and psychomotor functions. U.S. Department of Health and Human Services, National Institute
of Occupational Safety and Health, Washington, DC. Document no. HSM-099-71-62.
Chang, L. and H.A. Hartmann. 1972. Blood-brain barrier dysfunction in experimental mercury
intoxication. Acta. Neuropathol. (Berlin) 21:179-184.
Chang, L.W., S. Yamaguchi and J.A.W. Dudley. 1974. Neurological changes in cats following
long-term diet of mercury contaminated tuna. Acta. Neuropathol. (Berlin) 27:171-176.
Charbonneau, S.M., I. Munro and E. Nera. 1976. Chronic toxicity of methyl mercury in the adult cat.
Toxicology. 5:337-340.
Chen, W.J., R.L. Body and N.K. Mottet. 1983. Biochemical and morphological studies of monkeys
chronically exposed to methyl mercury. J. Toxicol. Environ. Health. 12: 407-416.
Cherian, M.G., J.G. Hursh, T.W. Clarkson, et al. 1978. Radioactive mercury distribution in biological
fluids and excretion in human subjects after inhalation of mercury vapor. Arch. Environ. Health.
33:190-114.
Cheung, M.K. and M.A. Verity. 1985. Experimental methyl mercury neurotoxicity: Locus of mercurial
inhibition of brain protein synthesis in vivo and in vitro. J. Neurochem. 44: 1799-1808.
Choi, B.H. 1986. Methyl mercury poisoning in the developing nervous system. I. Pattern of neuronal
migration in the cerebral cortex. Neurotox. 7:591-600.
Choi, C.M., L.W. Lapham, L. Amin-Zaki, et al. 1978. Abnormal neuronal migration, deranged cerebral
cortical organization and diffuse white matter astrocytosis of human fetal brain: A major effect of
methyl mercury poisoning in utero. J. Neuropathol. Exp. Neurol. 37:719-732.
Christie, N.T., O. Cantoni, R.M. Evans, et al. 1984. Use of mammalian DNA repair-deficient mutants to
assess the effects of toxic metal compounds on DNA. Biochem. Pharmacol. 33:1661-1670.
8-7
-------
Christie, N.T., O. Cantoni, M. Sugiyama, et al. 1985. Differences in the effects of Hg(II) on DNA repair
induced in Chinese hamster ovary cells by ultraviolet or X-rays. Mol. Pharmacol. 29:173-178.
Cinca, I., I. Dumitrescu, P. Onaca, et al. 1979. Accidental ethyl mercury poisoning with nervous system,
skeletal muscle, and myocardium injury. J. Neurol. Neurosurg. Psychiat. 43:143-149.
Clarkson, T.W. 1972. The Pharmacology of mercury compounds. Ann. Rev. Pharmacol. 12:375-406.
Clarkson, T.W. 1989. Mercury. J. Am. Coll. Toxicol. 8(7): 1291-1296.
Clarkson, T.W. 1992. Mercury: Major issues in environmental health. Environ. Health Pespect.
100:31-38.
Clarkson, T.W. 1993. Molecular and ionic mimicry of toxic metals. Annu. Rev. Pharmacol. Toxicol.
32:545-571.
Clarkson, T.W. and A. Rothstein. 1964. The excretion of volatile mercury by rats injected with mercuric
salts. Health Phys. 10:1115-1121.
Clarkson, T.W., L. Magos and M.R. Greenwood. 1972. The transport of elemental mercury into fetal
tissues. Biol. Neonate. 21:239-244.
Clarkson, T.W., L. Amin-Zaki and S. Al-Tikriti. 1976. An outbreak of methyl mercury poisoning due to
consumption of contaminated grain. Fed. Proc. 35:2395-2399.
Clarkson, T.W., J.B. Hursh, P.R. Sager, et al. 1988. Biological Monitoring of Toxic Metals. Plenum
Press, New York, NY. p. 199-246.
Cordier, S., F. Deplan, L. Mandereau, et al. 1991. Paternal exposure to mercury and spontaneous
abortions. Br. J. Ind. Med. 48(6):375-381.
Costa, M., N.T. Christie, O. Cantoni, et al. 1991. DNA damage by mercury compounds: An overview.
In: Advances in Mercury Toxicology, T. Suzuki, N. Imura, T.W. Clarkson, Ed. Plenum Press, New
York, NY. p. 255-273.
Cox, C., T.W. Clarkson, D.E. Marsh, et al. 1989. Dose-response analysis of infants prenatally exposed
to methyl mercury: An Appl.ication of a single compartment model to single-strand hair analysis.
Environ. Res. 49(2):318-332.
Cox, C., D. Marsh, G. Meyers, and T. Clarkson. 1995. Analysis of data on delayed development from
the 1971-72 outbreak of methylmercury poisoning in Iraq: assessment of influential points.
NeuroToxicology 16:727-730.
Cragle, D., D. Hollis, J. Qualters, et al. 1984. A mortality study of men exposed to elemental mercury.
J. Occup. Med. 26:817-821.
Crump, K., J. Viren, A. Silvers, H. Clewell, J. Gearhart, and A. Shipp. 1995. Reanalysis of dose-
response data from the Iraqi methylmercury poisoning episode. Risk Analysis 15:523-5 32.
-------
Cuomo, V., L. Ambrosi, Z. Annau, et al. 1984. Behavioral and neurocChemical changes in offspring of
rats exposed to methyl mercury during gestation. Neurobehav. Toxicol. Teratol. 6:249-254.
Cuvin-Aralar, M.L.A. and R.W. Furness. 1991. Mercury and selenium interaction: A review.
EcoToxicol. Environ. Saf 21(3):348-364.
Danielsson, B.R.G., A. Fredriksson, L. Dahlgren, A. Teiling Gardlund, L. Olsson, L. Dencker and T.
Archer. 1993. Behavioural effects of prenatal metallic mercury inhalation exposure in rats.
Neurotoxicol. Teratol. 15:391-396.
Daston, G.P., B.F. Rehnberg, L.L Hall, et al. 1986. Toxicity of mercuric chloride to the developing rat
kidney: III. Distribution and elimination of mercury during postnatal maturation. Toxicol. Appl.
Pharmacol. 85:39-48.
Davidson, P.W., G. Myers, C.C. Cox, J. Sloane-Reeves, D. Marsh, C. Shamlaye, and T. Clarkson. 1994.
Prenatal exposure to methyl mercury through maternal fish consumption: Neurodevelopment of
Seychelles child development cohort at 19 and 29 months of age. Abstract presented at the Twelfth
International Neurotoxicology Conf. "Neurotoxicity of Mercury: Indicators and Effect of Low-Level
Exposure". Hot Springs, AK. November 1.
Davidson, P.W., G. Myers, C.C. Cox,C.F. Shamlaye, D.O.Marsh, M.A.Tanner, M. Berlin, J. Sloane-
Reeves, E. Chernichiari,, O. Choisy, A. Choi and T.W. Clarkson. 1995. Longitudinal neurodevelopment
study of Seychellois children following in utero exposure to methylemrcury form maternal fish ingestion:
outcomes at 19 and 29 months. NeuroToxicology 16:677-688.
Davis, L.E., J.R. Wands, S.A. Weiss, et al. 1974. Central nervous system intoxication from mercurous
chloride laxatives ~ quantitative, histochemical and ultrastructure studies. Arch. Neurol. 30:428-431.
Dietrich, K.N., Krafft, K.M., Bornschein, R.L., Hammond, P.B., Berger, O., Succop, P.A., and M. Bier
1987. Low level fetal lead exposure effect on neurobehavioral development in early infancy. Pediatrics
80: 721-730.
Dencker, L., B. Danielsson, A. Khayat, etal. 1983. Deposition of metals in the embryo and fetus. In:
Reproductive and Developmental Toxicity of Metals, T.W. Clarkson, G.G. Nordberg, P.R. Sager, Ed.
Plenum Press, New York, NY. p. 607-631.
Denny, M.F., M.F. Hare and W.D. Atchison. 1993. Methyl mercury alters intrasynaptosomal
concentrations of endogenous polyvalent cations. Toxicol. Appl. Pharmacol. 122:222-232.
Derobert, L. and S. Tara. 1950. [Mercury intoxication in pregnant women.] Ann. Med. Leg.
30:222-225. (French)
De Rosis, F., S.P Anastasio, L. Selvaggi, et al. 1985. Female reproductive health in two lamp factories:
Effects of exposure to inorganic mercury vapour and stress factors. Br. J. Ind. Med. 42:488-494.
DHHS. 1993. Dental amalgam: A scientific review and recommended public health service strategy for
research, education and regulation. Department of Health and Human Services, Public Health Service,
Washington, DC.
8-9
-------
Dieter, M.P., M.I. Luster, G.A. Boorman, et al. 1983. Immunological and biochemical responses in mice
treated with mercuric chloride. Toxicol. Appl. Pharmacol. 68:218-228.
Dieter, M.P., G.A. Boorman, C.W. Jameson, et al. 1992. Development of renal toxicity in F344 rats
gavaged with mercuric-chloride for 2 weeks, or 2, 4, 6, 15, and 24 months. J. Toxicol. Environ. Health.
36(4):319-340.
DiSimplicio, P., M. Gorelli, R. Vignani, et al. 1993. The differential modulation of the enzymes of
glutathione metabolism. Biol. Trace Elem. Res. 36(2): 167-181.
Druckrey, H., H. Hamperl and D. Schmahl. 1957. Cancerogene wirkung von metallischem. Quecksilber
nach intraperitonealer gabe bei ratten. [German]
Druet, P., E. Druet, F. Potdevin, et al. 1978. Immune type glomerulonephritis induced by HgC12 in the
Brown Norway rat. Ann. Immunol. 129C:777-792.
Dunn, J.D., T.W. Clarkson and L. Magos. 198 la. Ethanol reveals novel mercury detoxification step in
tissues. Science. 213:1123-1125.
Dunn, J.D., T.W. Clarkson and L. Magos. 1981b. Interaction of ethanol and inorganic mercury:
Generation of mercury vapor in vivo. J. Pharmacol. Exp. Ther. 216:19-23.
Dyer, R.S., C.U. Eccias and Z. Annau. 1978. Evoked potential alterations following prenatal methyl
mercury exposure. Pharmacol. Biochem. Behav. 8:137-141.
Ehrenberg, R.L., R.L. Vogt, A.B. Smith, et al. 1991. Effects of elemental mercury exposure at a
thermometer plant. Am. J. Ind. Med. 19(4):495-507.
Eisner, J., B. Hodel, K.E. Suter, et al. 1988. Detection limits of different approaches in behavioral
teratology, and correlation of effects with neurochemical parameters. Neurotoxicol. Teratol. 10:155-167.
Ellingsen, D., A. Andersen, H.P. Nordhagen, J. Efskind and H. Kjuus. 1992. Cancer incidence and
mortality among workers exposed to mercury in the Norwegian chloralkali industry. 8th International
Symposium on Epidemiology in Occupational Health, Paris, France, September 10-12, 1991. Rev.
Epidemiol. Sante Publique 40(Suppl. 1):S93-S94.
Ellingsen, D.G., R.I. Holland, Y. Thomassen, et al. 1993. Mercury and selenium in workers previously
exposed to mercury vapour at a chloralkili plant. Br. J. Ind. Med. 50(8):745-752.
Eisner, J. 1991. Tactile-kinesthetic system of rats as an animal model for minimal brain dysfunction.
Arch. Toxicol. 65(6):465-473.
Endo, T., S. Nakaya and R. Kimura. 1989. Factors involved in absorption of organic mercuric
compounds from rat small intestine: Comparative study with mercuric chloride in situ. Pharmacol.
Toxicol. 65(2): 128-135.
8-10
-------
Endo, T., S. Nakaya and R. Kimura. 1990. Mechanisms of absorption of inorganic mercury from rat
small intestine: III. Comparative absorption studies of inorganic mercuric compounds in vitro.
Pharmacol. Toxicol. 66(5):347-353.
Erfurth, E.M., A. Schutz, A. Nilsson, et al. 1990. Normal pituitary hormone response to thyrotropin and
gonadotropin releasing hormones in subjects exposed to elemental mercury vapour. Br. J. Ind. Med.
47:639-644.
Ericson, A. and B. Kallen. 1988. Pregnancy outcome in women working as dentists, dental assistants or
dental technicians. Arch. Occup. Environ. Health. 61(5):329-333.
Ericson, A. and B. Kallen. 1989. Pregnancy outcome in women working as dentists, dental assistants or
dental technicians. Int. Arch. Occup. Environ. Health. 61:329-333.
Eriksson, M., L. Hardell, N.O. Berg, T. Moller and O. Axelson. 1981. Soft-tissue sarcomas and
exposure to chemical substances: A case-control study. Br. J. Ind. Med. 38:27-33.
Eriksson, M., L. Hardell and H.-O. Adami. 1990. Exposure to dioxins as a risk factor for soft-tissue
sarcoma: A population-based case-control study. J. Natl. Cancer Inst. 82:486-490.
Evans, H.L., R. Garman and B. Weiss. 1977. Methyl mercury: Exposure duration and regional
distribution as determinants of neurotoxicity in nonhuman primates. Toxicol. Appl. Pharmacol.
41:15-33.
Fagala, G.E. and C.L. Wigg. 1992. Psychiatric manifestations of mercury poisoning. J. Am. Acad.
Child. Adolesc. Psychiat. 31(2):306-311.
Falconer, M.M., A. Vaillant, K.R. Reuhl, et al. 1994. The molecular basis of microtubule stability in
neurons. NEED JOURNAL. 15(1):109-122.
Fang, S.C. 1980. Comparative study of uptake and tissue distribution of methyl mercury in female rats
by inhalation and oral routes of administration. Bull. Environ. Contam. Toxicol. 24:65-72.
Farris, F.F., R.L. Dedrick, P.V. Allen, et al. 1993. Physiological model for the pharmacokinetics of
methyl mercury in the growing rat. Toxicol. Appl. Pharmacol. 119:74-90.
Faustman, E.M., B.C. Allen, R.J. Kavlock and C.A. Kimmel. 1994. Dose-response assessment for
developmental toxicity I. Characterization of database and determination of no observed effect levels.
Fund. Appl. Toxicol. 23:478-486.
Fawer, R.F., Y. DeRibaupierre, M. Guillemin, et al. 1983. Measurement of hand tremor induced by
industrial exposure to metallic mercury. Br. J. Ind. Med. 40:204-208.
FEDRIP. 1993. Federal Research in Progress. Dialog Information Service, Inc., Palo Alto, CA.
FEDRIP. 1994. Federal Research in Progress. Dialog Information Service, Inc., Palo Alto, CA.
8-11
-------
Fiskesjo, G. 1979. Two organic mercury compounds tested for mutagenicity in mammalian cells by use
of the cell line V 79-4. Hereditas. 90:103-110.
Fitzhugh, O.G., A.A. Nelson, E.P. Laug, et al. 1950. Chronic oral toxicities of mercuric-phenyl and
mercuric salts. Arch. Ind. Hyg. Occup. Med. 2:433-442.
Foa, V., L. Caimi. L. Amante, et al. 1976. Patterns of some lysosomal enzymes in the plasma and of
proteins in urine of wirkers exposed to inorganic mercury. Int. Arch. Occup. Environ. Health. 37:115-
124.
Forzi, M., M.G. Cassitto, C. Bulgheroni and V. Foa. 1978. Psychological measures in workers
occupationally exposed to mercury vapours: A validation study. In: Adverse Effects on Environmental
Chemicals and Psychotropic Drugs: Neurophysiological and Behaviour Tests, Vol. 2. H.J. Zimmerman,
Ed. Appleton-Century-Crofts, New York, NY. p. 165-171.
Foulds, D., K. Copeland and R. Franks. 1987. Mercury poisoning and acrodynia. Am. J. Dis. Child.
141:124-125.
Foulkes, E.G. and D. Bergman. 1993. Inorganic mercury absorption in mature and immature rat
jejunum: Transcellular and intercellular pathways in vivo and in everted sacs. Toxicol. Appl.
Pharmacol. 120:89-95.
Fowler, B.A. 1972. Ultrastructural evidence for neuropathy induced by long-term exposure to small
amounts of methyl mercury. Science. 175:780-781.
Fowler, B.A. and J.S. Woods. 1977. The transplacental toxicity of methyl mercury to fetal rat liver
mitochondria. Lab. Invest. 36:122-130.
Franchi, E., G. Loprieno, M. Ballardin, L. Petrozzi and L. Migliore. 1994. Cytogenetic monitoring of
fishermen with environmental mercury exposure. Mutat. Res. 320:23-29.
Francis, P., W. Birge, B. Roberts, et al. 1982. Mercury content of human hair: a survey of dental
personnel. J. Toxicol. Environ. Health. 10:667-672.
Fredriksson, A., L. Dahlgren, B. Danielsson, et al. 1992. Behavioral effects of neonatal metallic
mercury exposure in rats. Toxicology. 74(2-3): 151-160.
Fredriksson, A., T. Gardlund, K. Bergman, A. Oskarsson, B. Ohlin, B. Danielson and T. Archer. 1993.
Effects of maternal dietary supplementation with selenite on the postnatal development of rat offspring
exposed to methyl mercury in utero. J. Pharmacol. Toxicol. 72:377-382
Friberg, L., S. Hammarstrom and A. Nystrom. 1953. Kidney injury after chronic exposure to inorganic
mercury. Arch. Ind. Hyg. Occup. Med. 8:149-153.
Fukino, H., M. Hirai, K. Ideura, et al. 1992. Effect of the administration of mercuric chloride on zinc
deficiency in rats. J. Food. Hyg. Soc. Jpn. 33(l):31-38.
8-12
-------
Fukuda, K. 1971. Metallic mercury induced tremor in rabbits and mercury content of the central
nervous system. Br. J. Ind. Med. 28:308-311.
Fuyuta, M., T. Fujimoto and S. Hirata. 1978. Embrotoxic effects of methylmercuric chloride
administrated to mice and rats during organogenesis. Teratology. 18:353-366.
Fuyuta, M., T. Fujimoto E. Kiyofuji. 1979. Teratogenic effects of a single oral administration of
methylmercuric chloride in mice. Acta. Anat. 104:356-362.
Gage, J.C. 1961. The distribution and excretion of inhaled mercury vapour. Br. J. Ind. Med.
18:287-294.
Gale, T.F. 1974. Embryopathic effects of different routes of administration of mercuric acetate on the
hamster. Environ. Res. 8:207-213.
Gale, T.F. 1981. The embryotoxic response produced by inorganic mercury in different strains of
hamsters. Environ. Res. 24: 152-161.
Gale, T. and V. Perm. 1971. Embryopathic effects of mercuric salts. Life Sci. 10: 1341-1347.
Ganser, A.L. and D.A. Kirschner. 1985. The interaction of mercurials with myelin: Comparison of in
vitro and in vivo effects. Neurotoxicology. 6:63-78.
Ganther, H.E. 1978. Modification of methyl mercury toxicity and metabolism by selenium and vitamin
E: Possible mechanisms. Environ. Health Perspect. 25:71-76.
Ganther, H.E. 1980. Interactions of vitamin E and selenium with mercury and silver. Ann. NY Acad.
Sci. 355:212-226.
Garrett, N.E., R.J. Burriss Garrett, J.W. Archdeacon. 1972. Placental transmission of mercury to the
fetal rat. Toxicol. Appl. Pharmacol. 22:649-654.
Gearhart, J., H. Clewell, K. Crump, A. Shipp and A. Silvers. 1995. Pharmacokinetic dose estimates of
mercury in children and dose response curves of performance tests in a large epidemiological study.
Water, Air, and Soil Pollution 80:49-58.
Gerstner, H. and J. Huff. 1977. Clinical Toxicology of mercury. J. Toxicol. Environ. Health. 2:491-526.
Geyer, M.A., R.E. Butcher and K. Fite. 1985. A study of startle and locomotor activity in rats exposed
prenatally to methyl mercury. Neurobehav. Toxicol. Teratol. 7:759-765.
Ghosh, A.K., S. Sen, A. Sharma, et al. 1991. Effect of chlorophyllin on mercuric chloride-induced
clastogenicity in mice. Food Chem. Toxicol. 29(11):777-779.
Gilbert, S.G., T.M. Burbacher and B.C. Rice. 1993. Effects of in utero methyl mercury exposure on a
spatial delayed alternation task in monkeys. Toxicol. Appl. Pharmacol. 123:130-136.
8-13
-------
Girardi, G. and M.M. Elias. 1991. Effectiveness of n-acetylcysteine in protecting against mercuric
chloride-induced nephrotoxicity. Toxicology. 67(2): 155-164.
Gleason, M.N., R.E. Gosselin and H.C. Hodge. 1957. Clinical Toxicology of Commercial Products.
Williams and Wilkins Co., Baltimore, MD. p. 154.
Goldman, M. and P. Blackburn. 1979. The effect of mercuric chloride on thyroid function in the rat.
Toxicol. Appl. Pharmacol. 48:49-55.
Goldman, M., P. Druet and E. Gleichmann. 1991. Th2 cells in systemic autoimmunity: Insights from
allogenic diseases and chemically-induced autoimmunity. Immunol. Today. 12:223-227.
Gotelli, C.A., E. Astolfi, C. Cox, et al. 1985. Early biochemical effects of an organic mercury fungicide
on infants: "Dose makes the poison." Science. 277:638-640.
Grandjean, P. and P. Weihe. 1993. Neurobehavioral effects of intruterine mercury exposrue:potential
sources of bias. Envrion. Res. 61:176-183.
Grandjean, P., P. Weihe, P.J. Jorgensen, et al. 1992a. Impact of maternal seafood diet on fetal exposure
to mercury, selenium and lead. Arch. Environ. Health. 47:185-195.
Grandjean, P., Nielsen, P.J. Jorgensen, et al. 1992b. Reference intervals for trace elements in blood:
Significance of risk factors. Scand. J. Clin. Lab. Invest. 52:321-337.
Grandjean, P., P.J. Jorgensen and P. Weihe. 1994. Human milk as a source of methyl mercury exposure
in infants. Environ. Health Perspect. 102(l):74-77.
Grandjean, P., P. Weihe, and R.F. White. 1995. Milestone development in infants exposed to
methylmercury from human milk. Neurotoxicol. 16:27-34.
Grandjean, P., P. Weihe, R.F. White, F. Debes, S. Araki, K. Yokoyama, K.Murata, N. S orensen, R.
Dahl and P.J.Jorgensen. 1997. Cognitive deficit in 7-year-old children with prenatal exposure to
methylmercury. Neurotoxicol. and Teratol. 20:1-12.
Greenwood, M.R. 1985. Methyl mercury poisoning in Iraq. An epidemiological study of the 1971-1972
outbreak. J. Appl. Toxicol. 5: 148-159.
Greenwood, M.R., T.W. Clarkson, R.A. Doherty, et al. 1978. Blood clearance half-times in lactating
and nonlactating members of a population exposed to methyl mercury. Environ. Res. 16:48-54.
Gross, M.B., and W.C. Wilson. 1974. Minimal Brain Dysfunction: A Clinical Study of Incidence,
Diagnosis, and Treatment in over 1000 Children. New York: Brunner/Mazel
Gstraunthaler, G., W. Pfaller and P. Kotanko. 1983. Glutathione depletion and in vitro lipid
peroxidation in mercury or maleate-induced acute renal failure. Biochem. Pharmacol. 32:2969-2972.
Guidetti, P., A. Giacobazzi, P. Zanoli and M. Baraldi. 1992. Prenatal exposure of rats to methyl mercury
increased sensitivity of the GABA-benzodiazepine receptor functions. In: Metal Compounds in
8-14
-------
Environment and Life, 4. Interrelation Between Chemistry and Biology; Fourth IAEAC (International
Association of Environmental Analytical Chemistry) Workshop on Toxic Metal Compounds, les
Diablerets, Switzerland, March 4-8, 1991. xi plus 514 p. Science and Technology Letters: Northwood,
England, UK; Science Reviews Inc.: Wilmington, DE, USA. p. 365-371.
Gunderson, V.M., K.S. Grant, T.M. Burbacher, J.F. Pagan, 3rd and N.K. Mottet. 1986. The effect of
low-level prenatal methyl mercury exposure on visual recognition memory in infant crab-eating
macaques. Child Dev. 57(4): 1076-1083.
Gunderson, V.M., K.S. Grant-Webster, T.M. Burbacher, et al. 1988. Visual recognition memory deficits
in methyl mercury-exposedMacaca fascicularis infants. Neurotoxicol. Teratol. 10(4):373-379.
Haddad, J. and E. Stenberg. 1963. Bronchitis due to acute mercury inhalation: Report of two cases.
Am. Rev. Respir. Dis. 88:543-545.
Halbach, S. 1985. The octanol/water distribution of mercury compounds. Arch. Toxicol. 57:139-141.
Halbach, S. and T.W. Clarkson. 1978. Enzymatic oxidation of mercury vapor by erythrocytes.
Biochem. Biophys. Acta. 523:522-531.
Hall, L.L., P.V. Allen, H.L. Fisher, et al. 1994. The kinetics of intravenously-administered inorganic
mercury in humans. In: Kinetic Models of Trace Elements and Mineral Metabolism During
Development, K.M.S. Subramanian and M.E. Wastney, Ed. CRC Press, Boca Raton, FL. p. 1-21.
Hallee, T.J. 1969. Diffuse lung disease caused by inhalation of mercury vapor. Am. Rev. Respir. Dis.
99:430-436.
Hansen, J. 1988. Blood mercury concentrations in birthgiving Greenlandic women. Arctic. Med. Res.
47(1):175-178.
Hansen, J., E. Reske-Nielsen, O. Thorlacius-Ussing, et al. 1989. Distribution of dietary mercury in a
dog. Quantitation and localization of total mercury in organs and central nervous system. Sci. Total
Environ. 78:23-43.
Hansen, J.C., U. Tarp and J. Bohm. 1990. Prenatal exposure to methyl mercury among Greenlandic
polar Inuits. Arch. Environ. Health. 45(6):355-358.
Harada, H. 1978. Congenital Minamata disease: Intrauterine methyl mercury poisoning. Teratology.
18:285-288.
Harvey, P., M. Hamlin, and R. Kumar. 1984. Blood lead, behavior and intelligence test performance in
pre-school children. Sci Total Environ, 40: 45-60.
Hatch, W.R. and W.L. Ott. 1968. Determination of sub-microgram quantities of mercury by atomic
absorption spectrophotometry. Anal. Chem. 40(14):2085-2087.
8-15
-------
Heddle, J.R. and W.R. Bruce. 1977. Comparison of the micronucleus and sperm assay for mutagenicity
with the carcinogenic activities of 61 different agents. In: Origins of Human Cancer, H.H. Hiatt, J.D.
Watson, J.A. Winsten, Ed. Vol. 4. Cold Spring Harbor Conferences.
Heidam, L.Z. 1984. Spontaneous abortions among dental assistants, factory workers, painters, and
gardening workers: A follow up study. J. Epidemiol. Comm. Health. 38:149-155.
Hill, W. 1943. Two deaths from exposure to fumes of di-ethyl mercury. Can. J. Pub. Health. 34:158.
Hirano, M., K. Mitsumori, K. Maita, et al. 1986. Further carcinogenicity study on methyl mercury
chloride in ICRmice. Jap. J. Vet. Sci. 48(1): 127-135.
Hirayama, K., A. Yasutake and M. Inoue. 1987. Effect of sex hormones on the fate of methyl mercury
and on glutathione metabolism in mice. Biochem. Pharmacol. 36(12): 1919-1924.
Hollins, J., R. Willes, F. Bryce, et al. 1975. The whole body retention and tissue distribution of
[203Hg]methyl mercury in adult cats. Toxicol. Appl. Pharmacol. 33:438-449.
Hook, O., K-D. Lundgren and A. Swensson. 1954. On alkyl mercury poisoning. Acta. Med. Scand.
150:131-137.
Hoover, S.H., R. A. Hill, and T.A. Watson. 1997. Exposure of aboriginals in British Cvolun=mbia to
methylmercury in freshwater fish: a comparison ot reference doses and estimated thresholds. Human and
Ecological Risk Assessment 3:439-463.
Hoskins, B.B. and E.W. Hupp. 1978. Methyl mercury effects in rat hamster and squirrel monkey.
Environ. Res. 15: 5-19.
Howard, W., B. Leonard, W. Moody, et al. 1991. Induction of chromosome changes by metal
compounds in cultured CHO cells. Toxicol. Lett. 56(1-2): 179-186.
Hua, J., L. Pelletier, M. Berlin, et al. 1993. Autoimmune glomerulonephritis induced by mercury vapour
exposure in the Brown Norway rat. Toxicol. 79(2): 119-129.
Hudson, P.J., R.L. Vogt, J. Brondum, et al. 1987. Elemental mercury exposure among children of
thermometer plant workers. Pediatrics. 79:935-938.
Hughes, J.A. and Z. Annau. 1976. Postnatal behavioral effects in mice after prenatal exposure to
methylmercury. Pharmacol. Biochem. Behav. 4:385-391.
Hultman, P. and S. Enestrom. 1992. Dose-response studies in murine mercury-induced autoimmunity
and immune-complex disease. Toxicol. Appl. Pharmacol. 113(2): 199-208.
Hultman, P. and U. Johansson. 1991. Strain differences in the effect of mercury on murine cell-
mediated immune reactions. Food Chem. Toxicol. 29(9):633-638.
8-16
-------
Hultman, P., L.J. Bell, S. Enestrom, et al. 1992. Murine susceptibility to mercury: I. Autoantibody
profiles and systemic immune deposits in inbred, congenic, and intra-H-2 recombinant strains. Clin.
Immunol. Immunopathol. 65(2):98-109.
Hunter, D., R.R. Bomford, D.S. Russell. 1940. Poisoning by methyl mercury compounds. Quart.!.
Med. 9:193-213.
Hursh, J.B. 1985. Partition coefficients of mercury (203Hg) vapor between air and biological fluids. 1985.
J. Appl. Toxicol. 5:327-332.
Hursh, J.B, T.W. Clarkson, M.G. Cherian, J.J. Vostal and P.V. Mallie. 1976. Clearance of mercury
(Hg-197, Hg-203) vapor inhaled by human subjects. Arch. Environ. Health. 31:302-309.
Hursh, J.D., M.R. Greenwood, T.W. Clarkson, et al. 1980. The effect of ethanol on the fate of mercury
vapor inhaled by man. J. Pharmacol. Exp. Ther. 214:520-527.
Hursh, J.B., T.W. Clarkson, T.V. Nowak, R.C. Pabico, B.A. Mckenna, E. Miles and F.R. Gibb. 1985.
Prediction of kidney mercury content by isotope techniques. Kidney Int. 27:898-907.
Hursh, J.B., S.P. Sichak, and T.W. Clarkson. 1988. In vitro oxidation of mercury by the blood.
Pharmacol. Toxicol. 63:266-273.
Hursh, J.B., T.W. Clarkson, E.F. Miles, et al. 1989. Percutaneous absorption of mercury vapor by man.
Arch. Environ. Health. 44:120-127.
Igata, A. 1993. Epidemiological and clinical features of Minamata Disease. Environ. Res. 63: 157-169.
Ilback, N.G. 1991. Effects of methyl mercury exposure on spleen and blood natural-killer (NK) cell-
activity in the mouse. Toxicology. 67(1): 117-124.
Imura, N. and A. Naganuma. 1991. Possible mechanism of detoxifying effect of selenium on the
toxicity of mercury compounds. In: Advances in Mercury Toxicology, T. Suzuki, N. Imura and T.W.
Clark, Ed. Plenum Press, New York, NY. p. 275-288.
Inglis, J., and J.S. Lawson. 1981. Sex differences in the effects of unilateral brain damage on
intelligence. Science 212: 693-695.
Inouye, M. and Y. Kajiwara. 1988a. An attempt to assess the inheritable effect of methyl mercury
toxicity subsequent to prenatal exposure of mice. Bull. Environ. Contam. Toxicol. 41(4):508-514.
Inouye, M. and Y. Kajiwara. 1988b. Developmental disturbances of the fetal brain in guinea-pigs
caused by methyl mercury. Arch. Toxicol. 62(1): 15-21.
Inouye, M. and U. Murakami. 1975. Teratogenic effect of orally administered methylmercuric chloride
in rats and mice. Congenital Anom. 15:1-9.
Inouye, M., K. Murao, Y. Kajiwara. 1985. Behavioral and neuropathological effects of prenatal methyl
mercury exposure in mice. Neurobehav. Toxicol. Teratol. 7:227-232.
8-17
-------
Jaffe, K.M., D.B. Shurtleff, and W.O. Robertson. 1983. Survival after acute mercury vapor poisoning ~
Role of intensive supportive care. Am. J. Dis. Child. 137:749-751.
Jagiello, G. and J.S. Lin. 1973. An assessment of the effects of mercury on the meiosis of mouse ova.
Mutat. Res. 17:93-99.
Janicki, K., J. Dobrowolski and K. Drasnicki. 1987. Correlation between contamination of the rural
environment with mercury and occurrence of leukemia in men and cattle. Chemosphere. 16:253-257.
Jenssen, O. and C. Ramel. 1980. The micronucleus test as part of a short-term mutagenicity test
program for the prediction of carcinogenicity evaluated by 143 agents tested. Mutat. Res. 75:191-202.
Jonker, D., R.A. Woutersen, P.J. van Bladeren, et al. 1993. Subacute (4-wk) oral toxicity of a
combination of four nephrotoxins in rats: Comparison with the toxicity of the individual compounds.
Food Chem. Toxicol. (England) 31(2): 125-136.
Joselow, M.M., R. Ruiz and L. Goldwater. 1968. Absorption and excretion of mercury in man: XIV.
Salivary excretion of mercury and its relationship to blood and urine. Arch. Environ. Health. 17:35-38.
Jugo, S. 1976. Retention and distribution of 203HgCl2 in suckling and adult rats. Health Phys.
30:240-241.
Kajiwara, Y. and M. Inouye. 1986. Effects of methyl mercury and mercuric chloride on preimplantation
mouse embryos in vivo. Teratol. 33:231-237.
Kajiwara, Y. and M. Inouye. 1992. Inhibition of implantation caused by methyl mercury and mercuric
chloride in mouse embryos in vivo. Bull. Environ. Contam. Toxicol. 49(4):541-546.
Kanematsu, N., M. Kara and T. Kada. 1980. Rec assay and mutagenicity studies on metal compounds.
Mutat Res. 77:109-116.
Kang-Yum, E. and S.H. Oransky. 1992. Chinese patient medicine as a potential source of mercury
poisoning. Vet. Hum. Toxicol. 34(3):235-238.
Kanluen, S. and C.A. Gottlieb. 1991. A clinical pathologic study of four adult cases of acute mercury
inhalation toxicity. Arch. Pathol. Lab. Med. 115(1):56-60.
Karpathios, T., A. Zervoudakis, C. Thodoridis, et al. 1991. Mercury vapor poisoning associated with
hyperthyroidism in a child. Acta. Paediatr. Scand. 80(5):551-552.
Kavlock, R.J., T. Logsdon and J.A. Gray. 1993. Fetal development in the rat following disruption of the
maternal renal function during pregnancy. Teratol. 48(3):247-258.
Kazantzis, G., K. Schiller, A. Asscher, et al. 1962. Albuminuria and the nephrotic syndrome following
exposure to mercury and its compounds. Q. J. Med. 3:403-419.
Kerper, L.E., N. Ballatori and T.W. Clarkson. 1992. Methyl mercury transport across the blood-brain
barrier by an amino acid carrier. Am. J. Physiol. 262(5):R761-R765.
8-18
-------
Kershaw, T.G., T. W. Clarkson and P.H. Dhahir. 1980. The relationship between blood levels and dose
of methylmercury in man. Arch. Environ. Health. 35(l):28-36.
Khayat, A. and L. Dencker. 1982. Fetal uptake and distribution of metallic mercury vapor in the mouse:
Influence of ethanol and aminotriazole. Int. J. Biol. Res. Pregnancy:38-46.
Khayat, A. and L. Dencker. 1984a. Interactions between tellerium and mercury in murine lung and other
organs after metallic mercury inhalation: A comparison with selenium. Chem. Biol. Interact. 50:123-
133.
Khayat, A. and L. Dencker. 1984b. Organ and cellular distribution of inhaled metallic mercury in the rat
and Marmoset monkey (Callithrix jacchus): Influence of ethyl alcohol pretreatment. Acta. Pharmacol.
Toxicol. 55:145-152.
Khera, S. 1973. Reproductive capability of male rats and mice treated with methyl mercury. Toxicol.
Appl. Pharmacol. 24:167-177.
Khera, K.S. and S.A. Tabacova. 1973. Effects of methylmercuric chloride on the progeny of mice and
rats treated before or during gestation. Food Cosmet. Toxicol. 11:245-254.
Kinjo, Y., Y. Takizawa, Y. Shibata, M. Watanabe and H. Kato. 1995. Threshold dose for adults exposed
to methylmercury in Niigata Minmata Disease outbreak. Environ. Sci. 3:91-101.
Kishi, R., K. Hashimoto, S. Shimizu, et al. 1978. Behavioral changes and mercury concentrations in
tissues of rats exposed to mercury vapor. Toxicol. Appl. Pharmacol. 46:555-566.
Kishi, R., R. Doi, Y. Fukuchi, et al. 1993. Subjective symptoms and neurobehavioral performances of
ex-mercury miners at an average of 18 years after the cessation of chronic exposure to mercury vapor.
Environ. Res. 62(2):289-302.
Kitamura, S., L. Sumino, K. Hayakawa and T. Shibata. 1975. Mercury content in normal human tissues.
In: Studies on the Health Effects of Alkylmercury in Japan, T. Tsubaki, Ed. Environment Agency, Japan.
p. 20-27.
Kjellstrom, T., P. Kennedy, S. Wallis and C. Mantell. 1986a. Physical and mental development of
children with prenatal exposure to mercury from fish. Stage 1: Preliminary test at age 4. National
Swedish Environmental Protection Board, Report 3080 (Solna, Sweden).
Kjellstrom, T., P. Kennedy, S. Wallis et al. 1986b. Physical and mental development of children with
prenatal exposure to mercury from fish. Stage 2: Interviews and psychological tests at age 6. National
Swedish Environmental Protection Board, Report 3642 (Solna, Sweden).
Kjellstrom, T., P. Kennedy, S. Wallis, et al. 1989. Physical and mental development of children with
prenatal exposure to mercury from fish: Stage 2: Interviews and psychological tests at age 6. National
Swedish Environmental Protection Board, Report 3642. (Solna, Sweden).
Klaussen, C.D., M.O. Amdur, and J. Doull. 1986. Casarett and Doull Toxicology, 3rd ed. Macmillan
Publishing Co., New York. p. 605.
8-19
-------
Koizumi, A., A. Takehiko, M. Tsukada, M. Naruse and N. Saitoh. 1994. Mercury, not sulphur dioxide,
poisoning as cause of smelter disease in industrial plants producing sulphuric acid. Lancet. 343:1411-
1412.
Koller, L.D., J.H. Exon and B. Arbogast. 1977. Methyl mercury: Effect on serum enzymes and humoral
antibody. J. Toxicol. Environ. Health. 2:1115-1123.
Kosta, L., A.R. Byrne and V. Zelenko. 1975. Correlation between selenium and mercury in man and all
following exposure to inorganic mercury. Nature (London). 254:238-239.
Kostial, K., D. Kello, S. Jugo, et al. 1978. Influence of age on metal metabolism and toxicity. Environ.
Health Perspect. 25:81-86.
Kutsuna, M., Ed. 1968. Minamata disease: Study group of Minamata disease. Kumamota University,
Japan, p. 1-4.
Ladd, A.C., E. Zuskin, R. Valic, et al. 1966. Absorption and excretion of mercury in miners. J. Occup.
Med. 8:122-131.
Lamperti, A.A. and R.H. Printz. 1973. Effects of mercuric chloride on the reproductive cycle of the
female hamster. Biol. Reprod. 8:378-387.
Lansdown, R., W. Yule, M. Urbanowicz, and J. Hunter. 1986. The relationship between blood-lead
concentrations, intelligence, attainment and behavior in a school population: the second London study.
Int Arch Occup Environ Hlth 57: 225-235.
Landworth, S. 1987. Renal function in workers exposed to inorganic mercury. In: Occupational Health
in the Chemical Industry: Papers presented at the XXIIICOH Congress, Sidney, Australia, September
27-October 2, 1987. Copenhagen, World Health Organization, p. 237.
Langolf, G.D., D.B. Chaffin, R. Henderson, et al. 1978. Evaluation of workers exposed to elemental
mercury using quantitative tests. Am. Ind. Hyg. Assoc. J. 39:976-984.
Langworth, S., O. Almkvist, E. Soderman, et al. 1992a. Effects of occupational exposure to mercury
vapour on the central nervous system. Br. J. Ind. Med. 49(8):545-555.
Langworth, S., C.G. Blinder, K.G. Sundquist, et al. 1992b. Renal and immunological effects of
occupational exposure to inorganic mercury. Br. J. Ind. Med. 49(6):394-401.
Lauwerys, R., A. Bernard, H. Roels, et al. 1983. Anti-laminin antibodies in workers exposed to mercury
vapor. Toxicol. Lett. 17:113-116.
Lauwerys, R., H. Roels P. Genet, et al. 1985. Fertility of male workers exposed to mercury vapor or to
manganese dust: A questionnaire study. Am. J. Ind. Med. 7:171-176.
LeBel, C.P., S.F. Ali, M. McKee, et al. 1990. Organometal-induced increases in oxygen reactive
species: The potential of 2',7'-dichlorofluorescin diacetate as an index of neurotoxic damage. Toxicol.
Appl. Pharmacol. 104:17-34.
8-20
-------
LeBel, C.P., S.F. Ali and S.C. Bondy. 1992. Deferoxamine inhibits methyl mercury-induced increases in
reactive oxygen species formation in rat brain. Toxicol. Appl. Pharmacol. 112(1): 161-165.
Lebel, J., D. Mergler, M. Lucotte,. M/ Amorim, J. Dolbec, D. Miranda, G. Arantes, I. Rheault and P.
Pichet. 1996. Evidence of early nervous sustem dysfunction in Amazonian popuations exposed to low-
levels of methylmercury. NeuroToxicology. 17:157-168.
Lee, I.P. and R.L. Dixon. 1975. Effects of mercury on spermatogenesis studies by velocity
sedimentation cell separation and serial mating. J. Pharmacol. Exp. Ther. 194:171-181.
Levin, M., J. Jacobs, and P.G. Polos. 1988. Acute mercury poisoning and mercurial pneumonitis from
gold ore purification. Chest. 94: 554-556.
Levine, S.P., G.D. Cavender, G.D. Langolf, et al. 1982. Elemental mercury exposure: Peripheral
neurotoxicity. Br. J. Ind. Med. 39:136-139.
Leyshon, K. and A.J. Morgan. 1991. An integrated study of the morphological and gross-elemental
consequences of methyl mercury intoxication in rats, with particular attention on the cerebellum.
Scanning Microsc. 5(3):895-904.
Liang, Y.X., R.K. Sun, Y. Sun, et al. 1993. Psychological effects of low exposure to mercury vapor:
Application of a computer-administered neurobehavioral evaluation system. Environ. Res.
60(2):320-327.
Lilis, R., A. Miller and Y. Lerman. 1985. Acute mercury poisoning with severe chronic pulmonary
manifestations. Chest. 88:306-309.
Lipfert F.W., P.D. Moskowitz, V. Fthenakis and L. Saroff 1994. Probabilisitc assessment of health risks
of methylmercury from burning coal. NeuroToxicology 17:197-212.
Livardjani, F., M. Ledig, P. Kopp, et al. 1991. Lung and blood superoxide dismutase activity in mercury
vapor exposed rats: Effect of N-acetylcysteine treatment. Toxicology. 66(3):289-295.
Lok, E. 1983. The effect of weaning on blood, hair, fecal and urinary mercury after chronic ingestion of
methylmercuric chloride by infant monkeys. Toxicol. Lett. 15:147-152.
Lonky, E., J. Reihman, T. Darvill, J. Mather Sr., and H. Daly. 1996. Neonatl behaviorla assessment
scale performance I humnas influenced by maternal consumption of environmnetally contaminated Lake
Ontario fish. J. Great Lakes Res. 22:198-212.
Lovejoy, H.B., Z.G. Bell and T.R. Vizena. 1974. Mercury exposure evaluations and their correlation
with urine mercury excretion. J. Occup. Med. 15:590-591.
Mabille, V., H. Roels, P. Jacquet, et al. 1984. Cytogenetic examination of leukocytes of workers
exposed to mercury vapor. Int. Arch. Occup. Environ. Health. 53:257-260.
MacDonald, J.S. and R.D. Harbison. 1977. Methyl mercury-induced encephalopathy in mice. Toxicol.
Appl. Pharmacol. 39:195-205.
8-21
-------
Magos, L. 1973. Factors affecting the uptake and retention of mercury by kidneys in rats. In: Mercury,
Mercurials and Mercaptans, M.W. Miller and T.W. Clarkson, Ed. CC Thomas, Springfield, IL.
p. 67-186.
Magos, L. 1987. The absorption, distribution, and excretion of methyl mercury. In: The Toxicity of
Methyl Mercury, C.U. Eccles and Z. Annau, Ed. Johns Hopkins, Baltimore, MD. p. 24-44.
Magos, L. and T.W. Clarkson. 1972. Atomic absorption and determination of total, inorganic, and
organic mercury in blood. J. AOAC 55(5):966-971.
Magos, L. and M. Webb. 1979. Synergism and antagonism in the toxicology of mercury. In: The
Biogeochemistry of Mercury in the Environment, J.O. Nriagu, Ed. Elsevier/North Holland Biomedical
Press, New York, NY. p. 581-99.
Magos, L., A.W. Brown, S. Sparrow, et al. 1985. The comparative Toxicology of ethyl and
methylmercury. Arch. Toxicol. 57:260-267.
Magos, L., T.W. Clarkson and A.R. Hudson. 1989. The effects of dose of elemental mercury and first-
pass circulation time on exhalation and organ distribution of inorganic mercury in rats. Biochem.
Biophys. Acta. 991(l):85-89.
Mailhes, J.B. 1983. Methyl mercury effects on Syrian hamster metaphase II oocyte chromosomes.
Environ. Mutagen. 5:679-686.
Malm, O., W.C. Pfeiffer, C.M.M. Souza and R. Reuther. 1990. Mercury pollution due to gold-mining in
the Madera River Basin, Brazil. Ambio. 19:11-15.
Marsh, D.O., G.J. Myers, T.W. Clarkson, L. Amin-Zaki, S. Tikrit, M.A.Majeed and A.R. Dabbagh.
1981. Dose-response relationship for human fetal exposure to methyl mercury. Clin. Toxicol.
10:1311-1318.
Marsh, D.O., T.W. Clarkson, C. Cox, L. Amin-Zaki and S. Al-Tikriti. 1987. Fetal methyl mercury
poisoning: Relationship between concentration in single strands of maternal hair and child effects. Arch.
Neurol. 44: 1017-1022.
Marsh, D.O., M. Turner, and G. Myers. 1994. Prenatal exposures in Peruvian fishing villages. Abstract
presented at the Twelfth International Neurotoxicology Conf. "Neurotoxicity of Mercury: Indicators and
Effect of Low-Level Exposure". Hot Springs, AR. November 1.
Marsh, D., M. Turner, J. Crispin Smith, P. Allen andN. Richdale. 1995. Fetal methylmercury study in a
Peruvian fish-eating population. NeuroToxiclogy 116:717-726.
Marsh, D., T. Clarkson, G. Myeres, P.Davidson, C. Cox, E. Cernichiari, M. Tanner, W. Lednar, C.
Shamlaye, O. Choisy, C. Horeau and M. Berlin. 1995a. The Seychelles study of fetal methylmercury
exposure and child development: introduction. NeuroToxicology. 16:582-596.
Marzin, D.R., and H.V. Phi. 1985. Study of the mutagenicity of metal derivatives with Salmonella
typhimurium TA102. Mutat. Res. 155:49-51.
8-22
-------
Mathieson, P.W. 1992. Mercuric chloride-induced autoimmunity. Autoimmunity (Switzerland).
13(3):243-247.
Matsumoto, N. and A. Spindle. 1982. Sensitivity of early mouse embryos to methyl mercury toxicity.
Toxicol. Appl. Pharmacol. 64:108-117.
McAnulty, P.A., J.M. Tesh, A.L. Pritchard, et al. 1982. Effects of mercury on foetal development.
Teratology. 25:26A.
McFarland, R. and H. Reigel. 1978. Chronic mercury poisoning from a single brief exposure. J. Occup.
Med. 20:534-534.
McGregor, A.J. and H.J. Mason. 1991. Occupational mercury vapour exposure and testicular, pituitary
and thyroid endocrine function. Hum. Exp. Toxicol. 10(3): 199-203.
McKeown-Eyssen, G.E., J. Ruedy and A. Neims. 1983. Methyl mercury exposure in northern Quebec:
II. Neurologic findings in children. Am. J. Epidemiol. 118:470-479.
Melkonian, R. and D. Baker. 1988. Risks of industrial mercury exposure in pregnancy. Obstet.
Gynecol. Surv. 43(11):637-641.
Mengel, H. and O. Karlog. 1980. Studies on the interaction and distribution of selenite, mercuric,
methocyethyl mercuric and methylmercuric chloride in rats. Acta. Pharmacol. Toxicol. 46:25-31.
Merigan, W.H. 1980. Visual fields and flicker thresholds in methyl mercury-poisoned monkeys. In:
Neurotoxicity of the Visual System, W.H. Merigan and B. Weiss, Ed. Raven Press, New York, NY. p.
149-183.
Meydani, M. and J.N. Hathcock. 1984. Effect of dietary methionine on methyl mercury and atrazine
toxicity. Drug-Nutr. Interact. 2:217-233.
Michaelson, J.H., J.P. McCoy, P. Hirzel, et al. 1985. Mercury-induced autoimmune glomerulonephritis
in inbred rats—part I: Kinetics and species specificity of autoimmune responses. Surv. Synth. Pathol.
Res. 4:401-411.
Miettinen, J.K. 1973. Absorption and elimination of dietary (Hg++) and methyl mercury in man. In:
Mercury, Mercurial, and Mercaptans, M.W. Miller and T.W. Clarkson, Ed. Springfield, IL. p. 233-243.
Miettinen, J.K., T. Rahola, T. Hattula, K. Rissanen and M. Tillander. 1971. Elimination of 203-Hg-
methylmercury in man. Ann. Clin. Res. 3(2): 116-122.
Miller, J., D. Chaffin and R. Smith. 1975. Subclinical psychomotor and neuromuscular changes in
workers exposed to inorganic mercury. Am. Ind. Hyg. Assoc. J. 36:725-733.
Miller, C.T., Z. Zawidska, E. Nagy, et al. 1979. Indicators of genetic toxicity in leukocytes and
granulocytic precursors after chronic methyl mercury ingestion by cats. Bull. Environ. Contain. Toxicol.
21:296-303.
8-23
-------
Milne, J., A. Christophers, and P. De Silva. 1970. Acute mercurial pneumonitis. Br. J. Ind. Med.
27:334-338.
Mishonova, V.N., P.A. Stepanova, and V.V. Zarudin. 1980. Characteristics of the course of pregnancy
and births in women with occupational contact with small concentrations of metallic mercury vapors in
industrial facilities. Gig. Truda Prof. Zabol. 24(2):21-23.
Mitani, K. and T. Nakata. 1978. Environmental hygienic studies on mercury compounds. V.
Accumulating sites of inorganic mercury compounds in human placenta. Eisei Kagaku. 24(2): 102-106.
Mitsumori, K., K. Maita, T. Saito, et al. 1981. Carcinogenicity of methyl mercury chloride in ICRmice:
Preliminary note on renal carcinogenesis. Cancer Lett. 12:305-310.
Mitsumori, K., K. Takahashi, O. Matano, S. Goto and Y. Shirasu. 1983. Chronic toxicity of methyl
mercury chloride in rats: Clinical study and chemical analysis. Jpn. J. Vet. Sci. 45(6):747-757.
Mitsumori, K., K. Maita and Y. Shirasu. 1984. Chronic toxicity of methyl mercury chloride in rats:
Pathological study. Jpn. J. Vet. Sci. 46(4):549-557.
Mitsumori, K., M. Hirano, H. Ueda, et al. 1990. Chronic toxicity and carcinogenicity of methyl mercury
chloride in B6C3F1 mice. Fund. Appl. Toxicol. 14:179-190.
Miyakawa, T., E. Murayama, S. Sumiyoshi, et al. 1976. Late changes in human sural nerves in
Minamata disease and in nerves of rats with experimental organic mercury poisoning. Acta. Neuropath.
(Berlin). 35:131-138.
Mohamed, M., T. Burbacher and N. Mottet. 1987. Effects of methyl mercury on testicular functions in
Macaca fascicularis monkeys. Pharmacol. Toxicol. 60(l):29-36.
Morimoto, K., S. lijima and A. Koizumi. 1982. Selenite prevents the induction of sister-chromatid
exchanges by methyl mercury and mercuric chloride in human whole-blood cultures. Mutat. Res.
102:183-192.
Moszczynski, P., J. Lisiewicz, R. Bartus, et al. 1990. The serum immunoglobulins in workers after
prolonged occupational exposure to the mercury vapors. Rev. Roum. Med. Intern. 28(1):25-30.
Munro, I.C., E.A. Nera, S.M. Charbonneau, et al. 1980. Chronic toxicity of methyl mercury in the rat.
J. Environ. Path. Toxicol. 3:437-447.
Murphy, M.J., E.J. Culliford, V. Parsons. 1979. A case of poisoning with mercuric chloride.
Resuscitation. 7:35-44.
Musch, H.R., M. Bornhausen, H. Kreigel, et al. 1978. Methyl mercury chloride induces learning deficits
in prenatally treated rats. Arch. Toxicol. 40:103-108.
Mutti, A., S. Lucertini, M. Fornari, J. Franchini, A. Bernard, H. Roels, R. Lauverys. 1985. Urinary
excretion of a brush-border antigen revealed by monoclonal antibodies in subjects occupationally
exposed to heavy metals. Heavy Met Environ. International Conference 5th. Vol. 1, p. 565-567.
8-24
-------
Myers, G., D. Marsh, C. Cox, P. Davidson, E. Cernichiari, T. Clarkson and C. Shamlaye. 1994.
Enrollment and neurodevelopment at 6 months. Abstract presented at the Twelfth International
Neurotoxicology Conf. "Neurotoxicity of Mercury: Indicators and Effect of Low-Level Exposure". Hot
Springs, AR. November 1.
Myers, G., P. Davidson, C. Cox, C. Shamlaye, M. Tanner, O. Chooisy, J. Sloane-Reeves, D. Marsh, E.
Cernichiari, A. Choi, M. Berlin and T. Clarkson. 1995a. Neurodevelopmental outcomes of seychellois
children sixty-six months after in utero exposure to methylemrcury from a maternal fish diet: pilot study.
NeurotToxicology 16: 639-652.
Myers, G., D. Marsh, C.Cox, P. Davidson, C. Shamlaye, M. Tanner, A. Choi, E. Cernichiari, O. Choisy,
and T. Clarkson. 1995b. A pilot neurodevelopmental study of seychellois children following in utero
exposure to methylemrcury from a maternal fish diet. NeuroToxicology 16:629-638.
Myers, G., D. Marsh, P. Davidson, C. Cox, C. Shamlaye, M. Tanner, A. Choi, E. Cerncihicari, O. Choisy,
and T. Clarkson. 1995c. Main neurodevelopmental study of Seychellois children following in utero
exposure to methylemrcury from a maternal fish diet: Outcome at six months. NeuroToxicology 19 653-
664.
Myers, G., P. Davidson, C. Cox, C. Shamlaye, M. Tanner, D. Marsh, E. Cernichiari, L. Lapham, M.
Berlin, and T. Clarkson. 1995d. Summary of the Seychelles child development study on the relationship
of fetal methylmercury exposure to neurodevelopment. NeuroToxicology 16: 711 -716.
Naeye, R.L., L.S. Burt, D.L. Wright, D.L., et al. 1971. Neonatal mortality: The male disadvantage.
Pediatrics 48: 902-906.
Naganuma, A. and N. Imura. 1981. Properties of mercury and selenium in a high-molecular weight
substance in rabbit tissues formed by simultaneous administration. Pharmacol. Biochem. Behav.
15:449-454.
Nakai, S. and I. Machida. 1973. Genetic effect of organic mercury on yeast. Mutat. Res. 21:348.
Nakamura, I., K. Hosokawa, H. Tamra, et al. 1977. Reduced mercury excretion with feces in germfree
mice after oral administration of methyl mercury chloride. Bull. Environ. Contam. Toxicol. 17:528-533.
Nakazawa, N., F. Makino and S. Okada. 1975. Acute effects of mercuric compounds on cultured
mammalian cells. Biochem. Pharmacol. 24:489-493.
NAS. 1991. Methylmercury: FDA Risk Assessment and Current Regulations. In: Seafood Safety.
Committee on Evaluation of the Safety of Fishery Products, National Academy Press, Washington, DC.
p.196-221.
Newberne, P.M., O. Glaser, L. Friedman, et al. 1972. Chronic exposure of rats to methyl mercury in fish
protein. Nature. 237:40-41.
Newland, M.C., S. Yezhou, B. Logdberg and M. Berlin. 1994. Prolonged behavioral effects of in utero
exposure to lead or methyl mercury: Reduced sensitivity to changes in reinforcement contingencies
during behavioral transitions and in steady state. Toxicol. Appl. Pharmacol. 126:6-15.
8-25
-------
Newland, M.C., K. WarfVinge and M. Berlin. 1996. Behavioral consequences of in utero exposure to
mercury vapor: alterations in lever-press durations and learning in squirrel monkeys. Toxicol. Appl.
Pharmacol. 139: 374-386.
Newton, D. and F. Fry. 1978. The retention and distribution of radioactive mercuric oxide following
accidental inhalation. Ann. Occup. Hyg. 21:21-32.
Ngim, C.H., S.C. Foo, K.W., Boey, et al. 1992. Chronic neurobehavioural effects of elemental mercury
in dentists. Br. J. Ind. Med. 49(11):782-790.
Nielsen, J.B. 1992. Toxicokinetics of mercuric-chloride and methylmercuric chloride in mice. J.
Toxicol. Environ. Health. 37(1):85-122.
Nielsen, J.B. and O. Andersen. 1991a. Methyl mercuric chloride toxicokinetics in mice. I: Effects of
strain, sex, route of administration and dose. Pharmacol. Toxicol. 68(3):201-207.
Nielsen, J.B. and O. Andersen. 1991b. Methyl mercuric chloride toxicokinetics in mice. II: Sexual
differences in whole-body retention and deposition in blood, hair, skin, muscles and fat. Pharmacol.
Toxicol. 68(3):208-211.
Nielsen, J.B., H.R. Andersen, O. Andersen, et al. 1991. Mercuric chloride-induced kidney damage in
mice: Time course and effect of dose. J. Toxicol. Environ. Health. 34(4):469-483.
Nielsen, J.B., O. Andersen and P. Grandjean. 1994. Evaluation of mercury in hair, blood and muscle as
biomarkers for methyl mercury exposure in male and female mice. Arch. Toxicol. 68:317-321.
Nielsen-Kudsk, F. 1965. The influence of ethyl alcohol on the absorption of methyl mercury vapor from
the lungs of man. Acta. Pharmacol. Toxicol. 23:263-274.
Nielsen-Kudsk, F. 1973. Biological oxidation of elemental mercury. In: Mercury, Mercurials and
Mercaptans, M.W. Miller and T.W. Clarkson, Ed. Charles C Thomas, Springfield, IL. p. 355.
Nilsson, B. and B. Nilsson. 1986. Mercury in dental practice: I. The working environment of dental
personnel and their exposure to mercury vapor. Swed. Dent. J. 10:1-14.
Nilsson, B., L. Gerhardsson and G.F. Nordberg. 1990. Urine mercury levels and associated symptoms in
dental personnel. Sci. Total Environ. 94(3): 179-185.
Nishikido, N., K. Furuyashiki, A. Naganuma, et al. 1987. Material selenium deficiency enhances the
fetolethal toxicity of methylmercruy. Toxicol. Appl. Pharmacol. 88:322-328.
Nolen, G.A., E.V. Buchler, R.G. Geil, et al. 1972. Effects of trisodium nitrotriacetate on cadmium and
methyl mercury toxicity and teratogenicity in rats. Toxicol. Appl. Pharmacol. 23:222-237.
Nordberg, G. and F. Serenius. 1969. Distribution of inorganic mercury in the guinea pig brain. Acta.
Pharmacol. Toxicol. 27:269-283.
8-26
-------
Nordberg, G.F. and P. Strangert. 1976. Estimations of a dose-response curve for long-term exposure to
methylmercuric compounds in human being taking into account availability of critical organ
concentration and biological half-time: A preliminary communication. In: Effects and Dose-Response
Relationships of Toxic Metals, G.F. Nordberg, Ed. Elsevier, Amsterdam, p. 273-282.
Nordberg, G.F., D. Brune, L. Gerhardsson, et al. 1992. The ICOH and IUPAC international programme
for establishing reference values of metals. Sci. Total. Environ. 120(1-2):17-21.
Norseth, T. and T.W. Clarkson. 1970. Studies on the biotransformation of Hg-203-labeled methyl
mercury chloride. Arch. Environ. Health. 21:717-727'.
Norseth, T. and T.W. Clarkson. 1971. Intestinal transport of Hg-203-labeled methyl mercury chloride:
Role of biotransformation in rats. Arch. Environ. Health. 22:568-577.
NRC (National Research Council). 1983. Risk Assessment in the Federal Government: Managing the
Process. National Academy Press, Washington, DC. 388 p.
NTP. 1993. Toxicology and carcinogenesis studies of mercuric chloride (CAS no. 7487-94-7) in
F344/N rats and B6C3F1 mice (gavage studies) (as modified based on peer review). National
Toxicology Program, U.S. Department of Health and Human Services, Public Health Service, National
Institutes of Health, Research Triangle Park, NC. NTP TR 408. NIH Publ. No. 91-3139.
Nylander, M. and J. Weiner. 1991. Mercury and selenium concentrations and their interrelations in
organs from dental staff and the general-population. Br. J. Ind. Med. 48(11):729-734.
Oberly, T.J., C.E. Piper and D.S. McDonald. 1982. Mutagenicity of metal salts in the L5178Y mouse
lymphoma assay. J. Toxicol. Environ. Health. 9:367-376.
Ogata, M., K. Kenmotsu, N. Hirota, et al. 1987. Reduction of mercuric ion and exhalation of mercury in
acatalasemic and normal mice. Arch. Environ. Health. 42(1):26-30.
Ohi, G., M. Fukuda, H. Seto, et al. 1976. Effect of methyl mercury on humoral immune responses in
mice under conditions simulated to practical situations. Bull. Environ. Contain. Toxicol. 15(2): 175-180.
Ohi, G., S. Nishigaki, H. Seki, et al. 1980. The protective potency of marine animal meat against the
neurotoxicity of methyl mercury: Its relationship with the organ distribution of mercury and selenium in
rat. Food Cosmet. Toxicol. 18:139-145.
Ohno, H., F. Hnanoka and M. Yamada. 1982. Inducibility of sister-chromatid exchanges by heavy-metal
ions. Mutat. Res. 104:141-145.
Oikawa, K. H. Saito, I. Kufune, et al. 1982. Mercury absorption by inhaling through the nose and
expiring through the mouth at various concentrations. Chemosphere. 11(9):943-951.
Olson, K. and G.M. Boush. 1975. Decreased learning capacity in rats exposed prenatally and
postnatally to low doses of mercury. Bull. Environ. Contain. Toxicol. 13:73-79.
8-27
-------
Parizek, J. and I. Ostadolva. 1967. The protective effect of small amounts of selenite in sublimate
intoxication. Experientia. 23:142.
Paton, G.R. and A.C. Allison. 1972. Chromosome damage in human cell cultures induced by metal salts.
Mutat. Res. 16:332-336.
Pesce, A.J., I. Hanenson and K. Sethi. 1977. B2 microglobulinuria in a patient with nephrotoxicity
secondary to mercuric chloride ingestion. Clin. Toxicol. 11:309-315.
Pfeiffer, W.C., L.D. Lacerda, O. Malm, C.M.M. Souza, E.J. Silviera, and W.R. Bastos. 1989. Mercury
concentration in inland waters of gold-mining areas in Rondonia, Brazil. Sci. Tot. Envrion. 87:233-240.
Phelps, R.W., T.W. Clarkson, T.G. Kershaw and B. Wheatley. 1980. Interrelationships of blood and
hair mercury concentrations in a North American population exposed to methyl mercury. Arch. Environ.
Health. 35:161-168.
Piikivi, L. 1989. Cardiovascular reflexes and low long-term exposure to mercury vapour. Int. Arch.
Occup. Environ. Health. 61(6):391-395.
Piikivi, L. and H. Hanninen. 1989. Subjective symptoms and psychological performance of chlor-alkali
workers. Scand. J. Work Environ. Health. 15(l):69-74.
Piikivi, L. and A. Ruokonen. 1989. Renal function and long-term low mercury vapor exposure. Arch.
Environ. Health. 44(3): 146-149.
Piikivi, L. and U. Tolonen. 1989. EEG findings in chlor-alkali workers subjected to low long term
exposure to mercury vapour. Br. J. Ind. Med. 46(6):370-375.
Piikivi, L., H. Hanninen, T. Martelin, et al. 1984. Psychological performance and long term exposure to
mercury vapors. Scand. J. Work Environ. Health. 10:35-41.
Piotrowski, J., B. Trojanowska, J. Wisneiwska-Knypl, et al. 1973. Further investigations on binding and
release of mercury in the rat. In: Mercury, Mercurials and Mercaptans, M. Miller and T. Clarkson, Ed.
Springfield, IL. p. 247-263.
Piotrowski, J.K., B. Trojanowska and E.M. Mogilnicka. 1975. Excretion kinetics and variability of
urinary mercury in workers exposed to mercury vapor. Int. Arch. Occup. Environ. Health. 35:245-256.
Poma, K., M. Kirsch-Volders and C. Susanne. 1981. Mutagenicity study of mice given mercuric
chloride. J. Appl. Toxicol. 1:314-316.
Popescu, H.I., L. Negru and I. Lancranjan. 1979. Chromosome aberrations induced by occupational
exposure to mercury. Arch. Environ. Health. 34:461-463.
Post, E.M., M.G. Yang, J.A. King, et al. 1973. Behavioral changes of young rats force-fed methyl
mercury chloride (37480). Proc. Soc. Exp. Biol. Med. 143(4): 1113-1116.
Prem, A.S., K.D. Vachhrajani, M. Bose, et al. 1992. Action of mercuric chloride during one cycle of
seminiferous epithelium in the rat. Bull. Environ. Contain. Toxicol. 48(6):865-868.
8-28
-------
Pritchard, A.L., M.J. Collier, P.A. McAnulty, et al. 1982a. The effects of peri- and post-natal exposure
to inorganic mercury on growth, development and behaviour of rats. Teratology. 26:20A.
Pritchard, A.L., P.A. McAnulty, M.J. Collier, et al. 1982b. The effects of inorganic mercury on fertility
and survival in utero in the rat. Teratology. 26:20A.
Rahola, T., T. Hattula, A. Korolainen and J.K. Miettinen. 1973. Elimination of free and protein-bound
ionic mercury in man. Ann. Clin. Res. 5:214-219.
Ramel, C. 1972. Genetic effects. In: Mercury in the Environment, L. Friberg and J. Vostal, Ed. CRC
Press, Cleveland, OH. p. 169-181.
Rana, S.V.S. and P.R. Boora. 1992. Antiperoxidative mechanisms offered by selenium against liver
injury caused by cadmium and mercury in rat. Bull. Environ. Contam. Toxicol. 48(1): 120-124.
Rentes, P. and E. Saligman. 1968. Relationship between environmental exposure to mercury and
clinical observation. Arch. Environ. Health. 16:794-800.
Reuhl, K.R., L.W. Chang and J.W. Townsend. 1981. Pathological effects of in utero methyl mercury
exposure on the cerebellum of the Golden hamster. I. Early effects upon the neonatal cerebellar cortex.
Environ. Res. 26:281-306.
Reuhl, K.R., L.A. Lagunowich and D.L. Brown. 1994. Cytoskeleton and cell adhesion molecules:
Critical targets of toxic agents. Neurotoxicology. 15(1): 133-146.
Rice, B.C. 1989a. Delayed neurotoxicity in monkeys exposed developmentally to methyl mercury.
Neurotoxicology. 10(4):645-650.
Rice, B.C. 1989b. Brain and tissue levels of mercury after chronic methyl mercury exposure in the
monkey. J. Toxicol. Environ. Health. 27(2): 189-198.
Rice, B.C. 1992. Effects of pre-plus postnatal exposure to methyl mercury in the monkey on fixed
interval and discrimination reversal performance. Neurotoxicology. 13(2):443-452.
Rice, B.C. 1996. Sensory and cognitive effects of developmenatl methylmercury exposure in monkeys,
and a comparison to effects in rodents. NeuroToxicol. 17:139-154.
Rice, B.C. and S.G. Gilbert. 1982. Early chronic low-level methyl mercury poisoning in monkeys
impairs spatial vision. Science. 216:759-761.
Rice, B.C. and S.G. Gilbert. 1990. Effects of developmental exposure to methyl mercury on spatial and
temporal visual function in monkeys. Toxicol. Appl. Pharmacol. 102(1): 151-163.
Rice, B.C. and S.G. Gilbert. 1992. Exposure to methyl mercury from birth to adulthood impairs high-
frequency hearing in monkeys. Toxicol. Appl. Pharmacol. 115(1):6-10.
Rice, B.C., B. Krewski, B.T. Collins and R.F. Willes. 1989. Pharmacokinetics of methylmercury in the
blood of monkeys (Macacafascicularis). Fund. Appl. Toxicol. 12(l):23-33.
8-29
-------
Rizzo, A.M. and A. Furst. 1972. Mercury teratogenesis in the rat. Proc. West. Pharmacol. Soc.
15:52-54.
Robison, S.H., O. Cantoni and M. Costa. 1982. Strand breakage and decreased molecular weight of
DNA induced by specific metal compounds. Carcinogenesis. 3:657-662.
Robison, S.H., O. Cantoni and M. Costa. 1984. Analysis of metal-induced DNA lesions and DNA-repair
replication in mammalian cells. Mutat. Res. 131:173-181.
Rodier, P.M., M. Aschner and P.R. Sager. 1984. Mitotic arrest in the developing CNS after prenatal
exposure to methyl mercury. Neurobehav. Toxicol. Teratol. 6:379-385.
Roels, H.A., R. Lauwerys, J.P. Buchet, et al. 1982. Comparison of renal function and psychomotor
performance in workers exposed. Int. Arch. Occup. Environ. Health. 50:77-93.
Roels, H., J.-P. Gennart, R. Lauwerys, et al. 1985. Surveillance of workers exposed to mercury vapour:
Validation of a previously proposed biological threshold limit value for mercury concentration in urine.
Am. J. Ind. Med. 7:45-71.
Roels, H., S. Abdeladim, E. Ceulemans, et al. 1987. Relationships between the concentrations of
mercury in air and in blood or urine in workers exposed to mercury vapour. Ind. Occup. Hyg.
31(2):135-145.
Roels, H., S. Abdeladim, M. Braun, et al. 1989. Detection of hand tremor in workers exposed to
mercury vapor: A comparative study of three methods. Environ. Res. 49(2): 152-165.
Roels, H.A., M. Boeckyx, E. Ceulemans, et al. 1991. Urinary excretion of mercury after occupational
exposure to mercury vapor and influence of the chelating agent meso-2,3-dimercaptosuccinic acid
(DMSA). Br. J. Ind. Med. 48(4):247-253.
Rosenman, K.D., J.A. Valciukas, L. Glickman, et al. 1986. Sensitive indicators of inorganic mercury
toxicity. Arch. Environ. Health. 41:208-215.
Rothstein, A. and A.L. Hayes. 1960. The metabolism of mercury in the rat studied by isotope
techniques. J. Pharmacol. Exp. Ther. 130:166-176.
Rothstein, A. and A.L. Hayes. 1964. The turnover of mercury in rats exposed repeatedly to inhalation of
vapor. Health. Phys. 10:1099-1113.
Rowens, B., D. Guerrero-Betancourt, C.A. Gottlieb, et al. 1991. Respiratory failure and death following
acute inhalation of mercury vapor: A clinical and histologic perspective. Chest. 99(1): 185-190.
Rowland, I., M. Davies and P. Grasso. 1977. Biosynthesis of methylmercury compounds by the
intestinal flora of the rat. Arch. Environ. Health. 32(l):24-28.
Rowland, I., M. Davies and J. Evans. 1980. Tissue content of mercury in rats given methyl mercury
chloride orally: Influence of intestinal flora. Arch. Environ. Health. 35:155-160.
8-30
-------
Rumbeiha, W.K., P.A. Gentry and M.K. Bhatnagar. 1992. The effects of administering methyl mercury
in combination with ethanol in the rat. Vet. Hum. Toxicol. 34(l):21-25.
Rustam, H. and T. Hamdi. 1974. Methyl mercury poisoning in Iraq: A neurological study. Brain.
97:499-510.
Sager, P.R., Aschner and P.M. Rodier. 1984. Persistent differential alterations in developing cerebellar
cortex of male and female mice after methylmercury exposure. Brain Research and Developmental
Brain Research, 12, 1-11.
Sager, P.R., R.A. Doherty and P.M. Rodier. 1982. Effects of methyl mercury on developing mouse
cerebellar cortex. Exp. Neurol. 77:179-183.
Sapin, C., F. Hirsch, J.-P. Delaporte, et al. 1984. Polyclonal IgE increase after HgC12 injections in BN
and LEW rats: A genetic analysis. Immunogenetics. 20:227-236.
Sato, T. and F. Ikuta. 1975. Long-term studies on the neurotoxicity of small amount of methyl mercury
in monkeys (first report). In: Studies on the Health Effects of Alkylmercury in Japan, T. Tsubaki, Ed.
Environment Agency, Japan, p. 63-70.
Satoh, H., N. Yasuda and S. Shimai. 1985. Development of reflexes in neonatal mice prenatally exposed
to methyl mercury and selenite. Toxicol. Lett. 25: 199-203.
Schamberg, J., J. Kolmer and G. Raiziss. 1918. Experimental studies of the mode of absorption of
mercury when Appl.ied by injection. J. Am. Med. Assoc. 70:142.
Schantz, S.L., A.M. Sweeney, J.C. Gardiner, H.E.B. Humphrey, R.J. McCaffrey, D.M. Gasior, K.R.
Srikanth, and M.L. Budd. 1996. Neuropsychological assessment of Great Lakes Fisheaters. Toxicol. And
Indust. Health. 12:403-417.
Schreiner, G., B. Ulbrich and R. Bass. 1986. Testing strategies in behavioral teratology: II.
Discrimination learning. Neurobehav. Toxicol. Teratol. 8:567-572.
Schroeder, H. and M. Mitchener. 1975. Life-time effects of mercury, methyl mercury, and nine other
trace metals in mice. J. Nutr. 105:452-458.
Schwartz, J.G., T.E. Snider and M.M. Montiel. 1992. Toxicity of a family from vacuumed mercury.
Am. J. Emerg. Med. 10(3):258-261.
Scragg, R.K.R., A.J. McMichael and P.O. Clark. 1977. Pollution and sex ratio of births. MedJAust 2:
68.
Sebe, E. and Y. Itsuno. 1962. Organo-mercury compounds and Minamata disease. Nisshin. Igaku.
49:607-631.
Selypes, A., L. Nagymajtenyi and G. Berencsi. 1984. Study of the mutagenic and teratogenic effect of
aerogenic mercury exposition in mouse. Collect. Med. Leg. Toxicol. Med. 125:65-69.
8-31
-------
Sexton, D., K. Powell, J. Liddle, et al. 1976. A nonoccupational outbreak of inorganic mercury vapor
poisoning. Arch. Environ. Health. 33:186-191.
Sexton, D., K. Powell, J. Liddle, et al. 1978. A nonoccupational outbreak of inorganic mercury vapor
poisoning. Arch. Environ. Health. (July/August): 186-191.
Shafer, T.J. and W.D. Atchison. 1989. Block of 45Ca uptake into synaptosomes by methyl mercury:
Ca++- and Na+-dependence. J. Pharmacol. Exp. Therap. 248(2):696-702.
Shafer, T.J. and W.D. Atchison. 1991. Methyl mercury blocks N- and L-type Ca++ channels in nerve
growth factor-differentiated pheochromocytoma (PC12) cells. J. Pharmacol. Exp. Therap. 258(1): 149-
157.
Shafer, T.J. and W.D. Atchison. 1992. Effects of methyl mercury on perineurial Na+ and Ca2+-
dependent potentials at neuromuscular junctions of the mouse. Brain. Res. 595(2):215-219.
Shafer, T.J., M.L. Contreras and W.D. Atchison. 1990. Characterization of interactions of methyl
mercury with Ca2+ channels in synaptosomes and pheochromocytoma cells: Radiotracer flux and binding
studies. Molec. Pharmacol. 38:102-113.
Shamlaye, C.F., D.O. Marsh, G.J.Myers, C. Cox, P.W. Davidson, O. Choisy, E. Cernichiari, A. Choi,
M.A. Tanner and T.W. Clarkson. 1995. The Seychelles child development study of neurodevelopmental
outcomes in children following in utero exposure to methylmercury from a maternal fish diet:
background and demographics. NeuroToxicology. 16:596-612.
Shapiro, I.M., A.J. Sumner, L.K. Spitz, et al. 1982. Neurophysiological and neuropsychological
function in mercury exposed dentists. Lancet, i: 1147-1150.
Sherlock, J.C. and M.J. Quinn. 1988. Underestimation of dose-response relationship with particular
reference to the relationship between the dietary intake of mercury and its concentration in blood. Hum.
Toxicol. 7(2): 129-132.
Sherlock, J.C., D.G. Lindsay, W.H. Evans, J.E. Hislop and T.R. Collier. 1982. Duplication diet study on
mercury intake by fish consumers in the UK. Arch. Environ. Health. 37(5):271-278.
Siblerud, R.L. 1990. The relationship between mercury from dental amalgam and the cardiovascular
system. Sci. Total Environ. 99(l-2):23-36.
Sikorski, R., T. Juszkiewicz, T. Paszkowski, et al. 1987. Women in dental surgeries: Reproductive
hazards in occupational exposure to metallic mercury. Int. Arch. Occup. Environ. Health. 59:551-557.
Sin, Y.M., Y.F. Lim and M.K. Wong. 1983. Uptake and distribution of mercury in mice from ingesting
soluble and insoluble mercury compounds. Bull. Environ. Contam. Toxicol. 31(5):605-612.
Singer, R., J.A. Valciukas and K.D. Rosenman. 1987. Peripheral neurotoxicity in workers exposed to
inorganic mercury compounds. Arch. Environ. Health. 42(4): 181-184.
8-32
-------
Skerfving, S. 1974. Methyl mercury exposure, mercury levels in blood and hair, and health status in
Swedes consuming contaminated fish. Toxicology. 2:3-23.
Skerfving, S. 1988. Mercury in women exposed to melthylmercury through fish consumption, and in
their newborn babies and breast milk. Bull. Environ. Contam. Toxicol. 41(4):475-482.
Skerfving, S., K. Hansson and J. Lindsten. 1970. Chromosome breakage in humans exposed to methyl
mercury through fish consumption. Arch. Environ. Health. 21:133-139.
Skerfving, S., K. Hansson, C. Mangs, J. Lindsten and N. Ryman. 1974. Methyl mercury-induced
chromosome damage in man. Environ. Res. 7:83-98.
Skog, E. and J. Wahlberg. 1964. A comparative investigation of the percutaneous absorption of metal
compounds in the guinea pig by means of the radioactive isotopes: 51Cr, 58Co, 65Zn, 110Ag, 115Cd, 203Hg.
J. Invest Dermatol. 43:187-192.
Slemr, F. and E. Langer. 1992. Increase in global atmospheric concentrations of mercury inferred from
measurements over the Atlantic Ocean. Nature. 355:434-437.
Smith, R.G., A.J. Vorwald, L.S. Patel, et al. 1970. Effects of exposure to mercury in the manufacture of
chlorine. Am. Ind. Hyg. Assoc. J. 31:687-700.
Smith, J., P. Allen, M. Turner, et al. 1994. The kinetics of intravenously administered methyl mercury
in man. Toxicol. Appl. Pharmacol. (128:251-256)
Snodgrass, W., J.B. Sullivan, B.H. Rumack, et al. 1981. Mercury poisoning from home gold ore
processing: Use of penicillamine and dimercaprol. J. Am. Med. Assoc. 246:1929-1931.
Snyder, R. and D. Seelinger. 1976. Methyl mercury poisoning: Clinical follow-up and sensory nerve
conduction studies. J. Neurol. Neurosurg. Psychiat. 39:701-704.
Snyder, W.S., M.T. Cook, L.R. Karhausen et al. 1981. International Commission of Radiological
Protection. No. 23: Report of a Task Group on Reference Man. Pergamon Press, NY.
Soni, J.P., R.U. Singhania, A. Bansal, et al. 1992. Acute mercury vapor poisoning. Indian Pediatr.
29(3):365-368.
Soria, M.L., P. Sanz, D. Martinez, et al. 1992. Total mercury and methyl mercury in hair, maternal and
umbilical blood, and placenta from women in the Seville area. Bull. Environ. Contam. Toxicol.
48(4):494-501.
Spyker, J.M. 1975. Assessing the impact of low level chemicals on development: behavioral and latent
effects. Fed. Proc. 34:1835-1844.
Spyker, J.M., S.B. Sparber and A.M. Goldberg. 1972. Subtle consequences of methyl mercury
exposure: behavioral deviations in offspring of treated mothers. Science. 177:621-623.
8-33
-------
Steffek, A.J., R. Clayton, C. Siew, et al. 1987. Effects of elemental mercury vapor exposure on pregnant
Sprague-Dawley rats [Abstract]. Teratology. 35:59A.
Stern, A.H. 1993. Re-evaluation of the reference dose for methyl mercury and assessment of current
exposure levels. Risk Anal. 13(3):355-364.
Stewart, W., H. Gurgis, J. Sanderson, et al. 1977. Urinary mercury excretion and proteinuria in
pathology laboratory staff. Br. J. Ind. Med. 34:26-31.
Stoltenburg-Didinger, G. and S. Markwort. 1990. Prenatal methyl mercury exposure results in dendritic
spine dysgenesis in rats. Neurotoxicol. Teratol. 12(6):573-576.
Stonard, M.D., B.R. Chater, D.P. Duffield, et al. 1983. An evaluation of renal function in workers
occupationally exposed to mercury vapor. Int. Arch. Occup. Environ. Health. 52:177-189.
Stopford, W., S.d. Bundy, L.J. Goldwater, et al. 1978. Microenvironmental exposure to mercury vapor.
Am. Ind. Hyg. Assoc. J. 39:378-384.
Su, M.O. and G.T. Okita. 1976. Behavioral effects on the progeny of mice treated with methyl mercury.
Toxicol. Appl. Pharmacol. 38:195-205.
Suda, I. and K. Hirayama. 1992. Degradation of methyl- and ethylmercury into inorganic mercury by
hydroxyl radical produced from rat liver microsomes. Arch. Toxicol. 66(6):398-402.
Suda, I. and H. Takahashi. 1986. Enhanced and inhibited biotransformation of methyl mercury in the rat
spleen. Toxicol. Appl. Pharmacol. 82:45-52.
Suda, I. and H. Takahashi. 1992. Degradation of methyl and ethyl mercury into inorganic mercury by
other reactive oxygen species besides hydroxyl radical. Arch. Toxicol. 66(l):34-39.
Sugata, Y. and T.W. Clarkson. 1979. Exhalation of mercury: Further evidence for an oxidation-
reduction cycle in mammalian tissues. Biochem. Pharmacol. 28:3474-3476.
Sundberg, J. and A. Oskarsson. 1992. Placental and lactational transfer of mercury from rats exposed to
methyl mercury in their diet: Speciation of mercury in the offspring. J. Trace Elem. Exp. Med.
5(l):47-56.
Suter, K.E. 1975. Studies on the dominant lethal and fertility effects of the heavy metal compounds
methyl mercuric hydroxide, mercuric chloride, and cadmium chloride in male and female mice. Mutat.
Res. 30: 365-374.
Suter, K.E. and H. Schon. 1986. Testing strategies in behavioral teratology: I. Testing battery approach.
Neurobehav. Toxicol. Teratol. 8:561-566.
Suzuki, T., T. Miyama and H. Katsunuma 1963. Comparative study of bodily distribution of mercury in
mice after subcutaneous administration of methyl, ethyl and n-propyl mercury acetate. Japan J. Exptl.
Med. 33:277-282.
8-34
-------
Suzuki, T., T. Miyama and H. Katsunuma 1964. Comparative study of bodily distribution of mercury in
mice after subcutaneous administration of propyl, butyl and amyl mercury acetate. Japan J. Exptl. Med.
34:211-216.
Suzuki, T., J. Yonemoto, H. Satoh, et al. 1984. Normal organic and inorganic mercury levels in the
human feto-placental system. J. Appl. Toxicol. 4(5):249-252.
Suzuki, T., T. Hongo, N. Matsuo, et al. 1992. An acute mercuric mercury poisoning: Chemical
speciation of hair mercury shows a peak of inorganic mercury value. Hum. Exp. Toxicol. 1 l(l):53-57.
Suzuki, T., T. Hongo, J. Yoshinaga, H. Imai, M. Nakazawa, N. Matsuo and H. Akagi. 1993. The hair-
organ relationship in mercury concentration in contemporary Japanese. Arch. Environ. Health.
48:221-229.
Swedish Expert Group. 1971. Methyl mercury in fish. A Toxicol.ogical-epidemiological evaluation of
risks. Norw. Hyg. Tidschr. 4(Suppl):19-364.
Syversen, T.L.M. 1977. Effects of methyl mercury on in vivo protein synthesis in isolated cerebral and
cerebellar neurons. Toxicol. Lett. 10:31-34.
Takahata, N., H. Hayashi, B. Watanabe, et al. 1970. Accumulation of mercury in the brains of two
autopsy cases with chronic inorganic mercury poisoning. Folia. Psychiatr. Neurol. Jpn. 24:59-69.
Takeuchi, T. and K. Eto. 1975. Minamata disease. Chronic occurrence from pathological view points.
In: Studies on the Health Effects of Alkylmercury in Japan. Tokyo, Japan Environment Agency.
p. 28-62.
Takeuchi, T., N. Morikawa, H. Matsumoto, et al. 1962. A pathological study of Minamata disease in
Japan. Acta. Neuropathol. 2:40-57.
Tamashiro, H., H. Akagi, M. Arakaki, et al. 1984. Causes of death in Minamata disease: Analysis of
death certificates. Int. Arch. Occup. Environ. Health. 54:135-146.
Tamashiro, H., M. Arakaki, H. Akagi, et al. 1986. Effects of ethanol on methyl mercury toxicity in rats.
J. Toxicol. Environ. Health. 18:595-605.
Tamashiro, H., M. Arakaki, H. Akagi, M.Futasuka, and L.H. Roht. 1985. Mortality and sutvival for
Minmata disease. Inj. J. Epidemiol. 14:582-588.
Tamashiro, H., M. Arakaki, M. Futatsuka and E.S. Lee. 1986. Methylmercury exposure and mortality in
southern Japan: A close look at causes of death. J. Epidemiol. Commun. Health 40:181-185.
Tanaka, T., A. Naganuma, K. Kobayashi et al. 1991. An explanation for strain and sex differences in
renal uptake of methyl mercury in mice. Toxicology. 69:317-329.
Tanaka, T., A. Naganuma, N. Miura et al. 1992. Role of testosterone in gamma-glutamyl transpeptidase-
dependent renal methyl mercury uptake in mice. Toxicol. Appl. Pharmacol. 112:58-63.
8-35
-------
Taueg, C., D.J. Sanfilippo, B. Rowens, et al. 1992. Acute and chronic poisoning from residential
exposures to elemental mercury. J. Toxicol. Clin. Toxicol. 30(l):63-67.
Teisinger, J. and V. Fiserova-Bergerova. 1965. Pulmonary retention and excretion of mercury vapors in
man. Ind. Med. Surg. 34:580.
Thomas, D., H. Fisher, L. Hall, et al. 1982. Effects of age and sex on retention of mercury by methyl
mercury-treated rats. Toxicol. Appl. Pharmacol. 62:445-454.
Thomas, D.J., H.L. Fisher, M.R. Sumler, et al. 1986. Sexual differences in the distribution and retention
of organic and inorganic mercury in methyl mercury-treated rats. Environ. Res. 41:219-234.
Thomas, D., H. Fisher, M. Sumler, et al. 1987. Sexual differences in the excretion of organic and
inorganic mercury by methyl mercury-treated rats. Environ. Res. 43:203-216.
Thomas, D.J., H.L. Fisher, M.R. Sumler, et al. 1988. Distribution and retention of organic and inorganic
mercury in methyl mercury-treated neonatal rats. Environ. Res. 47:59-71.
Thorp, J.M. Jr., D. Boyette, W.J. Watson, et al. 1992. Elemental mercury exposure in early pregnancy.
Obstet. Gynecol. 79(5 Pt 2):874-876.
Troen, P., S.A. Kaufman and K.H. Katz. 1951. Mercuric bichloride poisoning. N. Engl. J. Med.
244:459-463.
Tsubaki, T.K. and K. Irukayama. 1977. Minamata Disease: Methyl mercury Poisoning in Minamata and
Niigata, Japan. Elsevier Scientific Publ. Co., New York. p. 143-253.
Tsubaki, T. and H. Takahashi. 1986. Recent advances in Minamata disease studies. Kodansha, Ltd.,
Tokyo,Japan.
Tsubaki, T., K. Hiroto, K. Shirakawa, et al. 1978. Clinical epidemiological and Toxicol.ogical studies
on methyl mercury poisoning. In: Proc of 1st Intl. Conference in Toxicology: Toxicology as Predictive
Science, G.L. Plan and W.A.M. Duncan, Ed. Academy Press, New York, NY. p. 339-357.
Tubbs, R., D. Gordon, N. Gephardt, et al. 1982. Membranous glomerulonephritis associated with
industrial mercury exposure ~ Study of pathogenic mechanisms. Am. J. Clin. Pathol. 77:409-413.
Turner, C.J., M.K. Bhatnagar and S. Yamashiro. 1981. Ethanol potentiation of methyl mercury toxicity:
A preliminary report. J. Toxicol. Environ. Health. 7:665-668.
Umeda, M. and M. Nishimura. 1979. Inducibility of chromosomal aberrations by metal compounds in
cultured mammalian cells. Mutat. Res. 67:221-229.
U.S. EPA. 1980. Ambient Water Quality Criteria Document for Mercury. Prepared by the Office of
Health and Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH
for the Office of Water Regulation and Standards, Washington, DC. EPA 440/5-80-058. NTIS PB 81-
117699.
8-36
-------
U.S. EPA. 1984. Mercury Health Effects Update, Health Issue Assessment: Final report. Office of
Research and Development, Office of Health and Environmental Assessment, Environmental Criteria and
Assessment Office, Research Triangle Park, NC. EPA 600/8-84-019F.
U.S. EPA. 1985. Drinking Water Criteria Document for Mercury. U.S. Environmental Protection
Agency, Office of Drinking Water, Washington, DC.
U.S. EPA. 1986. Guidelines for Mutagenicity Risk Assessment. Federal Register. 51:34006-34012.
U.S. EPA. 1987a. The Risk Assesment Guidelines of 1986. Office of Health and Environmental
Assessment, Washington DC 20460. EPA/600/8-87/045.
U.S. EPA. 1987b. Peer Review Workshop on Mercury Issues. October 26-27, 1987, Summary Report.
Environmental Criteria and Assessment Office, Cincinnati, OH.
U.S. EPA. 1988. Drinking Water Criteria Document for Inorganic Mercury. Prepared by the Office of
Health and Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH
for the Office of Drinking Water, Washington, DC.
U.S. EPA. 1989. Proposed Amendments to the Guidelines for the Health Assessment of Suspect
Developmental Toxicants. Federal Register. 54:9386-9403.
U.S. EPA. 1990. Interim Methods for Development of Inhalation Reference Concentrations.
EPA 600/8-90/066A.
U.S. EPA. 1991. Guidelines for Developmental Toxicity Risk Assessment. Federal Register. 56:63798-
63826.
U.S. EPA. 1992. Dermal Exposure Assessment: Principles and Applications. EPA/600/8-91/01 IB,
January.
U.S. EPA. 1994. Integrated Risk Information System (IRIS). Online. Office of Health and
Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH.
U.S. EPA. 1997. The National Survey of Mercury Concentrations in Fish. Database Summary. 1990-
1995. September 29, 1997.
Vahter, M.E., N.K.ottet, L.T. Friberg, S.B. Lind,, J.S. Charleston, and T.M. Burbacher. 1995. Retention
of inorganic mercury in different brain sites of Macaca fascicularis monkeys during long-term
subclinical exposure to methylmercury. Toxicol.^4/?/>/. Pharmacol. 134:273-284.
Verbeck, M., H. Salle and C. Kemper. 1986. Tremor in workers with low exposure to metallic mercury.
Am. Ind. Hyg. Assoc. J. 47:559-562.
Verity, M.A. and T. Sarafian. 1991. Role of oxidative injury in the pathogenesis of methyl mercury
neurotoxicity. In: Advances in Mercury Toxicology, T. Suzuki, N. Imura and T.W. Clarkson, Ed.
Plenum Press, New York, NY. p. 209-222.
8-37
-------
Verschaeve, L., M. Kirsch-Volders, C. Susanne, et al. 1976. Genetic damage induced by occupationally
low mercury exposure. Environ. Res. 12:306-316.
Verschaeve, L., J.-P. Tassignon, M. Lefevre, et al. 1979. Cytogenic investigation on leukocytes of
workers exposed to metallic mercury. Environ. Mutagen. 1:259-268.
Verschaeve, L., M. Kirsch-Volders and C. Susanne. 1983. Mercury chloride- and methyl mercury
chloride-induced inhibition in NOR activity. Teratol. Carcinogen. Mutagen. 3:447-456.
Verschaeve, L., M. Kirsch-Volders and C. Susanne. 1984. Mercury-induced segregational errors of
chromosomes in human lymphocytes and in Indian muntjac cells. Toxicol. Lett. 21:247-253.
Verschuuren, H.G., R. Kroes, E.M. Den Tonkelaar, et al. 1976. Toxicity of methyl mercury chloride in
rats. III. Long-term toxicity study. Toxicol. 6:107-123.
Von Burg, R. and H. Rustam. 1974a. Electrophysiological investigations of methyl mercury intoxication
in humans: Evaluation of peripheral nerve by conduction velocity and electromyography.
Electroenceph. Clin. Neurophysiol. 37:381-392.
Von Burg, R. and H. Rustam. 1974b. Conduction velocities in methyl mercury poisoned patients. Bull.
Environ. Contam. Toxicol. 12(l):81-85.
Vorhees, C. 1985. Behavioral effects of prenatal methyl mercury in rats: A parallel trial to the
collaborative behavioral teratology study. Neurobehav. Toxicol. Teratol. 7:717-725.
Vroom, F.Q. and M. Greer. 1972. Mercury vapor intoxication. Brain. 95:305-318.
Wada, O., K. Toyokawa, T. Suzuki, et al. 1969. Response to a low concentration of mercury vapor:
Relation to human porphyrin metabolism. Arch. Environ. Health. 19:485-488.
Wakita, Y. 1987. Hypertension induced by methyl mercury in rats. Toxicol. Appl. Pharmacol.
89:144-147.
Walsh, C.T. 1982. The influence of age on the gastrointestinal absorption of mercuric chloride and
methyl mercury chloride in the rat. Environ. Res. 27: 412-420.
Warkany, J. and D.M. Hubbard. 1953. Acrodynia and mercury. J. Pediat 42:365-386.
Watanabe, T., T. Shimada and A. Endo. 1982. Effects of mercury compounds on ovulation and meiotic
and mitotic chromosomes in female golden hamsters. Teratology. 25:381-384.
Weiss, B. and W. Simon. 1975. Quantitative perspectives on the long-term toxicity of methyl mercury
and similar poisons. In: Environmental Science Research, Vol. 5. Behavioral Toxicology, B. Weiss and
V.G. Laties, Ed. Plenum Press, New York, NY. p. 429-437.
Welsh, S.O. 1979. The protective effect of vitamin E and N, n-diphenyl-p-phenylenediamine against
methyl mercury toxicity in the rat. J. Nutr. 109:1673-1681.
8-38
-------
West, I. and J. Lim. 1968. Mercury poisoning among workers in California mercury mills—preliminary
report. J. Occup. Med. 10:697-701.
WHO (World Health Organization). 1972. Evaluation of certain food additive and the contaminants
mercury, lead and cadmium. Sixteenth Report of the join FAO/WHO Committee on Food Additives.
WHO Technical Report Series No. 505. Geneva, Switzerland.
WHO (World Health Organization). 1976. Environmental Health Criteria: Mercury. World Health
Organization, Geneva, Switzerland, 121.
WHO (World Health Organization). 1980. Health risk evaluation of methyl mercury: An interim
review. EHE/EHC/80.22, World Health Organization, Geneva, Switzerland.
WHO (World Health Organization). 1990. Methyl mercury. Vol.101. World Health Organization,
International Programme on Chemical Safety, Geneva, Switzerland.
WHO (World Health Organization). 1991. Inorganic mercury. Vol.118. World Health Organization,
International Programme on Chemical Safety, Geneva, Switzerland, 168.
Williams, M.V., T. Winters and K.S. Waddel. 1987. In vivo effects of mercury (II) on deoxyuridine
triphosphate nucleotidohydrolase, DNA polymerase (alpha, beta) and uracil-DNA glycosylase activities
in cultured human cells: Relationship to DNA damage, DNA repair, and cytotoxicity. Mol.
Pharmacol. 31:200-207.
Williamson, A.M., R. Teo and J. Sanderson. 1982. Occupational mercury exposure and its
consequences for behavior. Int. Arch. Occup. Environ. Health. 50:273-286.
Wong, P.K. 1988. Mutagenicity of heavy metals. Bull. Environ. Contam. Toxicol. 40(4):597-603.
Wulf, H.C., N. Kromann, N. Kousgaard, et al. 1986. Sister chromatid exchange (SCE) in Greenlandic
Eskimos: Dose-response relationship between SCE and seal diet, smoking, and blood cadmium and
mercury concentrations. Sci. Total Environ. 48:81-94.
Yamini, B. and S.D. Sleight. 1984. Effects of ascorbic acid deficiency on methyl mercury
dicyandiamide toxicosis in guinea pigs. JEPTO 5-4(5): 139-150.
Yang, M.G., J.H.C. Wang, J.D. Garcia, et al. 1973. Mammary transfer of203 Hg from mothers to brains
of nursing rats. Proc. Soc. Exp. Biol. Med. 142:723-727.
Yasuda, Y., A.R. Datu, S. Hirata, et al. 1985. Characteristics of growth and palatal shelf development in
ICR mice after exposure to methyl mercury. Teratology. 32:273-286.
Yasutake, A. and K. Hirayama. 1988. Sex and strain differences of susceptibility to methyl mercury
toxicity in mice. Toxicology. 51:47-55.
Yasutake, A., Y. Hirayama and M. Inouye. 1991. Sex differences of nephrotoxicity by methyl mercury
in mice. In: Nephrotoxicity: Mechanisms, Early Diagnosis, and Therapeutic Management, P.H. Bach,
8-39
-------
Ed. Fourth International Symposium on Nephrotoxicity, Guilford, England, UK, 1989. Marcel Dekker,
Inc., New York, NY. p. 389-396.
Yip, R.K. and L.W. Chang. 1981. Vulnerability of dorsal route neurons and fibers toward methyl
mercury toxicity: A morphological evaluation. Environ. Res. 26:152-167.
Yoshida, M., H. Satoh, H. Aoyama, et al. 1989. Distribution of mercury in neonatal guinea pigs after
exposure to mercury vapor. Bull. Environ. Contain. Toxicol. 43(5):697-704.
Yoshida, M., H. Satoh and T. Kishimoto. 1992. Exposure to mercury via breast milk in suckling
offspring of maternal guinea pigs exposed to mercury vapor after parturition. J. Toxicol. Environ.
Health. 35(2): 135-139.
Zalups, R.K. and M.G. Cherian. 1992. Renal metallothionein metabolism after a reduction of renal
mass: II. Effects of zinc pretreatment on the renal toxicity and intrarenal accumulation of inorganic
mercury. Toxicology. 71(1-2): 103-117.
Zasukhina, G.D., I.M. Vasilyeva, N.I. Sdirkova, et al. 1983. Mutagenic effect of thallium and mercury
salts on rodent cells with different repair activities. Mutat. Res. 124:163-173.
Zenick, H. 1976. Evoked potential alterations in methyl mercury chloride toxicity. Pharmacol.
Biochem. Behav. 5:253-255.
8-40
-------
APPENDIX A
DOSE CONVERSIONS
All doses in the tables in Section 4 were adjusted for the amount of mercury in the compound.
For example, for animals administered 1 mg/kg/day mercuric chloride:
Molecular weight of mercuric chloride = 271.5
Molecular weight of mercury = 200.6
Dose of mercury = 1 mg Hg/kg/day x 200.6/271.5 = 0.74 mg Hg/kg/day
(1) To convert from ppm in feed to mg/kg body weight/day, the following equation was used:
mg toxicant (T)/kg body weight/day = mg T/kg food x food factor
where food factor = kg food per day intake/kg body weight
(2) To convert from ppm in water to mg/kg body weight/day, the following equation was used:
mg T/kg body weight/day = mg T/L water x L water per day intake/kg body weight
where L is liters of water intake per day
Species
Mouse
Rat
Rabbit
Sample Values Used
Water
intake/day
(Liter/day)
0.0057
0.049
0.41
Body weight
(kg)
0.03
0.35
3.8
Foodfactor
(kg food/kg body
weight)
0.13
0.05
0.049
(3) To convert from ppm in air to mg/m3 for a vapor, the following equation was used:
1 mg/m3 = 1 ppm x molecular weight/24.45
A-l
-------
APPENDIX B
SUMMARIES FOR THE
INTEGRATED RISK INFORMATION SYSTEM (IRIS)
-------
_I.B. REFERENCE CONCENTRATION FOR CHRONIC INHALATION EXPOSURE (RfC)
Substance Name ~ Elemental mercury (Hg)
CASRN - 7439-97-6
Preparation date - 3/12/90
J.B.I. INHALATION RfC SUMMARY
Critical Effect Exposures* UF MF RfC
Hand tremor; increases NOAEL: None 30 1 3E-4
in memory disturbances; mg/cu.m
slight subjective and LOAEL: 0.025 mg/cu.m
objective evidence of (converted to LOAEL [ADJ]
autonomic dysfunction of 0.009 mg/cu.m
Human occupational
inhalation studies
Faweretal, 1983;
Piikivi and Tolonen, 1989;
Piikivi and Hanninen, 1989;
Piikivi, 1989;
Ngimetal., 1992;
Liang etal., 1993
* Conversion Factors and Assumptions: This is an extrarespiratory effect of a vapor (gas). The LOAEL
is based on an 8-hour TWA occupational exposure. MVho = 10 cu.m/day, MVh = 20 cu.m/day.
LOAEL(HEC) = LOAEL(ADJ) = 0.025 mg/cu.m x MVho/MVh x 5 days/7 days = 0.009 mg/cu.m. Air
concentrations (TWA) were measured in the Fawer et al. (1983), Ngim et al. (1992), and Liang et al.
(1993) studies. Air concentrations were extrapolated from blood levels based on the conversion factor of
Roels et al. (1987) as described in the Additional Comments section for the studies of Piikivi and
Tolonen (1989), Piikivi and Hanninen (1989), and Piikivi (1989).
J.B.2. PRINCIPAL AND SUPPORTING STUDIES (INHALATION RfC)
Fawer, R.F., U. DeRibaupierre, M.P. Guillemin, M. Berode and M. Lobe. 1983. Measurement of hand
tremor induced by industrial exposure to metallic mercury. J. Ind. Med. 40: 204-208.
Piikivi, L. and U. Tolonen. 1989. EEG findings in chlor-alkali workers subjected to low long term
exposure to mercury vapor. Br. J. Ind. Med. 46: 370-375.
Piikivi, L. and H. Hanninen. 1989. Subjective symptoms and psychological performance of
chlorine-alkali workers. Scand. J. Work Environ. Health. 15:69-74.
B-l
-------
Piikivi, L. 1989. Cardiovascular reflexes and low long-term exposure to mercury vapor. Int. Arch.
Occup. Environ. Health. 61: 391-395.
Ngim, C.H., S.C. Foo, K.W. Boey and J. Jeyaratnam. 1992. Chronic neurobehavioral effects of
elemental mercury in dentists. Br. J. Ind. Med. 49: 782-790.
Liang, Y-X., R-K. Sun, Y. Sun, Z-Q. Chen and L-H. Li. 1993. Psychological effects of low exposure to
mercury vapor: Application of a computer-administered neurobehavioral evaluation system. Environ.
Res. 60: 320-327.
Fawer et al. (1983) used a sensitive objective electronic measure of intention tremor (tremors that
occur at the initiation of voluntary movements) in 26 male workers (mean age of 44 years) exposed to
low levels of mercury vapor in various occupations: fluorescent tube manufacture (n=7), chloralkali
plants (n=12), and acetaldehyde production (n=7). Controls (n=25; mean age of 44.6 years) came from
the same factories but were not exposed occupationally. Personal air samples (two per subject) were
used to characterize an average exposure concentration of 0.026 mg/cu.m. It should be noted that it is
likely that the levels of mercury in the air varied during the period of exposure and historical data
indicate that previous exposures may have been higher. Exposure measurements for the control cohort
were not performed. The average duration of exposure was 15.3 years. The measures of tremor were
significantly increased in the exposed compared to control cohorts, and were shown to correspond to
exposure and not to chronologic age. These findings are consistent with neurophysiological impairments
that might result from accumulation of mercury in the cerebellum and basal ganglia. Thus, the TWA of
0.026 mg/cu.m was designated a LOAEL. Using the TWA and adjusting for occupational ventilation
rates and workweek, the resultant LOAEL(HEC) is 0.009 mg/cu.m.
Piikivi and Tolonen (1989) used EEGs to study the effects of long-term exposure to mercury
vapor in 41 chloralkali workers exposed for a mean of 15.6 +/- 8.9 years as compared with matched
referent controls. They found that the exposed workers, who had mean blood Hg levels of 12 ug/L and
mean urine Hg levels of 20 ug/L, tended to have an increased number of EEG abnormalities when
analyzed by visual inspection only. When the EEGs were analyzed by computer, however, the exposed
workers were found to have significantly slower and attenuated brain activity as compared with the
referents. These changes were observed in 15% of the exposed workers. The frequency of these changes
correlated with cortical Hg content (measured in other studies); the changes were most prominent in the
occipital cortex less prominent in the parietal cortex, and almost absent in the frontal cortex. The authors
extrapolated an exposure level associated with these EEG changes of 0.025 mg/cu.m from blood levels
based on the conversion factor calculated by Roels et al. (1987).
Piikivi and Hanninen (1989) studied the subjective symptoms and psychological performances
on a computer-administered test battery in 60 chloralkali workers exposed to mercury vapor for a mean
of 13.7 +/- 5.5 years as compared with matched referent controls. The exposed workers had mean blood
Hg levels of 10 ug/L and mean urine Hg levels of 17 ug/L. A
statistically significant increase in subjective measures of memory disturbance and sleep disorders was
found in the exposed workers. The exposed workers also reported more anger, fatigue and confusion.
No objective disturbances in perceptual motor, memory or learning abilities were found in the exposed
workers. The authors extrapolated an exposure level associated
with these subjective measures of memory disturbance of 0.025 mg/cu.m from blood levels based on the
conversion factor calculated by Roels et al. (1987).
B-2
-------
Both subjective and objective symptoms of autonomic dysfunction were investigated in 41
chloralkali workers exposed to mercury vapor for a mean of 15.6 +/- 8.9 years as compared with matched
referent controls (Piikivi, 1989). The quantitative non-invasive test battery consisted of measurements of
pulse rate variation in normal and deep breathing, in the Valsalva maneuver and in vertical tilt, as well as
blood pressure responses during standing and isometric work. The exposed workers had mean blood
levels of 11.6 ug/L and mean urine levels of 19.3 ug/L. The exposed workers complained of more
subjective symptoms of autonomic dysfunction than the controls, but the only statistically significant
difference was an increased reporting of palpitations in the exposed workers. The quantitative tests
revealed a slight decrease in pulse rate variations, indicative of autonomic reflex dysfunction, in the
exposed workers. The authors extrapolated an exposure level associated with these subjective and
objective measures of autonomic dysfunction of 0.030 mg/cu.m from blood levels based on the
conversion factor calculated by Roels et al. (1987).
Two more recent studies in other working populations corroborate the neurobehavioral toxicity
of low-level mercury exposures observed in the Fawer et al. (1983), Piikivi and Tolonen (1989), Piikivi
and Hanninen (1989), and Piikivi (1989) studies.
Ngim et al. (1992) assessed neurobehavioral performance in a cross-sectional study of 98 dentists
(38 female, 60 male; mean age 32, range 24-49 years) exposed to TWA concentrations of 0.014 mg/cu.m
(range 0.0007 to 0.042 mg/cu.m) versus 54 controls (27 female, 27 male; mean age 34, range 23-50
years) with no history of occupational exposure to mercury. Air concentrations were measured with
personal sampling badges over typical working hours (8-10 hours) and converted to an 8-hour TWA. No
details on the number of exposure samples or exposure histories were provided. Blood samples from the
exposed cohort were also taken and the data supported the correspondence calculated by Roels et al.
(1987). Based on extrapolation of the average blood mercury concentration (9.8 ug/L), the average
exposure concentration would be estimated at 0.023 mg/cu.m. The average duration of practice of the
exposed dentists was 5.5 years. Exposure measurements of the control cohort were not performed. The
exposed and control groups were adequately matched for age, amount offish consumption, and number
of amalgam dental fillings. The performance of the dentists was significantly worse than controls on a
number of neurobehavioural tests measuring motor speed (finger tapping), visual scaning, visumotor
coordination and concentration, visual memmory, and visuomotor coordination speed. These
neurobehavioral effects are consistent with central and peripheral neurotoxicity and the TWA is
considered a LOAEL. Using the TWA and adjusting for occupational ventilation rates and the reported
6-day workweek, the resultant LOAEL(HEC) is 0.006 mg/cu.m.
Liang et al. (1993) investigated workers in a fluorescent lamp factory with a
computer-adminstered neurobehavioral evaluation system and a mood inventory profile. The exposed
cohort (mean age 34.2 years) consisted of 19 females and 69 males exposed to ninterruptedly for at least
2 years prior to the study. Exposure was monitored with area samplers and ranged from 0.008 to 0.085
mg/cu.m across worksites. No details on how the exposure profiles to account for time spent in different
worksites were constructed. The average exposure was estimated at 0.033 mg/cu.m. (range 0.005 to 0.19
mg/cu.m). The average duration was of working was 15.8 years for the exposed cohort. Urinary
excretion was also monitored and reported to average 0.025 mg/L. The control cohort (mean age 35.1
years) consisted of 24 females and 46 males recruited from an embroidery factory. The controls were
matched for age, education, smoking and drinking habits. Exposure measurements for the control cohort
were not performed. The exposed cohort performed significantly worse than the control on tests of
finger tapping, mental arithmetic, two-digit searches, switiching attention, and visual reaction time. The
effect on performance persisted after the confounding factor of chronological age was controlled. Based
B-3
-------
on these neurobehavioral effects, the TWA of 0.033 mg/cu.m is designated as LOAEL. Using the TWA
and adjusting for occupational ventilation rates and workweek, the resultant LOAEL(HEC) is 0.012
mg/cu.m.
The above studies were taken together as evidence for a LOAEL based on neurobehavioral
effects of low-level mercury exposures. The LOAEL(HEC) levels calculated on measured air
concentration levels of the Ngim et al. (1992) and the Liang et al. (1993) studies bracket that calculated
based on the air concentrations measured by Fawer et al. (1983) as a median HEC level. Extrapolations
of blood levels, used as biological monitoring that accounts for variability in exposure levels, also
converge at 0.025 mg/cu.m as a TWA which results in the same HEC level. Thus, the TWA level of
0.025 mg/cu.m was used to represent the exposure for the synthesis of the studies described above.
Using this TWA and taking occupational ventilation rates and workweek into account results in a
LOAEL(HEC) of 0.009 mg/cu.m.
_I.B.3. UNCERTAINTY AND MODIFYING FACTORS (INHALATION RfC)
UF ~ An uncertainty factor of 10 was used for the protection of sensitive
human subpopulations (including concern for acrodynia - see Additional
Comments section) together with the use of a LOAEL. An uncertainty factor of
3 was used for lack of data base, particularly developmental and reproductive
studies.
MF ~ None
_I.B.4. ADDITIONAL COMMENTS (INHALATION RfC)
Probably the most widely recognized form of hypersensitivity to mercury poisoning is the
uncommon syndrome known as acrodynia, also called erythredema polyneuropathy or pink disease
(Warkany and Hubbard, 1953). Infantile acrodynia was first described in 1828, but adult cases have also
since been reported. While acrodynia has generally been associated with short-term exposures and with
urine levels of 50 ug/L or more, there are some cases in the literature in which mercury exposure was
known to have occurred, but no significant (above background) levels in urine were reported. There
could be many reasons for this, but the most likely is that urine levels are not a simple measure of body
burden or of target tissue (i.e., brain levels); however, they are the best means available for assessing the
extent of exposure. It was felt that the RfC level estimated for mercury vapor based on neurotoxicity of
chronic exposure in workers is adequate to protect children from risk of acrodynia because such
exposures of long duration would be expected to raise urine levels by only 0.12 ug/L against a
background level of up to 20 ug/L (i.e., such exposures would not add significantly to the background
level of mercury in those exposed).
Roels et al. (1987) investigated the relationships between the concentrations of metallic mercury
in air and levels monitored in blood or urine in workers exposed during manufacturing of dry alkaline
batteries. Breathing zone personal samples were used to characterize airborne mercury vapors. Total
mercury in blood and urine samples were analyzed using atomic absorption. The investigation controlled
B-4
-------
for several key factors including the use of reliable personal air monitoring, quality control for blood and
urine analyses, standardization of the urinary mercury concentration for creatinine concentration, and
stability of exposure conditions (examined subjects were exposed to mercury vapor for at least 1 year).
Strong correlations were found between the daily intensity of exposure to mercury vapor and the end of
workshift levels in blood (r=0.86; n=40) or urine (r=0.81; n=34). These relationships indicated a
conversion factor of 1:4.5 (airblood) and 1:1.22 (airurine as ug/g creatinine). These factors were used
to extrapolate blood or urine levels associated with effects in the reported studies to airborne mercury
levels.
Sensory and motor nerve conduction velocities were studied in 18 workers from a mercury cell
chlorine plant (Levine et al., 1982). Time-integrated urine Hg levels were used as an indicator of
mercury exposure. Using linearized regression analysis, the authors found that motor and sensory nerve
conduction velocity changes (i.e., prolonged distal latencies correlated with the time-integrated urinary
Hg levels in asymptomatic exposed workers) occurred when urinary Hg levels exceeded 25 ug/L. This
study demonstrates that mercury exposure can be associated with preclinical evidence of peripheral
neurotoxicity.
Singer et al. (1987) studied nerve conduction velocity of the median motor, median sensor and
sural nerves in 16 workers exposed to various inorganic mercury compounds (e.g., mercuric oxides,
mercurial chlorides, and phenyl mercuric acid) for an average of 7.3 +/- 7.1 years as compared with an
unexposed control group using t-tests. They found a slowing of nerve conduction velocity in motor, but
not sensory, nerves that correlated with increased blood and urine Hg levels and an increased number of
neurologic symptoms. The mean mercury levels in the exposed workers were 1.4 and 10 ug/L for blood
and urine, respectively. These urine levels are 2-fold less than those associated with peripheral
neurotoxicity in other studies (e.g., Levine et al., 1982). There was considerable variability in the data
presented by Singer et al. (1987), however, and the statistical analyses (t-test) were not as rigorous as
those employed by Levine et al. (1982) (linearized regression analysis). Furthermore, the subjects in the
Levine et al. (1982) study were asymptomatic at higher urinary levels than those reported to be
associated with subjective neurological complaints in the workers studied by Singer et al. (1987).
Therefore, these results are not considered to be as reliable as those reported by Levine et al. (1982).
Miller et al. (1975) investigated several subclinical parameters of neurological dysfunction in
142 workers exposed to inorganic mercury in either the chloralkali industry or a factory for the
manufacture of magnetic materials. They reported a significant increase in average forearm tremor
frequency in workers whose urinary Hg concentrations exceeded 50 ug/L as compared with unexposed
controls. Also observed were eyelid fasciculation, hyperactive deep-tendon reflexes and dermatographia,
but there was no correlation between the incidence of these findings and urinary Hg levels.
Roels et al. (1985) examined 131 male and 54 female workers occupationally exposed to
mercury vapor for an average duration of 4.8 years. Urinary mercury (52 and 37 ug/g creatinine for
males and females, respectively) and blood mercury levels (14 and 9 ug/L for males and females,
respectively) were recorded, but atmospheric mercury concentration was not provided. Symptoms
indicative of CNS disorders were reported but not related to mercury exposure. Minor renal tubular
effects were detected in mercury-exposed males and females and attributed to current exposure intensity
rather (urinary Hg >50 ug/g creatinine) than exposure duration. Male subjects with urinary mercury
levels of >50 ug/g creatinine exhibited preclinical signs of hand tremor. It was noted that females did not
exhibit this effect and that their urinary mercury never reached the level of 50 ug/g creatinine. A
companion study (Roels et al., 1987) related air mercury (Hg-air)levels to blood mercury (Hg-blood) and
B-5
-------
urinary mercury (Hg-U) values in 10 workers in a chloralkali battery plant. Duration of exposure was not
specified. A high correlation was reported for Hg-air and Hg-U for preshift exposure (r=0.70, p<0.001)
and post-shift (r=0.81, p<0.001) measurements. Based on these data and the results of their earlier
(1985) study, the investigators suggested that some mercury-induced effects may occur when Hg-U levels
exceed 50 ug/g creatinine, and that this value corresponds to a mercury TWA of about 40 ug/cu.m.
A survey of 567 workers at 21 chloralkali plants was conducted to ascertain the effects of
mercury vapor inhalation (Smith et al., 1970). Mercury levels ranged from <0.01 to 0.27 mg/cu.m and
chlorine concentrations ranged from 0.1 to 0.3 ppm at most of the working stations of these plants.
Worker exposure to mercury levels (TWA) varied, with 10.2% of the workers being exposed to <0.01
mg/cu.m, 48.7% exposed to 0.01 to 0.05 mg/cu.m, 25.6% exposed to 0.06 to 0.10 mg/cu.m and 4.8%
exposed to 0.24 to 0.27 mg/cu.m (approximately 85% were exposed to Hg levels less than or equal to 0.1
mg/cu.m). The duration of employment for the examined workers ranged from one year (13.3%) to >10
years (31%), with 55.7% of the workers being employed for 2 or 9 years. A group of 600 workers not
exposed to chlorine served as a control group for assessment of chlorine effects, and a group of 382
workers not exposed to either chlorine or mercury vapor served as the reference control group. A strong
positive correlation (p<0.001) was found between the mercury TWAs and the reporting of subjective
neuropsychiatric symptoms (nervousness, insomnia), occurrence of objective tremors, and weight and
appetite loss. A positive correlation (p<0.001) was also found between mercury exposure levels and
urinary and blood mercury levels of test subjects. No adverse alterations in cardiorespiratory,
gastrointestinal, renal or hepatic functions were attributed to the mercury vapor exposure. Additionally,
biochemical (hematologic data, enzyme activities) and clinical measurements (EKG, chest X-rays) were
no different between the mercury-exposed and non-exposed workers. No significant signs or symptoms
were noted for individuals exposed to mercury vapor concentrations less than or equal to 0.1 mg/cu.m.
This study provides data indicative of aNOAEL of 0.1 mg Hg/cu.m and a LOAEL of 0.18 mg Hg/cu.m.
In a followup study conducted by Bunn et al. (1986), however, no significant differences in the frequency
of objective or subjective findings such as weight loss and appetite loss were observed in workers
exposed to mercury at levels that ranged between 50 and 100 ug/L. The study by Bunn et al. (1986) was
limited, however, by the lack of information provided regarding several methodological questions such as
quality assurance measures and control of possible confounding variables.
The mercury levels reported to be associated with preclinical and symptomatic neurological
dysfunction are generally lower than those found to affect kidney function, as discussed below.
Piikivi and Ruokonen (1989) found no evidence of glomerular or tubular damage in 60
chloralkali workers exposed to mercury vapor for an average of 13.7 +/- 5.5 years as compared with their
matched referent controls. Renal function was assessed by measuring urinary albumin and
N-acetyl-beta-glucosaminidase (NAG) activity. The mean blood Hg level in the exposed workers was 14
ug/L and the mean urinary level was 17 ug/L. The authors extrapolated the NOAEL for kidney effects
based on these results of 0.025 mg/cu.m from blood levels using the conversion factor calculated by
Roelsetal. (1987).
Stewart et al. (1977) studied urinary protein excretion in 21 laboratory workers exposed to 10-50
ug/cu.m of mercury. Their urinary level of mercury was about 35 ug/L. Increased proteinuria was found
in the exposed workers as compared with unexposed controls. When preventive measure were instituted
to limit exposure to mercury, proteinuria was no longer observed in the exposed technicians.
B-6
-------
Lauwerys et al. (1983) found no change in several indices of renal function (e.g., proteinuria,
albuminuria, urinary excretion of retinol-binding protein, aminoaciduria, creatinine in serum,
beta-2-microglobulin in serum) in 62 workers exposed to mercury vapor for an average of 5.5 years. The
mean urinary Hg excretion in the exposed workers was 56 ug/g creatinine, which corresponds to an
exposure level of about 46 ug/cu.m according to a conversion factor of 1:1.22 (airurine [ug/g creatinine])
(Roels et al., 1987). Despite the lack of observed renal effects, 8 workers were found to have an
increased in serum anti-laminin antibodies, which can be indicative of immunological effects. In a
followup study conducted by Bernard et al. (1987), however, there was no evidence of increased serum
anti-laminin antibodies in 58 workers exposed to mercury vapor for an average of 7.9 years. These
workers had a mean urinary Hg excretion of 72 ug/g creatinine, which corresponds to an exposure levels
of about 0.059 mg/cu.m.
Stonard et al. (1983) studied renal function in 100 chloralkali workers exposed to inorganic
mercury vapor for an average of 8 years. No changes in the following urinary parameters of renal
function were observed at mean urinary Hg excretion rates of 67 ug/g creatinine: total protein, albumin,
alpha-1-acid glycoprotein, beta-2-microglobulin, NAG, and gamma-glutamyl transferase. When urinary
Hg excretion exceeded 100 ug/g creatinine, a small increase in the prevalence of higher activities of
NAG and gamma-glutamyl transferase was observed.
The mercury levels reported to be associated with preclinical and symptomatic neurological
dysfunction and kidney effects are lower than those found to pulmonary function, as discussed below.
McFarland and Reigel (1978) described the cases of 6 workers who were acutely exposed (4-8
hours) to calculated metallic mercury vapor levels of 1.1 to 44 mg/cu.m. These men exhibited a
combination of chest pains, dyspnea, cough, hemoptysis, impairment of pulmonary function (reduced
vital capacity), diffuse pulmonary infiltrates and evidence of interstitial pneumonitis. Although the
respiratory symptoms resolved, all six cases exhibited chronic neurological dysfunction, presumably as a
result of the acute, high-level exposure to mercury vapor.
Lilis et al. (1985) described the case of a 31-year-old male who was acutely exposed to high
levels of mercury vapor in a gold-extracting facility. Upon admission to the hospital, the patient
exhibited dyspnea, chest pain with deep inspiration, irregular infiltrates in the lungs and reduced
pulmonary function (forced vital capacity [FVC]). The level of mercury to which he was exposed is not
known, but a 24-hour urine collection contained 1900 ug Hg/L. Although the patient improved gradually
over the next several days, 11 months after exposure he still showed signs of pulmonary function
abnormalities (e.g., restriction and diffusion impairment).
Levin et al. (1988) described four cases of acute high-level mercury exposure during gold ore
purification. The respiratory symptoms observed in these four cases ranged from minimal shortness of
breath and cough to severe hypoxemia. The most severely affected patient exhibited mild interstitial lung
disease both radiographically and on pulmonary function testing. One patient had a urinary Hg level of
245 ug/L upon hospital admission. The occurrence of long-term respiratory effects in these patients
could not be evaluated since all but one refused follow-up treatment.
Ashe et al. (1953) reported that there was no histopathological evidence of respiratory damage in
24 rats exposed to 0.1 mg Hg/cu.m 7 hr/day, 5 days/week for 72 weeks. This is equivalent to a
NOAEL[HEC] of 0.07 mg/cu.m.
B-7
-------
Kishi et al. (1978) observed no histopathological changes in the lungs of rats exposed to 3
mg/cu.m of mercury vapor 3 hours/day, 5 days/week for 12-42 weeks.
Beliles et al. (1967) observed no histopathological changes in the lungs of pigeons exposed to 0.1
mg/cu.m of mercury vapor 6 hours/day, 5 days/week for 20 weeks.
Neurological signs and symptoms (i.e., tremors) were observed in 79 workers exposed to metallic
mercury vapor whose urinary mercury levels exceeded 500 ug/L. Short-term memory deficits were
reported in workers whose urine levels were less than 500 ug/L (Langolf et al., 1978).
Impaired performance in mechanical and visual memory tasks and psychomotor ability tests was
reported by Forzi et al. (1978) in exposed workers whose urinary Hg levels exceeded 100 ug/L.
Decreased strength, decreased coordination, increased tremor, decreased sensation and increased
prevalence of Babinski and snout reflexes were exhibited by 247 exposed workers whose urinary Hg
levels exceeded 600 ug/L. Evidence of clinical neuropathy was observed at urinary Hg levels that
exceeded 850 ug/L (Albers et al., 1988).
Preclinical psychomotor dysfunction was reported to occur at a higher incidence in 43 exposed
workers (mean exposure duration of 5 years) whose mean urinary excretion of Hg was 50 ug/L. Workers
in the same study whose mean urinary Hg excretion was 71 ug/L had a higher incidence of total
proteinuria and albuminuria (Roels et al., 1982).
Postural and intention tremor was observed in 54 exposed workers (mean exposure duration of
7.7 years) whose mean urinary excretion of Hg was 63 ug/L (Roels et al., 1989).
Verbeck et al. (1986) observed an increase in tremor parameters with increasing urinary
excretion of mercury in 21 workers exposed to mercury vapor for 0.5-19 years. The LOAEL for this
effect was a mean urinary excretion of 35 ug/g creatinine.
Rosenman et al. (1986) evaluated routine clinical parameters (physical exams, blood chemistry,
urinalysis), neuropsychological disorders, urinary NAG, motor nerve conduction velocities and
occurrence of lenticular opacities in 42 workers of a chemical plant producing mercury compounds. A
positive correlation (p<0.05 to p<0.001) was noted between urinary mercury (levels ranged from 100-250
ug/L) and the number of neuropsychological symptoms, and NAG excretions and the decrease in motor
nerve conduction velocities.
Evidence of renal dysfunction (e.g., increased plasma and urinary concentrations of
beta-galactosidase, increased urinary excretion of high-molecular weight proteins and a slightly increased
plasma beta-2-microglobulin concentration) was observed in 63 chloralkali workers. The incidence of
these effects increased in workers whose urinary Hg excretion exceeded 50 ug/g
creatinine (Buchet et al., 1980).
Increased urinary NAG levels were found in workers whose urinary Hg levels exceeded 50 ug/L
(Langworth et al., 1992).
An increase in the concentration of urinary brush border proteins (BB-50) was observed in 20
workers whose mean urinary Hg excretion exceeded 50 ug/g creatinine (Mutti et al., 1985).
B-8
-------
Foaet al. (1976) found that 15 out of 81 chloralkali workers exposed to 60-300 ug/cu.m mercury
exhibited proteinuria.
An increased excretion of beta-glutamyl transpeptidase, indicative of renal dysfunction, was
found in 509 infants dermally exposed to phenylmercury via contaminated diapers (Gotelli et al., 1985).
Berlin et al. (1969) exposed rats, rabbits and monkeys to 1 mg/cu.m of mercury vapor for 4 hours
and measured the uptake and distribution of mercury in the brain as compared with animals injected
intravenously with the same doses of mercury as mercuric salts. Mercury accumulated in the brain
following inhalation exposure to metallic mercury vapor at levels that were 10 times higher than those
observed following intravenous injection of the same dose of mercury as mercuric salts. These results
demonstrate that mercury is taken up by the brain following inhalation of the vapor at higher levels than
other forms of mercury and that this occurs in all species studied.
Limited animal studies concerning inhalation exposure to inorganic mercury are available. The
results of a study conducted by Baranski and Szymczyk (1973) were reported in an English abstract.
Adult female rats were exposed to metallic mercury vapor at 2.5 mg/cu.m for 3 weeks prior to
fertilization and during gestation days 7-20. A decrease in the number of living fetuses was observed in
the dams compared with unexposed controls, and all pups born to the exposed dams died by the sixth day
after birth. However, no difference in the occurrence of developmental abnormalities was observed
between exposed and control groups. The cause of death of the pups in the mercury-exposed group was
unknown, although an unspecified percentage of the deaths was attributed by the authors to a failure of
lactation in the dams. Death of pups was also observed in another experiment where dams were only
exposed prior to fertilization (to 2.5 mg/cu.m), which supports the conclusion that the high mortality in
the first experiment was due at least in part to poor health of the mothers. Without further information,
this study must be considered inconclusive regarding developmental effects.
The only other study addressing the developmental toxicology of mercury is the one reported in
abstract form by Steffek et al. (1987) and, as such, is included as a supporting study. Sprague-Dawley
rats (number not specified) were exposed by inhalation to mercury vapor at concentrations of 0.1, 0.5 or
1.0 mg/cu.m throughout the period of gestation (days 1-20) or during the period of organogenesis (days
10-15). The authors indicated the exposure protocols to be chronic and acute exposure, respectively. At
either exposure protocol, the lowest mercury level produced no detectable adverse effect. At 0.5
mg/cu.m, an increase in the number of resorptions (5/41) was noted for the acute group, and two of 115
fetuses exhibited gross cranial defects in the chronic group. At 1.0 mg/cu.m, the number of resorptions
was increased in acute (7/71) and chronic (19/38) groups and a decrease in maternal and fetal weights
also was detected in the chronic exposure group. No statistical analysis for these data was provided. A
LOAEL of 0.5 mg/cu.m is provided based on these data.
Mishinova et al. (1980) investigated the course of pregnancy and parturition in 349 women
exposed via inhalation to metallic mercury vapors (unspecified concentrations) in the workplace as
compared to 215 unexposed women. The authors concluded that the rates of pregnancy and labor
complication were high among women exposed to mercury and that the effects depended on "the length
of service and concentration of mercury vapors." Lack of sufficient details preclude the evaluation of
dose-response relationships.
In a questionnaire that assessed the fertility of male workers exposed to mercury vapor,
Lauwerys et al. (1985) found no statistically significant change in the observed number of children born
B-9
-------
to the exposed group compared with a matched control group. The urinary excretion of mercury in the
exposed workers ranged from 5.1 to 272.1 ug/g creatinine.
Another study found that exposure to metallic mercury vapor caused prolongation of estrus
cycles in animals. Baranski and Szymczyk (1973) reported that female rats exposed via inhalation to
mercury vapor at an average of 2.5 mg/cu.m, 6 hours/day, 5 days/week for 21 days experienced longer
estrus cycles than unexposed animals. In addition, estrus cycles during mercury exposure were longer
than normal estrus cycles in the same animals prior to exposure. Although the initial phase of the cycle
was protracted, complete inhibition of the cycle did not occur. During the second and third weeks of
exposure, these rats developed signs of mercury poisoning including restlessness, seizures and trembling
of the entire body. The authors speculated that the effects on the estrus cycle were caused by the action
of mercury on the CNS (i.e., damage to the hypothalamic regions involved in the control of estrus
cycling).
Renal toxicity has been reported following oral exposure to inorganic mercury salts in animals,
with the Brown-Norway rat appearing to be uniquely sensitive to this effect. These mercury-induced
renal effects in the Brown-Norway rat are the basis for the oral RfD for mercurial mercury. Several
investigators have produced autoimmune glomerulonephritis by administering HgC12 to Brown-Norway
rats (Druet et al, 1978).
The current OSHA standard for mercury vapor is 0.05 mg/cu.m. NIOSH recommends a TWA
Threshold Limit Value of 0.05 mg/cu.m for mercury vapor.
_I.B.5. CONFIDENCE IN THE INHALATION RfC
Study ~ Medium
Data Base ~ Medium
RfC - Medium
Due to the use of a sufficient number of human subjects, the inclusion of appropriate control
groups, the exposure duration, the significance level of the reported results and the fact that exposure
levels in a number of the studies had to be extrapolated from blood mercury levels, confidence in the key
studies is medium. The LOAEL values derived from these studies can be corroborated by other human
epidemiologic studies. The adverse effects reported in these studies are in accord with the
well-documented effects of mercury poisoning. The lack of human or multispecies
reproductive/developmental studies precludes assigning a high confidence rating to the data base and
inadequate quantification of exposure levels. Based on these considerations, the RfC for mercury is
assigned a confidence rating of medium.
_I.B.6. EPA DOCUMENTATION AND REVIEW OF THE INHALATION RfC
Source Document - U.S. EPA, 1995
This IRIS summary is included in The Mercury Study Report to Congress which was reviewed
by OHEA and EPA's Mercury Work Group in November 1994. An interagency review by scientists from
B-10
-------
other federal agencies took place in January 1995. The report was also reviewed by a panel of
non-federal external scientists in January 1995 who met in a public meeting on January 25-26. All
reviewers comments have been carefully evaluated and considered in the revision and fmalization of this
IMS summary. A record of these comments is summarized in the IMS documentation files.
Other EPA Documentation ~ None
Agency Work Group Review ~ 11/16/89, 03/22/90, 04/19/90
Verification Date - 04/19/90
_I.B.7. EPA CONTACTS (INHALATION RfC)
Annie M. Jarabek /NCEA - (919)541-4847
William F. Sette / OPP - (703)305-6375
REFERENCES
Albers, J.W., L.R. Kallenbach, L.J. Fine, et al. 1988. Neurological abnormalities associated with remote
occupational elemental mercury exposure. Ann. Neurol. 24(5): 651-659.
Ashe, W.F., E.J. Largent, F.R. Dutra, D.M. Hubbard and M. Blackstone. 1953. Behavior of mercury in
the animal organism following inhalation. Ind. Hyg. Occup. Med. 17: 19-43.
Baranski, B. and I. Szymczyk. 1973. [Effects of mercury vapor upon reproductive functions of female
white rats]. Med. Pr. 24(3): 249-261. (Czechoslovakia^
Beliles, R.P., R.S. Clark, P.R. Belluscio, C.L. Yuile and L.J. Leach. 1967. Behavioral effects in pigeons
exposed to mercury vapor at a concentration of 0.1 mg/cu.m. Am. Ind. Hyg. J. 28(5): 482-484.
Berlin, M., J. Fazackerley and G. Nordberg. 1969. The uptake of mercury in the brains of mammals
exposed to mercury vapor and to mercuric salts. Arch. Environ. Health. 18: 719-729.
Bernard, A.M., H.R. Roels, J.M. Foldart and R.L. Lauwerys. 1987. Search for anti-laminin antibodies in
the serum of workers exposed to cadmium, mercury vapour or lead. Int. Arch. Occup. Environ. Health.
59: 303-309.
Buchet, J.P., H. Roels, A. Bernard and R. Lauwerys 1980. Assessment of renal function of workers
exposed to inorganic lead, cadmium or mercury vapor. J. Occup. Med. 22(11): 741-750.
Bunn, W.B., C.M. McGill, T.E. Barber, J.W. Cromer and L.J. Goldwater. 1986. Mercury exposure in
chloralkali plants. Am. Ind. Hyg. Assoc. J. 47(5): 249-254.
Druet, P., E. Druet, F. Potdevin, et al. 1978. Immune type glomerulonephritis induced by HgC12 in the
Brown-Norway rat. Ann. Immunol. 129C: 777-792.
B-ll
-------
Fawer, R.F., Y. DeRibaupierre, M.P. Guillemin, M. Berode and M. Lob. 1983. Measurement of hand
tremor induced by industrial exposure to metallic mercury. J. Ind. Med. 40: 204-208.
Foa, V., L. Caimi, L. Amante, et al. 1976. Patterns of some lysosomal enzymes in the plasma and of
proteins in urine of workers exposed to inorganic mercury. Int. Arch. Occup. Environ. Health. 37:
115-124.
Forzi, M., M.G. Cassitto, C. Bulgheroni and V. Foa. 1978. Psychological measures in workers
occupationally exposed to mercury vapors: A validation study. In: Adverse Effects of Environmental
Chemicals and Psychotropic Drugs: Neurophysiological and Behavioral Tests, Vol. 2, H.J. Zimmerman,
Ed. Appleton-Century-Crofts, New York, NY. p. 165-171.
Gotelli, C.A., E. Astolfi, C. Cox, E. Cernichiari and T. Clarkson. 1985. Early biochemical effects of an
organic mercury funcigicide on infants: "Dose makes the poison". Science. 277:638-640.
Kishi, R., K. Hashimoto, S. Shimizu and M. Kobayashi. 1978. Behavioral changes and mercury
concentrations in tissues of rats exposed to mercury vapor. Toxicol. Appl. Pharmacol. 46(3): 555-566.
Langolf, G.D., D.B. Chaffin, R. Henderson and H.P. Whittle. 1978. Evaluation of workers exposed to
elemental mercury using quantitative tests of tremor and neuromuscular functions. Am. Ind. Hyg. Assoc.
J. 39: 976-984.
Langworth, S., C.G. Blinder, K.G. Sundquist and O. Vesterberg. 1992. Renal and immunological effects
of occupational exposure to inorganic mercury. Br. J. Ind. Med. 49: 394-401.
Lauwerys, R., A. Bernard, H. Roels, et al. 1983. Anti-laminin antibodies in workers exposed to mercury
vapour. Toxicol. Lett. 17: 113-116.
Lauwerys, R., H. Roels, P. Genet, G. Toussaint, A. Bouckaert and S. De Cooman. 1985. Fertility of
male workers exposed to mercury vapor or to manganese dust: A questionnaire study. Am. J. Ind. Med.
7(2): 171-176.
Levin, M., J. Jacobs and P.G. Polos. 1988. Acute mercury poisoning and mercurial pneumonitis from
gold ore purification. Chest. 94(3): 554-558.
Levine, S.P., G.D. Cavender, G.D. Langolf and J.W. Albers. 1982. Elemental mercury exposure:
Peripheral neurotoxicity. Br. J. Ind. Med. 39: 136-139.
Liang, Y-X., R-K. Sun, Y. Sun, Z-Q. Chen and L-H. Li. 1993. Psychological effects of low exposure to
mercury vapor: Application of a computer-administered neurobehavioral evaluation system. Environ.
Res. 60: 320-327.
Lilis, R., A. Miller and Y. Lerman. 1985. Acute mercury poisoning with severe chronic pulmonary
manifestations. Chest. 88(2): 306-309.
McFarland, R.B. and H. Reigel. 1978. Chronic mercury poisoning from a single brief exposure. J.
Occup. Med. 20(8): 532-534.
B-12
-------
Miller, J.M., D.B. Chaffin and R.G. Smith. 1975. Subclinical psychomotor and neuromuscular changes
in workers exposed to inorganic mercury. Am. Ind. Hyg. Assoc. J. 36: 725-733.
Mishonova, V.N., P.A. Stepanova and V.V. Zarudin. 1980. Characteristics of the course of pregnancy
and labor in women coming in contact with low concentrations of metallic mercury vapors in
manufacturing work places. Gig Tr Prof Zabol. Issue 2: 21-23.
Mutti, A., S. Lucertini, M. Fornari, et al. 1985. Urinary excretion of a brush-border antigen revealed by
monoclonal antibodies in subjects occupationally exposed to heavy metals. Heavy Met Environ.
International Conference 5th. Vol.1, p. 565-567.
Ngim, C.H., S.C. Foo, K.W. Boey and J. Jeyaratnam. 1992. Chronic neurobehavioral effects of
elemental mercury in dentists. Br. J. Ind. Med. 49: 782-790.
Piikivi, L. 1989. Cardiovascular reflexes and low long-term exposure to mercury vapor. Int. Arch.
Occup. Environ. Health. 61: 391-395.
Piikivi, L. and H. Hanninen. 1989. Subjective symptoms and psychological performance of
chlorine-alkali workers. Scand. J. Work Environ. Health. 15: 69-74.
Piikivi, L. and A. Ruokonen. 1989. Renal function and long-term low mercuryvapor exposure. Arch.
Environ. Health. 44(3): 146-149.
Piikivi, L. and U. Tolonen. 1989. EEG findings in chlor-alkali workers subjected to low long term
exposure to mercury vapor. Br. J. Ind. Med. 46: 370-375.
Roels, H., R. Lauwerys, J.P. Buchet, et al. 1982. Comparison of renal function and psychomotor
performance in workers exposed to elemental mercury. Int. Arch. Occup. Environ. Health. 50: 77-93.
Roels, H., J.P. Gennart, R. Lauwreys, J.P. Buchet, J. Malchaire and A. Bernard. 1985. Surveillance of
workers exposed to mercury vapor: validation of a previously proposed biological threshold limit value
for mercury concentration in urine. Am. J. Ind. Med. 7: 45-71.
Roels, H., S. Abdeladim, E. Ceulemans and R. Lauwreys. 1987. Relationships between the
concentrations of mercury in air and in blood or urine in workers exposed to mercury vapour. Ann.
Occup. Hyg. 31(2): 135-145.
Roels, H., S. Abdeladim, M. Braun, J. Malchaire and R. Lauwerys. 1989. Detection of hand tremor in
workers exposed to mercury vapor: A comparative study of three methods. Environ. Res. 49:152-165.
Rosenman, K.D., J.A. Valciukas, L. Glickman, B.R. Meyers and A. Cinotti. 1986. Sensitive indicators
of inorganic mercury toxicity. Arch. Environ. Health. 41(4): 208-215.
Singer, R., J.A. Valciukas and K.D. Rosenman. 1987. Peripheral neurotoxicity in workers exposed to
inorganic mercury compounds. Arch. Environ. Health. 42(4): 181-184.
Smith, R.G., A.J. Vorwald, L.S. Patil and T.F. Mooney, Jr. 1970. Effects of exposure to mercury in the
manufacture of chlorine. Am. Ind. Hyg. Assoc. J. 31(1): 687-700.
B-13
-------
Steffek, A.J., R. Clayton, C. Siew and A.C. Verrusio. 1987. Effects of elemental mercury vapor
exposure on pregnant Sprague-Dawley rats (abstract only). Teratology. 35: 59A.
Stewart, W.K., H.A. Guirgis, J. Sanderson and W. Taylor. 1977. Urinary mercury excretion and
proteinuria in pathology laboratory staff. Br. J. Ind. Med. 34: 26-31.
Stonard, M.D., B.V. Chater, D.P. Duffield, A.L. Nevitt, J.J. O'Sullivan and G.T. Steel. 1983. An
evaluation of renal function in workers occupationally exposed to mercury vapor. Int. Arch. Occup.
Environ. Health. 52: 177-189.
U.S. EPA. 1995. Mercury Study Report to Congress. Office of Research and Development, Washington
DC 20460. EPA/600/P-94/002Ab. External Review Draft.
Verbeck, M.M., H.J.A. Salle and C.H. Kemper. 1986. Tremor in workers with low exposure to metallic
mercury. Hyg. Assoc. J. 47(8): 559-562.
Warkany, J. andD.M. Hubbard. 1953. Acrodynia and mercury. J. Pediat. 42:365-386.
B-14
-------
_II. CARCINOGENICITY ASSESSMENT FOR LIFETIME EXPOSURE
Substance Name ~ Mercury, elemental
CASRN - 7439-97-6
Preparation Date ~ 5/24/94
_II.A. EVIDENCE FOR CLASSIFICATION AS TO HUMAN CARCINOGENICITY
_II.A.l WEIGHT-OF-EVIDENCE CLASSIFICATION
Classification ~ D; not classifiable as to human carcinogenicity
Basis ~ Based on inadequate human and animal data. Epidemiologic studies failed to show a correlation
between exposure to elemental mercury vapor and carcinogenicity; the findings in these studies were
confounded by possible or known concurrent exposures to other chemicals, including human
carcinogens, as well as lifestyle factors (e.g., smoking). Findings from genotoxicity tests are severely
limited and provide equivocal evidence that mercury adversely affects the number or structure of
chromosomes in human somatic cells.
_II.A.2 HUMAN CARCINOGENICITY DATA
Inadequate. A number of epidemiological studies were conducted that examined mortality
among elemental mercury vapor-exposed workers. Conflicting data regarding a correlation between
mercury exposure and an increased incidence of cancer mortalities have been obtained. All of the studies
have limitations that complicate interpretation of their results for associations between mercury exposure
and induction of cancer; increased cancer rates were attributable to other concurrent exposures or
lifestyle factors.
A retrospective cohort study examined mortality among 5663 white males who worked between
1953 and 1963 at a plant in Oak Ridge, Tennessee, where elemental mercury was used for lithium isotope
separation (Cragle et al., 1984). The workers were divided into three cohorts: exposed workers who had
been monitored on a quarterly basis for mercury levels in urine (n=2,133); workers exposed in the
mercury process section for whom urinalysis monitoring data were not collected (n=270); and unexposed
workers from other sections of the nuclear weapons production facility (n=3260). The study subjects
worked at least 4 months during 1953-1958 (a period when mercury exposures were likely to be high);
mortality data from death certificates were followed through the end of 1978. The mean age of the men
at first employment at the
facility was 33 years, and the average length of their employment was >16 years with a mean of 3.73
years of estimated mercury exposure. Air mercury levels were monitored beginning in 1955; during
1955 through the third quarter of 1956, air mercury levels were reportedly above 100 ug/cu.m in 30-80%
of the samples. Thereafter, air mercury levels decreased to concentrations below 100 ug/cu.m. The
mortality experience (i.e., the SMR) of each group was compared with the age-adjusted mortality
experience of the U.S. white male population. Among exposed and monitored workers, no significant
increases in mortality from cancer at any site were reported, even after the level or length of exposure
B-15
-------
was considered. A significantly lower mortality from all causes was observed. An excessive number of
deaths was reportedly due to lung cancer in the exposed and monitored workers (42 observed, 31.36
expected), but also in the unexposed workers (71 observed, 52.93 expected). The SMR for each group
was 1.34; the elevated incidence of lung cancer deaths was, therefore, attributed to some other factor at
the plant and/or to lifestyle factors (e.g., smoking) common to both the exposed and unexposed groups.
Study limitations include small cohort sizes for cancer mortality, which limited the statistical stability of
many comparisons.
Barregard et al. (1990) studied mortality and cancer morbidity between 1958 and 1984 in 1190
workers from eight Swedish chloralkali plants that used the mercury cell process in the production of
chlorine. The men included in the study had been monitored for urinary or blood mercury for more than
one year between 1946 and 1984. Vital status and cause of death were ascertained from the National
Population Register and the National Bureau of Statistics. The cancer incidence of the cohort was
obtained from the Swedish Cancer Register. The observed total mortality and cancer incidences were
compared with those of the general Swedish male population. Comparisons were not made between
exposed and unexposed workers. Mean urinary mercury levels indicated a decrease in exposure between
the 1950s and 1970s; the mean urinary mercury level was 200 ug/L during the 1950s, 150 ug/L during
the 1960s and 50 ug/L in the 1970s. Mortality from all causes was not significantly increased in exposed
workers. A significant increase in deaths from lung tumors was observed in exposed workers 10 years or
more after first exposure (rate ratio, 2.0; 95% CI, 1.0-3.8). Nine of the 10 observed cases of lung cancer
occurred among workers (457 of the 1190) possibly exposed to asbestos as well as to mercury. No dose
response was observed with respect to mercury exposure and lung tumors. This study is limited because
no quantitation was provided on smoking status, and results were confounded by exposure to asbestos.
Ahlbom et al. (1986) examined the cancer mortality during 1961-1979 of cohorts of Swedish
dentists and dental nurses aged 20-64 and employed in 1960 (3454 male dentists, 1125 female dentists,
4662 female dental nurses). Observed incidences were compared with those expected based on cancer
incidence during 1961-1979 among all Swedes employed during 1960 and the proportion of all Swedes
employed as dentists and dental nurses. Data were stratified by sex, age (5-year age groups) and county.
The incidence of glioblastomas among the dentists and dental nurses combined was significantly
increased compared to survival rates (SMR, 2.1; 95% CI, 1.3-3.4); the individual groups had apparently
elevated SMRs (2.0-2.5), but the 95%
confidence intervals of these groups included unity. By contrast, physicians and nurses had SMRs of
only 1.3 and 1.2, respectively. Exposure to mercury could not be established as the causative factor
because exposure to other chemicals and X-rays was not ruled out.
Amandus and Costello (1991) examined the association between silicosis and lung cancer
mortality between 1959 and 1975 in 9912 white male metal miners employed in the United States
between 1959 and 1961. Mercury exposures were not monitored. Exposures to specific metals among
the silicotic and nonsilicotic groups were analyzed separately. Lung cancer mortality in both silicotic
and nonsilicotic groups was compared with rates in white males in the U.S. population. Both silicotic
(n=l 1) and nonsilicotic mercury miners (n=263) had significantly increased lung cancer mortality (SMR,
14.03; 95% CI, 2.89-40.99 for silicotics. SMR, 2.66; 95% CI, 1.15-5.24 for nonsilicotics). The analysis
did not focus on mercury miners, and confounders such as smoking and radon exposure were not
analyzed with respect to mercury exposure. This study is also limited by the small sample size for
non-silicotic mercury miners.
B-16
-------
A case-control study of persons admitted to a hospital in Florence, Italy, with lung cancer
between 1981-1983 was performed to evaluate occupational risk factors (Buiatti et al., 1985). Cases
were matched with one or two controls (persons admitted to the hospital with diagnoses other than lung
cancer or suicide) with respect to sex, age, date of admission and smoking status. Women who had "ever
worked" as hat makers had a significantly increased risk of lung cancer. The duration of employment as
a hat maker averaged 22.2 years, and latency averaged 47.8 years. Workers in the Italian hat industry
were known to be occupationally exposed to mercury; however, the design of this study did not allow
evaluation of the relationship between cumulative exposure and cancer incidence. In addition,
interpretation of the results of this study is limited by the small sample size (only 6/376 cases reported
this occupation) and by exposure of hat makers to other pollutants including arsenic, a known lung
carcinogen.
Ellingsen et al. (1992) examined the total mortality and cancer incidence among 799 workers
employed for more than 1 year in two Norwegian chloralkali plants. Mortality incidence between 1953
and 1988 and cancer incidence between 1953 and 1989 were examined. Mortality and cancer incidence
were compared with that of the age-adjusted general male Norwegian population. No increase in total
cancer incidence was reported, but lung cancer was significantly elevated in the workers (rate ratio, 1.66;
95% CI, 1.0-2.6). No causal relationship can be drawn from the study between mercury exposure and
lung cancer because no correlation existed between cumulative mercury dose, years of employment or
latency time. Also, the prevalence of smoking was 10 20% higher in the exposed workers, and many
workers were also exposed to asbestos.
_II.A.3 ANIMAL CARCINOGENICITY DATA
Inadequate. Druckrey et al. (1957) administered 0.1 mL of metallic mercury to 39 male and
female rats (BD III and BD IV strains) via intraperitoneal injection. Among the rats surviving longer
than 22 months, 5/12 developed peritoneal sarcomas. The increase in the incidence of sarcomas was
observed only in those tissues that had been in direct contact with the mercury. Although severe kidney
damage was reported in all treated animals, no renal tumors or tumors at any site other than the peritoneal
cavity were observed.
_II.A.4 SUPPORTING DATA FOR CARCINOGENICITY
Cytogenetic monitoring studies of workers occupationally exposed to mercury by inhalation
provide very limited evidence that mercury adversely affects the number or structure of chromosomes in
human somatic cells. Popescu et al. (1979) compared four men exposed to elemental mercury vapor with
an unexposed group and found a statistically significant increase in the incidence of chromosome
aberrations in the WBCs from whole blood. Verschaeve et al. (1976) found an increase in aneuploidy
after exposure to low concentrations of vapor, but results could not be repeated in later studies
(Verschaeve et al., 1979). Mabille et al. (1984) did not find increases in structural chromosomal
aberrations of lymphocytes of exposed workers. Similarly, Barregard et al. (1991) found no increase in
the incidence or size of micronuclei and no correlation between micronuclei and blood or urinary
mercury levels of chloralkali workers. A statistically significant correlation was observed between
B-17
-------
cumulative exposure to mercury and micronuclei induction in T lymphocytes in exposed workers,
suggesting a genotoxic effect.
_II.B QUANTITATIVE ESTIMATE OF CARCINOGENIC RISK FROM ORAL EXPOSURE
None.
_II.C QUANTITATIVE ESTIMATE OF CARCINOGENIC RISK FROM INHALATION
EXPOSURE
None.
_II.D EPA DOCUMENTATION, REVIEW, AND CONTACTS (CARCINOGENICITY
ASSESSMENT)
_II.D.l EPA DOCUMENTATION
Source document --U.S. EPA, 1995
This IRIS summary is included in The Mercury Study Report to Congress which was reviewed
by OHEA and EPA's Mercury Work Group in November 1994. An interagency review by scientists from
other federal agencies took place in January 1995. The report was also reviewed by a panel of
non-federal external scientists in January 1995 who met in a public meeting on January 25-26. All
reviewers comments have been carefully evaluated and considered in the revision and finalization of this
IRIS summary. A record of these comments is
summarized in the IRIS documentation files.
_II.D.2 REVIEW (CARCINOGENICITY ASSESSMENT)
Agency Work Group Review - 01/13/88, 03/03/94
Verification Date - 03/03/94
_II.D.3 U.S. EPA CONTACTS (CARCINOGENICITY ASSESSMENT)
Rita Schoeny / NCEA ~ (513)569-7544
REFERENCES
B-18
-------
Ahlbom, A., S. Norell, Y. Rodvall and M. Nylander. 1986. Dentists, dental nurses, and brain tumours.
Br. Med. J. 292: 662.
Amandus, H. and J. Costello. 1991. Silicosis and lung cancer in U.S. metal miners. Arch. Environ.
Health. 46(2): 82-89.
Barregard, L., G. Sallsten and B. Jarvholm. 1990. Mortality and cancer incidence in chloralkali workers
exposed to inorganic mercury. Br. J. Ind. Med. 47(2): 99-104.
Barregard, L., B. Hogstedt, A. Schutz, A. Karlsson, G. Sallsten and G. Thiringer. 1991. Effects of
occupational exposure to mercury vapor on lymphocyte micronuclei. Scand. J. Work Environ. Health.
17: 263-268.
Buiatti, E., D. Kriebel, M. Geddes, M. Santucci and N. Pucci. 1985. A case control study of lung cancer
in Florence, Italy. I. Occupational risk factors. J. Epidemiol. Comm. Health. 39: 244-250.
Cragle, D.L., D.R. Hollis, J.R. Qualters, W.G. Tankersley and S.A. Fry. 1984. A mortality study of men
exposed to elemental mercury. J. Occup. Med. 26(11): 817-821.
Druckrey, H., H. Hamperl and D. Schmahl. 1957. Carcinogenic action of metallic mercury after
intraperitoneal administration in rats. Z. Krebsforsch. 61: 511-519. (Cited in U.S. EPA, 1985)
Ellingsen, D., A. Andersen, H.P. Nordhagen, J. Efskind and H. Kjuus. 1992. Cancer incidence and
mortality among workers exposed to mercury in the Norwegian chloralkali industry. 8th International
Symposium on Epidemiology in Occupational Health, Paris, France, September 10-12, 1991. Rev.
Epidemiol. Sante Publique. 40(1): S93-S94.
Mabille, V., H. Roels, P. Jacquet, A. Leonard and R. Lauwerys. 1984. Cytogenetic examination of
leucocytes of workers exposed to mercury vapor. Int. Arch. Occup. Environ. Health. 53: 257-260.
Popescu, H.I., L. Negru and I. Lancranjan. 1979. Chromosome aberrations induced by occupational
exposure to mercury. Arch. Environ. Health. 34(6): 461-463.
U.S. EPA. 1980. Ambient Water Quality Criteria Document for Mercury. Prepared by the Office of
Health and Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH
for the Office of Water Regulation and Standards, Washington, DC. EPA/440/5-80/058. NTIS PB
81-117699.
U.S. EPA. 1984a. Mercury Health Effects Update: Health Issue Assessment. Final Report. Prepared
by the Office of Health and Environmental Assessment, Environmental Criteria and Assessment Office,
Cincinnati, OH for the Office of Air Quality Planning and Standards, Research Triangle Park, NC.
EPA/600/8-84/019F. NTIS PB81-85-123925.
U.S. EPA. 1984b. Health Effects Assessment for Mercury. Prepared by the Office of Health and
Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH for the
Office of Emergency and Remedial Response, Washington, DC. EPA/540/1086/042. NTIS
PB86-134533/AS.
B-19
-------
U.S. EPA. 1985. Drinking Water Criteria Document for Mercury. Prepared by the Office of Health and
Environmental Assessment Office, Cincinnati, OH for the Office of Drinking Water, Washington, DC.
EPA/600/X-84/178. NTIS PB86-117827.
U.S. EPA. 1988. Drinking Water Criteria Document for Inorganic Mercury. Prepared by the Office of
Health and Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH
for the Office of Drinking Water, Washington, DC. EPA/600/X-84/178. NTIS PB89-192207.
U.S. EPA. 1993. Summary Review of Health Effects Associated with Mercuric Chloride: Health Issue
Assessment (Draft). Prepared by the Office of Health and Environmental Assessment, Environmental
Criteria and Assessment Office, Cincinnati, OH for the Office of Air Quality Planning and Standards,
Research Triangle Park, NC. EPA/600/R-92/199.
U.S. EPA. 1995. Mercury Study Report to Congress. Office of Research and Development,
Washington, DC. External Review Draft. EPA/600/P-94/002Ab.
Verschaeve, L., M. Kirsch-Volders, C. Susanne et al. 1976. Genetic damage induced by occupationally
low mercury exposure. Environ. Res. 12:303-316.
Verschaeve, L., J.P. Tassignon, M. Lefevre, P. De Stoop and C. Susanne. 1979. Cytogenetic
investigation on leukocytes of workers exposed to metallic mercury. Environ. Mutagen. 1: 259-268.
B-20
-------
_I.A REFERENCE DOSE FOR CHRONIC ORAL EXPOSURE (RfD)
Substance Name ~ Mercuric chloride (HgC12)
CASRN - 7487-94-7
Preparation Date - 11/01/88
_I.A.l ORAL RfD SUMMARY
Critical Effect Experimental Doses* UF MF RfD
Autoimmune effects NOAEL: None 1000 1 3E-4
mg/kg-day
Rat Subchronic LOAEL: 0.226 mg/kg-day
Feeding and
Subcutaneous LOAEL: 0.317 mg/kg-day
Studies
LOAEL: 0.633 mg/kg-day
U.S. EPA, 1987
* Conversion Factors and Assumptions ~ Dose conversions in the three studies employed a 0.739 factor
for HgC12 to Hg2+, a 100% factor for subcutaneous (s.c.) to oral route of exposure, and a time-weighted
average for days/week of dosing. This RfD is based on the back calculations from a Drinking Water
Equivalent Level (DWEL), recommended to and subsequently adopted by the Agency, of 0.010 mg/L:
(RfD = 0.010 mg/L x 2 L/day/70 kg bw = 0.0003 mg/kg bw/day). The LOAEL exposure levels, utilized
in the three studies selected as the basis of the recommended DWEL, are from Druet et al. (1978),
Bernaudin et al. (1981) and Andres (1984), respectively.
_I.A.2 PRINCIPAL AND SUPPORTING STUDIES (ORAL RfD)
U.S. EPA. 1987. Peer Review Workshop on Mercury Issues. Summary Report. Environmental Criteria
and Assessment Office, Cincinnati, OH. October 26-27'.
On October 26-27, 1987, a panel of mercury experts met at a Peer Review Workshop on Mercury
Issues in Cincinnati, Ohio, and reviewed outstanding issues concerning the health effects and risk
assessment of inorganic mercury (U.S. EPA, 1987). The following five consensus conclusions and
recommendations were agreed to as a result of this workshop:
1) The most sensitive adverse effect for mercury risk assessment is formation of
mercuric-mercury-induced autoimmune glomerulonephritis. The production and
deposition of IgG antibodies to the glomerular basement membrane can be considered
the first step in the formation of this mercuric-mercury-induced autoimmune
glomerulonephritis.
B-21
-------
2) The Brown Norway rat should be used for mercury risk assessment. The Brown Norway
rat is a good test species for the study of Hg2+-induced autoimmune glomerulonephritis.
The Brown Norway rat is not unique in this regard (this effect has also been observed in
rabbits).
3) The Brown Norway rat is a good surrogate for the study of mercury-induced kidney
damage in sensitive humans. For this reason, the uncertainty factor used to calculate
criteria and health advisories (based on risk assessments using the Brown Norway rat)
should be reduced by 10-fold.
4) Hg2+ absorption values of 7% from the oral route and 100% from the s.c. route should
be used to calculate criteria and health advisories.
5) A DWEL of 0.010 mg/L was recommended based on the weight of evidence from the
studies using Brown Norway rats and limited human tissue data.
Three studies using the Brown Norway rat as the test strain were chosen from a larger selection of studies
as the basis for the panel's recommendation of 0.010 mg/L as the DWEL for inorganic mercury. The
three studies are presented below for the sake of completeness. It must be kept in mind, however, that
the recommended DWEL of 0.010 mg/L and back calculated oral RfD of 0.0003 mg/kg-day were arrived
at from an intensive review and workshop discussions of the entire inorganic mercury data base, not just
from one study.
In the Druet et al. (1978) study, the duration of exposure was 8-12 weeks; s.c. injection was used
instead of oral exposure. In this study the development of kidney disease was evaluated. In the first
phase the rats developed anti-GBM antibodies. During the second phase, which is observed after 2-3
months, the patterns of fixation of antisera changed from linear to granular as the disease progressed.
The immune response was accompanied by proteinuria and in some cases by a nephrotic syndrome.
Both male and female Brown Norway rats 7-9 weeks of age were divided into groups of 6-20
animals each. The numbers of each sex were not stated. The animals received s.c. injections of mercuric
chloride (HgC12) 3 times weekly for 8 weeks, with doses of 0, 100, 250, 500, 1000 and 2000 ug/kg. An
additional group was injected with a 50 ug/kg dose for 12 weeks. Antibody formation was measured by
the use of kidney cryostat sections stained with a fluoresceinated sheep anti-rat IgG antiserum. Urinary
protein was assessed by the biuret method (Druet et al., 1978).
Tubular lesions were observed at the higher dose levels. Proteinuria was reported at doses of 100
ug/kg and above, but not at 50 ug/kg. Proteinuria was considered a highly deleterious effect, given that
affected animals developed hypoalbuminemia and many died. Fixation of IgG antiserum was detected in
all groups except controls (Druet et al., 1978).
Bernaudin et al. (1981) reported that mercurials administered by inhalation or ingestion to Brown
Norway rats developed a systemic autoimmune disease. The HgC12 ingestion portion of the study
involved the forcible feeding of either 0 or 3000 ug/kg-week of HgC12 to male and female Brown
Norway rats for up to 60 days. No abnormalities were reported using standard histological techniques in
either experimental or control rats. Immunofluorescence histology revealed that 80% (4/5) of the
mercuric-exposed rats were observed with a linear IgG deposition in the glomeruli after 15 days of
exposure. After 60 days of HgC12 exposure, 100% (5/5) of the rats were seen with a mixed linear and
B-22
-------
granular pattern of IgG deposition in the glomeruli and granular IgG deposition in the arteries. Weak
proteinuria was observed in 60% (3/5) of the rats fed HgC12 for 60 days. The control rats were observed
to have no deposition of IgG in the glomeruli or arteries as well as normal urine protein concentrations.
Andres (1984) administered HgC12 (3 mg/kg in 1 mL of water) by gavage to five Brown Norway
rats and two Lewis rats twice a week for 60 days. A sixth Brown Norway rat was given only 1 mL of
water by gavage twice a week for 60 days. All rats had free access to tap water and pellet food. After
2-3 weeks of exposure, the Brown Norway HgC12-treated rats started to lose weight and hair. Two of the
HgC12-treated Brown Norway rats died 30-40 days after beginning the study. No rats were observed to
develop detectable proteinuria during the 60-day study. The kidneys appeared normal in all animals
when evaluated using standard histological techniques, but examination by immunofluorescence showed
deposits of IgG present in the renal glomeruli of only the mercuric-treated Brown Norway rats. The
Brown Norway treated rats were also observed with mercury-induced morphological lesions of the ileum
and colon with abnormal deposits of IgA in the basement membranes of the intestinal glands and of IgG
in the basement membranes of the lomina propria. All observations in the Lewis rats and the control
Brown Norway rat appeared normal.
_I.A.3 UNCERTAINTY AND MODIFYING FACTORS (ORAL RfD)
UF ~ An uncertainty factor of 1000 was applied to the animal studies using Brown Norway rats as
recommended in U.S. EPA (1987). An uncertainty factor was applied for LOAEL to NOAEL
conversion: 10 for use of subchronic studies and a combined 10 for both animal to human and sensitive
human populations.
MF ~ None
_I.A.4 ADDITIONAL STUDIES / COMMENTS (ORAL RfD)
Kazantzis et al. (1962) performed renal biopsies in 2 (out of 4) workers with nephrotic syndrome
who had been occupationally exposed to mercuric oxide, mercuric acetate and probably mercury vapors.
Investigators reported that the nephrotic syndrome observed in 3 of the 4 workers may have been an
idiosyncratic reaction since many other workers in a factory survey had similarly high levels of urine
mercury without developing proteinuria. This conclusion was strengthened by work in Brown Norway
rats indicating a genetic (strain) susceptibility and that similar mercury-induced immune system
responses have been seen in affected humans and the susceptible Brown Norway rats (U.S. EPA, 1987).
The only chronic ingestion study designed to evaluate the toxicity of mercury salts was reported
by Fitzhugh et al. (1950). In this study, rats of both sexes (20-24/group) were given 0.5, 2.5, 10, 40 or
160 ppm mercury as mercuric acetate in their food for up to 2 years. Assuming food consumption was
equal to 5% bw/day, the daily intake would have been 0.025, 0.125, 0.50, 2.0 and 8.0 mg/kg for the five
groups, respectively. At the highest dose level, a slight depression of body weight was detected in male
rats only. The statistical significance of this body-weight depression was not stated. Kidney weights
were significantly (p<0.05) increased at the 2.0 and 8.0 mg/kg dose levels. Pathological changes
originating in the proximal convoluted tubules of the kidneys were also noted, with more severe effects
B-23
-------
in females than males. The primary weaknesses of this study were (1) the lack of reporting on which
adverse effects were observed with which dosing groups and (2) that the most sensitive strain, the Brown
Norway rat, was not used for evaluating the mercury-induced adverse health effects.
NTP (1993) conducted subchronic and chronic gavage toxicity studies on Fischer 344 rats and
B6C3F1 mice to evaluate the effects of HgC12, and the kidney appeared to be the major organ affected.
In the 6-month study, Fischer 344 rats (10/sex /group) were administered 0, 0.312, 0.625, 1.25, 2.5 or 5
mg/kg-day of HgC12 (0.23, 0.46, 0.92, 1.9 and 3.7 mg/kg-day) 5 days/week by gavage. Survival was not
affected, although body-weight gains were decreased in males at high dose and in females at or above the
0.46 mg/kg-day dose. Absolute and relative kidney weights were significantly increased in both sexes
with exposure to at least 0.46 mg/kg-day. In males, the incidence of nephropathy was 80% in the
controls and 100% for all treated groups; however, severity was minimal in the controls and two
low-dose groups and minimal to mild in the 0.92 mg/kg-day group and higher. In females, there was a
significant increased incidence of nephropathy only in the high-dose group (4/10 with minimal severity).
Nephropathy was characterized by foci of tubular regeneration, thickened tubular basement membrane
and scattered dilated tubules containing hyaline casts. No treatment-related effects were
observed in the other organs; however, histopathology on the other organs was performed only on control
and high-dose rats.
B6C3F1 mice (10/sex/group) were administered 0, 1.25, 2.5, 5, 10 or 20 mg/kg-day HgC12 (0,
0.92, 1.9,3.7, 7.4 or 14.8 mg/kg-day) 15 days/week by gavage for 6 months (NTP 1993). A decrease in
body-weight gain was reported in only the males at the highest dose tested. Significant increases
occurred in absolute kidney weights of male mice at 3.7 mg/kg-day or greater and relative kidney weights
of male mice at 7.4 and 14.8 mg/kg-day doses. The kidney weight changes corresponded to an increased
incidence of cytoplasmic vacuolation of renal tubule epithelium in males exposed to at least 3.7
mg/kg-day. The exposed female mice did not exhibit any histopathologic changes in the kidneys.
In the 2-year NTP study, Fischer 344 rats (60/sex/group) were administered 0, 2.5 and 5
mg/kg-day HgC12 (1.9 and 3.7 mg/kg-day) 5 days week by gavage (NTP, 1993). After 2 years, survival
was reduced in only the treated male rat groups compared with the control. Mean body weights were
decreased in both male and female treated groups. After 2 years, an increased incidence of nephropathy
of moderate-to-marked severity and increased incidence of tubule
hyperplasia was observed in the kidneys of exposed males compared with the controls. The control
males exhibited nephropathy, primarily of mild-to-moderate severity. Hyperparathyroidism,
mineralization of various tissues and fibrous osteodystrophy were observed and considered secondary to
the renal impairment. No significant differences were found in renal effects between exposed and
control females. Other nonneoplastic effects included an increased incidence of forestomach hyperplasia
in the exposed males and high-dose females.
NTP (1993) also administered to B6C3F1 mice (60/sex/group) daily oral gavage doses of 0, 5 or
10 mg/kg-day HgC12 (0, 3.7 and 7.4 mg/kg-day) 5 days/week by gavage for 2 years. Survival and body
weights of mice were slightly lower in HgC12-treated mice compared with controls. Absolute kidney
weights were significantly increased in the treated males, while relative kidney weights were
significantly increased in high-dose males and both low- and high-dose females. Histopathology
revealed an increase in the incidence and severity of nephropathy in exposed males and an increase in the
incidence of nephropathy in exposed females. Nephropathy was defined as foci of proximal convoluted
tubules with thickened basement membrane and basophilic cells with scant cytoplasm. Some affected
B-24
-------
convoluted tubules contained syaline casts. Also, an increase in nasal cavity inflammation (primarily
infiltration of granulocytes in nasal mucosa) was observed in the exposed animals.
Gale and Perm (1971) studied the teratogenic effects of mercuric acetate on Syrian golden
hamsters. Single doses of 2, 3 or 4 mg/kg were injected by the i.v. route on day 8 of gestation. Growth
retardation, increased resorption rates and edema of the fetuses were found at all three dose levels, while
an increase in the number of abnormalities was detected at the two higher doses. In a more recent study,
Gale (1981) compared the embryotoxic effects of a single s.c. dose of 15 mg/kg mercuric acetate on the
eighth day of gestation in five inbred strains and one noninbred strain of Syrian hamsters. While strain
differences were apparent, a variety of abnormalities were reported in all the strains. Gale (1974) also
compared the relative effectiveness of different exposure routes in Syrian hamsters. The following
sequence of decreasing efficacy was noted for mercuric acetate; i.p. > i.v. > s.c. > oral. The lowest doses
used, 2 mg/kg for i.p. and 4 mg/kg for the other three routes, were all effective in causing increased
resorption and percent abnormalities.
In male mice administered a single i.p. dose of 1 mg/kg HgC12, fertility decreased between days
28 and 49 post treatment with no obvious histological effects noted in the sperm (Lee and Dixon, 1975).
The period of decreased fertility indicated that spermatogonia and premeiotic spermatocytes were
affected. The effects were less severe than following a similar dose of methyl mercury. A single i.p.
dose of 2 mg/kg HgC12 in female mice resulted in a significant decrease in the total number of implants
and number of living embryos and a significant increase in the percentage of dead implants (Suter, 1975).
These effects suggest that mercury may be a weak inducer of dominant lethal mutations.
_I.A.5 CONFIDENCE IN THE ORAL RfD
Study - N/A
Data Base ~ High
RfD - High
No one study was found adequate for deriving an oral RfD; however, based on the weight of
evidence from the studies using Brown Norway rats and the entirety of the mercuric mercury data base,
an oral RfD of high confidence results.
_I.A.6 EPA DOCUMENTATION AND REVIEW OF THE ORAL RfD
Source Document -- U.S. EPA, 1988
This IRIS summary is included in The Mercury Study Report to Congress, which was reviewed
by OHEA and EPA's Mercury Work Group in November 1994. An interagency review by scientists from
other federal agencies took place in January 1995. The report was also reviewed by a panel of
non-federal external scientists in January 1995 who met in a public meeting on January 25-26. All
reviewers comments have been carefully evaluated and considered in the revision and finalization of this
IRIS summary. A record of these comments is
summarized in the IRIS documentation files.
B-25
-------
Other Docmentation -- U.S. EPA, 1987
Agency Work Group Review - 08/05/85, 02/05/86, 08/19/86, 11/16/88
Verification Date - 11/16/88
_I.A.7 EPA CONTACTS (ORAL RfD)
W. Bruce Peirano /NCEA - (513)569-7540
Krishan Khanna / OST -- (202)260-7588
REFERENCES
Andres, P. 1984. IgA-IgG disease in the intestine of Brown Norway rats ingesting mercuric chloride.
Clin. Immunol. Immunopathol. 30: 488-494.
Bernaudin, J.F., E. Druet, P. Druet and R. Masse. 1981. Inhalation or ingestion of organic or inorganic
mercurials produces auto-immune disease in rats. Clin. Immunol. Immunopathol. 20: 129-135.
Druet, P., E. Druet, F. Potdevin and C. Sapin. 1978. Immune type glomerulonephritis induced by HgC12
in the Brown Norway rat. Ann. Immunol. 129C: 777-792.
Fitzhugh, O.G., A.A. Nelson, E.P. Laug and P.M. Kunze. 1950. Chronic oral toxicants of
mercuric-phenyl and mercuric salts. Arch. Ind. Hyg. Occup. Med. 2: 433-442.
Gale, T.F. 1974. Embryopathic effects of different routes of administration of mercuric acetate in the
hamster. Environ. Res. 8: 207-213.
Gale, T.F. 1981. The embryotoxic response produced by inorganic mercury in different strains of
hamsters. Environ. Res. 24: 152-161.
Gale, T. and V. Perm. 1971. Embryopathic effects of mercuric salts. Life Sci. 10(2): 1341-1347.
Kazantzis, G., K.F.R. Schiller, A.W. Asscher and R.G. Drew. 1962. Albuminuria and the nephrotic
syndrome following exposure to mercury and its compounds. Q. J. Med. 31(124): 403-419.
Lee, I.D. and R.L. Dixon. 1975. Effects of mercury on spermatogenesis studied by velocity
sedimentation, cell separation and serial mating. J. Pharmacol. Exp. Ther. 194(1): 171-181.
NTP (National Toxicology Program). 1993. Toxicology and carcinogenesis studies of mercuric chloride
(CAS No. 7487-94-7) in F344 rats and B3C3F1 mice (gavage studies). NTP Technical Report Series No.
408. National Toxicology Program, U.S. Department of Health and Human Services, Public Health
Service, National Institutes of Health, Research Triangle Park, NC.
B-26
-------
Suter, K.E. 1975. Studies on the dominant lethal and fertility effects of he heavy metal compounds
methyl mercuric hydroxide, mercuric chloride, and cadmium chloride in male and female mice. Mutat.
Res. 30: 365-374.
U.S. EPA. 1987. Peer Review Workshop on Mercury Issues. Environmental Criteria and Assessment
Office, Cincinnati, OH. Summary report. October 26-27'.
U.S. EPA. 1988. Drinking Water Criteria Document for Inorganic Mercury. Prepared by the Office of
Health and Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH
for the Office of Drinking Water, Washington, DC. EPA/600/X-84/178. NTIS PB89-192207.
B-27
-------
_II. CARCINOGENICITY ASSESSMENT FOR LIFETIME EXPOSURE
Substance Name ~ Mercuric Chloride
CASRN - 7487-94-7
Preparation Date ~ 5/24/94
_II.A EVIDENCE FOR CLASSIFICATION AS TO HUMAN CARCINOGENICITY
_II.A.l WEIGHT-OF-EVIDENCE CLASSIFICATION
Classification ~ C; possible human carcinogen
Basis ~ Based on the absence of data in humans and limited evidence of carcinogenicity in rats and mice.
Focal papillary hyperplasia and squamous cell papillomas in the forestomach as well as thyroid follicular
cell adenomas and carcinomas were observed in male rats gavaged with mercuric chloride for 2 years.
The relevance of the forestomach papillomas to assessment of cancer in humans is questionable because
no evidence indicated that the papillomas progressed to malignancy. The relevance of the increase in
thyroid tumors has also been questioned because these tumors are generally considered to be secondary
to hyperplasia; this effect was not observed in the high-dose males. It should also be noted that the
authors considered the doses used in the study to exceed the MTD for male rats. In the same study,
evidence for increases in squamous cell papillomas in the forestomach of female rats was equivocal. In a
second study, equivocal evidence for renal adenomas and adenocarcinomas was observed in male mice;
there was a significant positive trend. This tumor type is rare in mice, and the increase in incidence was
statistically significant when compared with historic controls. Two other nonpositive lifetime rodent
studies were considered inadequate. Mercuric chloride showed mixed results in a number of
genotoxicity assays.
_II.A.2 HUMAN CARCINOGENICITY DATA
None. No data are available on the carcinogenic effects of mercuric chloride in humans.
_II.A.3 ANIMAL CARCINOGENICITY DATA
Limited. The results from a dietary study in rats and mice show equivocal evidence for
carcinogenic activity in male mice and female rats and some evidence for carcinogenic activity in male
rats. Two other dietary studies did not show any evidence for carcinogenicity, but these studies are
limited by inadequacies in the data and experimental design, including the small number of animals/dose
and/or a lack of complete histopathological examinations.
Mercuric chloride (purity >99%) was administered by gavage in water at doses of 0, 2.5 or 5
(mg/kg)/day (0, 1.9 and 3.7 (mg/kg)/day) to 60 F344 rats/sex/group, 5 days/week for 104 weeks (NTP,
B-28
-------
1993). An interim sacrifice (10/sex/dose) was conducted after 15 months of exposure. Complete
histopathological examinations were performed on all animals found dead, killed in extremis, or killed by
design. Survival after 24 months was lower in low- and high-dose males at a statistically significant rate;
survival was 43, 17 and 8% in control, low-, and high-dose males, respectively, and 58, 47 and 50% in
control, low-, and high-dose females, respectively. During the second year of the study, body weight
gains of low- and high-dose males were 91 and 85% of controls, respectively, and body weight gains of
low- and high-dose females were 90 and 86% of controls, respectively. At study termination,
nephropathy was evident in almost all male and female rats including controls, but the severity was much
greater in treated males. The incidence of "marked" nephropathy was 6/50, 29/50 and 29/50 in control,
low- and high-dose males, respectively. Squamous cell papillomas of the forestomach showed a
statistically significant positive trend with dose by life table adjusted analysis; the incidences were 0/50,
3/50 and 12/50 in control, low- and high-dose males, respectively. For females, the incidence was 0/50,
0/49 and 2/50 in control, low- and high-dose groups, respectively. These neoplasms are rare in male rats
and occurred in only 1/264 historical controls. The incidence of papillary hyperplasia of the stratified
squamous epithelium lining of the forestomach was elevated at a statistically significant rate in all dosed
males (3/49, 16/50 and 35/50 in control, low- and high-dose males, respectively) and in high-dose
females (5/50, 5/49 and 20/50 in control, low-and high-dose females, respectively). The incidence of
thyroid follicular cell carcinomas, adjusted for survival, showed a significant positive trend in males; the
incidence was 1/50, 2/50 and 6/50 in control, low- and high-dose groups, respectively. The combined
incidence of thyroid follicular cell neoplasms (adenoma and/or carcinoma) was not significantly
increased (2/50, 6/50 and 6/50 in control, low- and high-dose males, respectively). In female rats a
significant decrease in the incidence of mammary gland fibroadenomas was observed (15/50, 5/48 and
2/50 in control, low- and high-dose females, respectively). The high mortality in both groups of treated
males indicates that the MTD was exceeded in these groups and limits the value of the study for
assessment of carcinogenic risk. NTP (1993) considered the forestomach tumors to be of limited
relevance to humans because the tumors did not appear to progress to malignancy. NTP (1993) also
questioned the relevance of the thyroid carcinomas because these neoplasms are usually seen in
conjunction with increased incidences of hyperplasia and adenomas. In this study, however, no
increases in hyperplasia or adenomas were observed. Hyperplasia incidence was 2/50, 4/50 and 2/50 in
control, low- and high-dose males, respectively; adenoma incidence was 1/50, 4/50 and 0/50 in control,
low- and high-dose males, respectively.
In the same study, mercuric chloride was administered by gavage in water at doses of 0, 5 or 10
(mg/kg)/day (0, 3.7 and 7.4 (mg/kg)/day) to 60 B6C3F1 mice/sex/group 5 days/week for 104 weeks
(NTP, 1993). An interim sacrifice (10/sex/dose) was conducted after 15 months of exposure. Terminal
survival and body weight gain were not affected in either sex by the administration of mercuric chloride.
It should be noted that survival of high-dose females was lower than controls; female survival rates were
82, 70 and 62% in control, low- and high-dose females, respectively. Female mice exhibited a significant
increase in the incidence of nephropathy (21/49, 43/50 and 42/50 in control, low- and high-dose females,
respectively). Nephropathy was observed in 80-90% of the males in all groups. The severity of
nephropathy increased with increasing dose. The incidence of renal tubular hyperplasia was 1/50, 0/50
and 2/49 in control, low- and high-dose males. The combined incidence of renal tubular adenomas and
adenocarcinomas was 0/50, 0/50 and 3/49 in control, low- and high-dose males, respectively. Although
no tumors were seen in the low-dose males, a statistically significant positive trend for increased
incidence with increased dose was observed. These observations were considered important because
renal tubular hyperplasia and tumors in mice are rare. The 2-year historical incidence of renal tubular
adenomas or adenocarcinomas in males dosed by gavage with water was 0/205, and only 4 of the nearly
400 completed NTP studies have shown increased renal tubular neoplasms in mice. Data from this study
B-29
-------
were not statistically compared with historical control data by NTP. EPA's analysis of the reported data
with Fisher's Exact test showed that the incidence of renal tubular adenomas or adenocarcinomas in the
high-dose males was significantly elevated when compared with historical controls (Rice and Knauf,
1994).
A 2-year feeding study in rats (20 or 24/sex/group; strain not specified) was conducted in which
mercuric acetate was administered in the diet at doses of 0, 0.5, 2.5, 10, 40 and 160 ppm (0, 0.02, 0.1,
0.4, 1.7 and 6.9 (mg Hg/kg)/day (Fitzhugh et al., 1950). Survival was not adversely affected in the study.
Increases in kidney weight and renal tubular lesions were observed at the two highest doses. No
statement was made in the study regarding carcinogenicity. This study was not intended to be a
carcinogenicity assay, and the number of animals/dose was rather small. Histopathological analyses
were conducted on only 50% of the animals (complete histopathology conducted on only 31% of the
animals examined), and no quantitation of results or statistical analyses were performed.
No increase in tumor incidence was observed in a carcinogenicity study using white Swiss mice
(Schroeder and Mitchener, 1975). Groups of mice (54/sex/group) were exposed until death to mercuric
chloride in drinking water at 5 ppm Hg (0.95 (mg/kg)/day). No effects on survival or body weights were
observed. After dying, mice were weighed and dissected. The animals were examined for gross tumors,
and some sections were made of the heart, lung, liver, kidney and spleen for microscopic examination.
No toxic effects of mercuric chloride were reported in the study. No statistically significant differences
were observed in tumor incidences for treated animals and controls. This study is of limited use for
evaluation of carcinogenicity because complete histological examinations were not performed, only a
single dose was tested, and the MTD was not achieved.
_II.A.4 SUPPORTING DATA FOR CARCINOGENICITY
The increasing trend for renal tubular cell tumors in mice observed in the NTP (1993) study
receives some support from similar findings in mice after chronic dietary exposure to methylmercury
(Hirano et al., 1986; Mitsumori et al., 1981, 1990). In these studies, dietary exposure to methylmercuric
chloride resulted in increases in renal tubular tumors at doses wherein substantial nephrotoxicity was
observed (see methylmercury file on IRIS).
As summarized in NTP (1993) and U.S. EPA (1985), mercuric chloride has produced some
positive results for clastogenicity in a variety of in vitro and in vivo genotoxicity assays; mixed results
regarding its mutagenic activity have been reported. Mercuric chloride was negative in gene mutation
tests with Salmonella typhimurium (NTP, 1993; Wong, 1988) but produced DNA damage as measured in
the Bacillus subtilis rec assay (Kanematsu et al., 1980). A
weakly positive response for gene mutations was observed in mouse lymphoma (L5178Y) cells in the
presence of microsomal activation (Oberly et al., 1982). DNA damage has also been observed in assays
using rat and mouse embryo fibroblasts (Zasukhina et al., 1983), CHO cells and human KB cells
(Cantoni and Costa, 1983; Cantoni etal., 1982, 1984a,b; Christie etal, 1984, 1986;
NTP, 1993; Williams et al., 1987). Mercuric chloride also produced chromosome aberrations and SCEs
in CHO cells (Howard et al., 1991) and chromosome aberrations in human lymphocytes (Morimoto et al.,
1982). Sex-linked recessive lethal mutations were not observed in male Drosophila melanogaster (NTP,
1993).
B-30
-------
Although mice given intraperitoneal doses of mercuric chloride have shown no increase in
chromosomal aberrations in bone marrow cells (Poma et al., 1981) and no increase in aneuploidy in
spermatogonia (Jagiello and Lin, 1973), mercuric chloride administered to mice by gavage induced a
dose-related increase in chromosome aberrations and aberrant cells in the bone marrow (Ghosh et al.,
1991). Similarly, an increased incidence of chromosomal aberrations (primarily deletion and numeric
aberrations) was observed in livers of fetal mice exposed to mercury in utero as the result of maternal
inhalation of aerosols of mercuric chloride (Selypes et al., 1984). Positive dominant lethal results
(increased resorptions and post-implantation deaths in
untreated females) have been obtained in studies in which male rats were administered mercuric chloride
orally (Zasukhina et al., 1983). A slight increase in post-implantation deaths and a decrease in living
embryos were also reported in treated female mice mated to untreated males (Suter, 1975); however, it
was not clear whether these effects were the result of germ cell
mutations or were secondary to maternal toxicity.
The effects of mercuric chloride on genetic material has been suggested to be due to the ability of
mercury to inhibit the formation of the mitotic spindle, an event known as c-mitosis (U.S. EPA, 1985).
_II.B QUANTITATIVE ESTIMATE OF CARCINOGENIC RISK FROM ORAL EXPOSURE
None. The incidences of squamous cell papillomas of the forestomach and thyroid follicular
cell carcinomas were evaluated. No slope factor was derived using the forestomach tumors because these
tumors are probably the result of doses of mercuric chloride above-MTD resulting in irritation of the
forestomach and subsequent cell death and epithelial proliferation. The carcinogenic mechanism for
mercuric chloride at the high doses observed may be specific to effects of irritation of the forestomach.
Regarding the thyroid carcinomas, a variety of drugs, chemicals and physiological perturbations
result in the development of thyroid follicular tumors in rodents. For a number of chemicals, the
mechanism of tumor development appears to be a secondary effect of long-standing hypersecretion of
thyroid-stimulating hormone by the pituitary (Capen and Martin, 1989; McClain, 1989). In the absence
of such long-term stimulatory effects, induction of thyroid follicular cell cancer by such chemicals
usually does not occur (Hill, 1989). The mechanism whereby thyroid tumors developed in the NTP
(1993) assay is very unclear given that hyperplasia was not observed. The study reviewers concluded
that it was difficult to associate the increase in thyroid tumors with mercuric chloride administration.
Thus, it would be of questionable value to use the thyroid tumors in rats as the basis for a quantitative
cancer risk estimate for humans.
All tumors in rats were observed at doses equalling or exceeding the MTD. Kidney tumors in
mice were observed in only the high-dose males. The increased incidence was not statistically significant
in comparison to the concurrent controls, but was significant when compared with historical controls. A
linear low-dose extrapolation based on the male mouse kidney tumor data (three tumors in the high-dose
group only) is not appropriate.
_II.C QUANTITATIVE ESTIMATE OF CARCINOGENIC RISK FROM INHALATION
EXPOSURE
None.
B-31
-------
JI.D EPA DOCUMENTATION, REVIEW AND CONTACTS (CARCINOGENICITY
ASSESSMENT)
_II.D.l EPA DOCUMENTATION
Source Document -- U.S. EPA, 1995
This IRIS summary is included in The Mercury Study Report to Congress which was reviewed
by OHEA and EPA's Mercury Work Group in November 1994. An interagency review by scientists from
other federal agencies took place in January 1995. The report was also reviewed by a panel of
non-federal external scientists in January 1995 who met in a public meeting on January 25-26. All
reviewers comments have been carefully evaluated and considered in the revision and fmalization of this
IRIS summary. A record of these comments is summarized in the IRIS documentation files.
_II.D.2 REVIEW (CARCINOGENICITY ASSESSMENT)
Agency Work Group Review ~ 03/03/94
Verification Date - 03/03/94
_II.D.3 U.S. EPA CONTACTS (CARCINOGENICITY ASSESSMENT)
Rita Schoeny / NCEA ~ (513)569-7544
REFERENCES
Cantoni, O. and M. Costa. 1983. Correlations of DNA strand breaks and their repair with cell survival
following acute exposure to mercury (II) and X-rays. Mol. Pharmacol. 24(1): 84-89.
Cantoni, O., R.M. Evans and M. Costa. 1982. Similarity in the acute cytotoxic response of mammalian
cells to mercury (II) and X-rays: DNA damage and glutathione depletion. Biochem. Biophys. Res.
Commun. 108(2): 614-619.
Cantoni, O., N.T. Christie, A. Swann, D.B. Drath and M. Costa. 1984a. Mechanism of HgC12
cytotoxicity in cultured mammalian cells. Mol. Pharmacol. 26: 360-368.
Cantoni, O., N.T. Christie, S.H. Robinson and M. Costa. 1984b. Characterization of DNA lesions
produced by HgC12 in cell culture systems. Chem. Biol. Interact. 49: 209-224.
Capen, C.C. and S.L. Martin. 1989. The effects of xenobiotics on the structure and function of thyroid
follicular and C-cells. Toxicol. Pathol. 17(2): 266-293.
B-32
-------
Christie, N.T., O. Cantoni, R.M. Evans, R.E. Meyn and M. Costa. 1984. Use of mammalian DNA
repair-deficient mutants to assess the effects of toxic metal compounds on DNA. Biochem. Pharmacol.
33(10): 1661-1670.
Christie, N.T., O. Cantoni, M. Sugiyama, F. Cattabeni and M. Costa. 1986. Differences in the effects of
Hg(II) on DNA repair induced in Chinese hamster ovary cells by ultraviolet or X-rays. Mol. Pharmacol.
29: 173-178.
Fitzhugh, O.G., A.A. Nelson, E.P. Lauge and P.M. Kunze. 1950. Chronic oral toxicities of
mercuric-phenyl and mercuric salts. Arch. Ind. Hyg. Occup. Med. 2: 433-442.
Ghosh, A.K., S. Sen, A. Sharma and G. Talukder. 1991. Effect of chlorophyllin on mercuric
chloride-induced clastogenicity in mice. Food. Chem. Toxicol. 29(11): 777-779.
Hill, R.N., L.S. Erdreich, O.V. Paynter, P.A. Roberts, S.L. Rosenthal and C.F. Wilkinson. 1989.
Review. Thyroid follicular cell carcinogenesis. Fund. Appl. Toxicol. 12: 629-697.
Hirano, M., K. Mitsumori, K. Maita and Y. Shiraso. 1986. Further carcinogenicity study on
methylmercury chloride in ICRmice. Jap. J. Vet. Sci. 48(1): 127-135.
Howard, W., B. Leonard, W. Moody and T.S. Kochhar. 1991. Induction of chromosome changes by
metal compounds in cultured CHO cells. Toxicol. Lett. 56(1-2): 179-186.
Jagiello, G. and J.S. Lin. 1973. An assessment of the effects of mercury on the meiosis of mouse ova.
Mutat. Res. 17: 93-99.
Kanematsu, N., M. Kara and T. Kada. 1980. Rec assay and mutagencity studies on metal compounds.
Mutat. Res. 77: 109-116.
McClain, R.M. 1989. The significance of hepatic microsomal enzyme induction and altered thyroid
function in rats: Implications for thyroid gland neoplasia. Toxicol. Pathol. 17(2): 294-306.
Mitsumori, K., K. Maita, T. Saito, S. Tsuda and Y. Shirasu. 1981. Carcinogenicity of methylmercury
chloride in ICR mice: Preliminary note on renal carcinogenesis. Cancer Lett. 12:305-310.
Mitsumori, K., M. Hirano, H. Ueda, K. Maita and Y. Shirasu. 1990. Chronic toxicity and
carcinogenicity of methylmercury chloride in B6C3F1 mice. Fund. Appl. Toxicol. 14: 179-190.
Morimoto, K., S. lijima and A. Koizumi. 1982. Selenite prevents the induction of sister-chromatid
exchanges by methyl mercury and mercuric chloride in human whole-blood cultures. Mutat. Res. 102:
183-192.
NTP (National Toxicology Program). 1993. NTP technical report on the toxicology and carcinogenesis
studies of mercuric chloride (CAS No. 7487-94-7) in F344 rats and B6C3F1 mice (gavage studies). NTP
TR 408. National Toxicology Program, U.S. Department of Health and Human Services, Public Health
Service, National Institutes of Health, Research Triangle Park, NC.
B-33
-------
Oberly, T.J., C.E. Piper and D.S. McDonald. 1982. Mutagenicity of metal salts in the L5178Y mouse
lymphoma assay. J. Toxicol. Environ. Health. 9: 367-376.
Poma, K., M. Kirsch-Volders and C. Susanne. 1981. Mutagenicity study of mice given mercuric
chloride. J. Appl. Toxicol. 1(6): 314-316.
Rice, G. and L. Knauf. 1994. Further Statistical Evaluation of the NTP Mercuric Chloride Mouse
Bioassay. Memorandum to the U.S. EPA CRAVE File for Mercuric Chloride, March 1.
Schroeder, H. and M. Mitchener. 1975. Life-time effects of mercury, methyl mercury, and nine other
trace metals in mice. J. Nutr. 105:452-458.
Selypes, A., L. Nagymajtenyi and G. Berencsi. 1984. Study of the mutagenic and teratogenic effect of
aerogenic mercury exposition in mouse. Collect. Med. Leg. Toxicol. Med. 125: 65-69.
Suter, K.E. 1975. Studies on the dominant-lethal and fertility effects of the heavy metal compounds
methylmercuric hydroxide, mercuric chloride and cadmium chloride in male and female mice. Mutat.
Res. 30: 365-374.
U.S. EPA. 1980. Ambient Water Quality Criteria Document for Mercury. Prepared by the Office of
Health and Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH
for the Office of Water Regulation and Standards, Washington, DC. EPA/440/5-80/058. NTIS PB
81-117699.
U.S. EPA. 1984a. Mercury Health Effects Update: Health Issue Assessment. Final Report. Prepared
by the Office of Health and Environmental Assessment, Environmental Criteria and Assessment Office,
Cincinnati, OH for the Office of Air Quality Planning and Standards, Research Triangle Park, NC.
EPA/600/8-84/019F. NTIS PB81-85-123925.
U.S. EPA. 1984b. Health Effects Assessment for Mercury. Prepared by the Office of Health and
Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH for the
Office of Emergency and Remedial Response, Washington, DC. EPA/540/1086/042. NTIS
PB86-134533/AS.
U.S. EPA. 1985. Drinking Water Criteria Document for Mercury. Prepared by the Office of Health and
Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH for the
Office of Drinking Water, Washington, DC. EPA/600/X-84/178. NTIS PB86-117827.
U.S. EPA. 1988. Drinking Water Criteria Document for Inorganic Mercury. Prepared by the Office of
Health and Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH
for the Office of Drinking Water, Washington, DC. EPA/600/X-84/178. NTIS PB89-192207.
U.S. EPA. 1993. Summary Review of Health Effects Associated with Mercuric Chloride: Health Issue
Assessment (Draft). Prepared by the Office of Health and Environmental Assessment, Environmental
Criteria and Assessment Office, Cincinnati, OH for the Office of Air Quality Planning and Standards,
Research Triangle Park, NC. EPA/600/R-92/199.
U.S. EPA. 1995. Mercury Study Report to Congress. Office of Research and Development,
Washington, DC. External Review Draft. EPA/600/P-94/002Ab.
B-34
-------
Williams, M.V., T. Winters and K.S. Waddel. 1987. In vivo effects of mercury (II) on deoxyuridine
triphosphate nucleotidohydrolase, DNA polymerase (alpha, beta) and uracil-DNA glycosylase activities
in cultured human cells: Relationship to DNA damage, DNA repair, and cytotoxicity. Mol. Pharmacol.
31:200-207.
Wong, P.K. 1988. Mutagenicity of heavy metals. Bull. Environ. Contam. Toxicol. 40(4): 597-603.
Zasukhina, G.D., I.M. Vasilyeva, N.I. Sdirkova, G.N. Krasovsky, L.Y. Vasyukovich, U.I. Kenesariev and
P.G. Butenko. 1983. Mutagenic effect of thallium and mercury salts on rodent cells with different repair
activities. Mutat. Res. 124: 163-173.
B-35
-------
_I.A. REFERENCE DOSE FOR CHRONIC ORAL EXPOSURE (RfD)
Chemical ~ Methylmercury (MeHg)
CASRN - 22967-92-6
Preparation Date - 2/10/95
_I.A.l ORAL RfD SUMMARY
Critical Effect Experimental Doses* UF MF RfD
Developmental Benchmark Dose: llppm 10 1 1E-4
neurologic in hair; equivalent to mg/kg-day
abnormalities maternal blood levels
in human infants 44 ug/L and body
burdens of 69 ug or
Human epidemiologic daily intake of 1.1
studies ug/kg-day
Marsh et al., 1987: Seafood Safety, 1991
* Conversion Factors and Assumptions ~ Maternal daily dietary intake levels were used as the dose
surrogate for the observed developmental effects in the infants. The daily dietary intake levels were
calculated from hair concentrations measured in the mothers. This conversion is explained in the text
below. A benchmark dose approach was used rather than a NOAEL/LOAEL approach to analyze the
neurological effects in infants as the response variable. This analysis is also explained in the text below.
_I.A.2 PRINCIPAL STUDIES (ORAL RfD)
Marsh, D.O., T.W. Clarkson, C. Cox, L. Amin-Zaki and S. Al-Trkiriti. 1987. Fetal methylmercury
poisoning: Relationship between concentration in a single strand of maternal hair and child effects.
Arch. Neural. 44: 1017-1022.
Seafood Safety. 1991. Committee on Evaluation of the Safety of Fishery Products, Chapter on
Methylmercury: FDA Risk Assessment and Current Regulations, National Academy Press, Washington,
DC. p. 196-221.
In 1971-1972 many citizens in rural Iraq were exposed to MeHg-treated seed grain that was
mistakenly used in home-baked bread. Latent toxicity was observed in many adults and children who
had consumed bread over a 2- to 3-month period. Infants born to mothers who ate contaminated bread
during gestation were the most sensitive group. Often infants exhibited neurologic abnormalities while
their mothers showed no signs of toxicity. Some information indicates that male infants are more
sensitive than females. Among the signs noted in the infants exposed during fetal development were
cerebral palsy, altered muscle tone and deep tendon reflexes as well as delayed developmental
milestones, i.e., walking by 18 months and talking by 24 months. The neurologic signs noted in adults
B-36
-------
included paresthesia, ataxia, reduced visual fields and hearing impairment. Some mothers experienced
paresthesia and other sensory disturbances but these symptoms were not necessarily correlated with
neurologic effects in their children. Unique analytic features of mercury (Hg), that is, analysis of
segments of hair correlated to specific time periods in the past permitted approximation of maternal
blood levels that the fetuses were exposed to in utero. The data collected by Marsh et al. (1987)
summarizes clinical neurologic signs of 81 mother and child pairs. From x-ray fluorescent spectrometric
analysis of selected regions of maternal scalp hair, concentrations ranging from 1 to 674 ppm were
determined and correlated with clinical signs observed in the affected members of the mother-child pairs.
Among the exposed population were affected and unaffected individuals throughout the dose-exposure
range.
While the purpose of the Seafood Safety publication was to critique the quantitative risk
assessment that FDA had performed for MeHg, this material is included in the EPA risk assessment
because the Tables of Incidence of various clinical effects in children that were provided in the FDA
assessment readily lend themselves to a benchmark dose approach. Specifically the continuous data for
the Iraqi population that was reported in Marsh et al. (1987) are placed in five dose groups and incidence
rates are provided for delayed onset of walking, delayed onset of talking, mental symptoms, seizures,
neurological scores above 3 and neurological scores above 4 for affected children. Neurologic scores
were determined by clinical evaluation for cranial nerve signs, speech, involuntary movement, limb tone
strength, deep tendon reflexes, plantar responses, coordination, dexterity, primitive reflexes, sensation,
posture, and ability to sit, stand and run. This paper provided groupings of the 81 mother-infant pairs for
various effects, and the authors present the data in Tables 6-11 through 6-16B.
EQUATION USED FOR CALCULATION OF DAILY DOSE: From the concentration of Hg present in
maternal hair, a corresponding blood concentration value is determined. A hair concentration of 11 ppm
converts to a blood concentration of 44 ug/L; the following equation can then be used to determine the
daily dose that corresponds to that blood concentration of Hg. Use of this equation is based on the
assumption that steady-state conditions exist and that first-
order kinetics for Hg are being followed.
d = (C x b x V)/(A x f)
d (ug/day) = 44 ug/L multiplied by 0.014 multiplied by 5 liters divided by 0.95 then divided by
0.05 yields 65 ug/day
where:
d = daily dietary intake (expressed as ug of MeHg)
C = concentration in blood (expressed as ug/L)
b = elimination constant (expressed as days-1)
V = volume of blood in the body (expressed as liters)
A = absorption factor (unitless)
f = fraction of daily intake taken up by blood (unitless)
B-37
-------
The following sections provide the data and rationale supporting the choice of parameter values
used in the conversion equation. It should be noted that even if the upper or lower ranges of the
parameter values were used, the conversion factor precision remains the same due to rounding error. The
Agency realizes that new pharmacokinetic data may become available that warrant a change to some of
these parameters.
HAIR TO BLOOD CONCENTRATION RATIO: The hairblood concentration ratio for total Hg is
frequently cited as 250. The following description provides a justification of why we have chosen to use
the ratio of 250:1. Ratios reported in the literature range from 140 to 370, a difference of more than a
factor of 2.5. Differences in the location of hair sampled (head vs chest and distance from scalp) may
contribute to the differences observed. As much as a 3-fold seasonal variation in Hg levels was observed
in average hair levels for a group of individuals with moderate to high fish consumption rates, with
yearly highs occurring in the fall and early winter (Phelps et al., 1980; Suzuki et al., 1993). The high
slope reported by Tsubaki and Irukayama (1977) may have reflected the fact that Hg levels were
declining at the time of sampling so that the hair levels reflect earlier, higher blood levels. Phelps et al.
(1980) obtained multiple blood samples and sequentially analyzed lengths of hair from individuals. Both
hair and blood samples were taken for 339 individuals in Northwestern Ontario. After reviewing the
various reports for converting hair concentrations to blood concentrations, the Phelps paper was selected
because of the large sample size and the attention to sampling and analysis. The ratio Phelps observed
between the total Hg concentration in hair taken close to the scalp and simultaneous blood sampling for
this group was 296. To estimate the actual ratio, the authors assumed that blood and hair samples were
taken following complete cessation of MeHg intake. They also assumed a half-life of MeHg in blood of
52 days and a lag of 4 weeks for appearance of the relevant level in hair at the scalp. Phelps also
determined that 94% of the Hg in hair was MeHg. Based on these assumptions, they calculated that if
the actual hairblood ratio was 200, they would have observed a ratio of 290. Based on these and other
considerations, Phelps states that the actual ratio is "probably higher than 200, but less than the observed
value of 296." As the authors point out, 2/3 of the study population were sampled during the falling
phase of the seasonal variation (and 1/3 or less in the rising phase). This methodology would tend to
result in a lower observed ratio; therefore, the actual average is likely to be greater than 200.
In view of these limitations a value of 250 was considered acceptable for the purpose of
estimating average blood levels in the Iraqi population.
CALCULATION OF DIETARY INTAKE FROM BLOOD CONCENTRATION: The first step in this
process is to determine the fraction of Hg in diet that is absorbed. Radio-labeled methyl-mercuric nitrate
(MeHgNO3) was administered in water to three healthy volunteers (Aberg et al., 1969). The uptake was
>95%. Miettinen et al. (1971) incubated fish liver homogenate with radio-labeled MeHgNO3 to produce
methylmercury proteinate. The proteinate was then fed to fish that were killed after a week and then
cooked and fed to volunteers after confirmation of MeHg in the fish. Mean uptake exceeded 94%.
Based on these experimental results, this derivation used an absorption factor of 0.95.
The next step involves determining the fraction of the absorbed dose that is found in the blood.
There are three reports on the fraction of absorbed MeHg dose distributed to blood volume in humans.
Kershaw et al. (1980) report an average fraction of 0.059 of absorbed dose in total blood volume, based
on a study of five adult male subjects who ingested MeHg-contaminated tuna. In a group of nine male
and six female volunteers who had received 203 Hg-methylmercury in fish approximately 10% of the
total body burden was present in 1 liter of blood in the first few days after exposure, dropping to
approximately 5% over the first 100 days (Miettinen et al., 1971). In another study, an average value of
B-38
-------
1.14% for the percentage of absorbed dose in 1 kg of blood was derived from subjects who consumed a
known amount of MeHg in fish over a 3-month period (Sherlock et al., 1982). Average daily intake in
the study ranged from 43-233 ug/day and a dose-related effect on percentage of absorbed dose was
reported that ranged from 1.03-1.26% in 1 liter of blood (each of these values should be multiplied by 5
[since there are approximately 5 liters of blood in an adult human body] to yield the total amount in the
blood compartment). The value 0.05 has been used for this parameter in the past (WHO, 1990).
ELIMINATION CONSTANT: Based on data taken from four studies, reported clearance half-times
from blood or hair ranged from 48-65 days. Two of these studies included the Iraqi population exposed
during the 1971-1972 outbreak. The value from the Cox study (Cox et al., 1989) is derived from the
study group that included the mothers of the infants upon which this risk assessment is based. The
average elimination constant of the four studies is 0.014; the average of individual values reported for 20
volunteers ingesting from 42-233 ug Hg/day in fish for 3 months (Sherlock et al., 1982) is also 0.014.
VOLUME OF BLOOD IN THE BODY AND BODY WEIGHT: Blood volume is 7% of body weight as
has been determined by various experimental methods and there is an increase of 20 to 30% (to about 8.5
to 9%) during pregnancy (Best and Taylor, 1961). Specific data for the body weight of Iraqi women
were not found. Assuming an average body weight of 60 kg. (Snyder et al., 1981) and a blood volume of
9% of body weight during pregnancy, a blood volume of 5.4 liters is derived.
DERIVATION OF A BENCHMARK DOSE: Benchmark dose estimates were made for excess risk
above background based on a combination of all childhood neurologic end points. This method was
chosen since the Agency felt that any childhood neurologic abnormality is considered an adverse effect
and likely to have serious sequelae lasting throughout lifetime. In addition, grouping of all neurologic
endpoints provided a much better goodness of fit of the data than when any endpoint was used
individually. The endpoints that were grouped delayed the onset of walking and talking, neurologic
scores <3, mental symptoms, and seizures. Using these data sets taken from the Seafood Safety paper,
benchmark doses at the 1, 5 and 10% incidence levels were constructed using both Weibull and
polynomial models. The Weibull model places the maximum likelihood estimate with corresponding
95% confidence level at 11 ppm of MeHg in maternal hair. The Agency decided to use the lower 95%
confidence level for the 10% incidence rate. Recent research by Faustman et al. (1994) and Allen et al.
(1994a,b) suggests that the 10% level for the benchmark dose roughly correlates with a NOAEL for
quantal developmental toxicity data. The 95% lower confidence limits on doses corresponding to the 1,
5, and 10% levels were calculated using both models and the values determined using the polynomial
model always fell within 3% of the Weibull values. For final quantitative analysis the Weibull model
was chosen because of goodness of fit of the data and because this model has been used in the past by the
Agency for developmental effects. The experience of the Agency indicates that this model performs well
when modeling for developmental effects.
_I.A.3 UNCERTAINTY AND MODIFYING FACTORS (ORAL RfD)
UF ~ An uncertainty factor of 3 is applied for variability in the human population, in particular the
variation in the biological half-life of MeHg and the variation that occurs in the hairblood ratio for Hg.
In addition, a factor of 3 is applied for lack of a two-generation reproductive study and lack of data for
the effect of exposure duration on sequelae of the developmental neurotoxicity effects and on adult
paresthesia. The total UF is 10.
B-39
-------
MF ~ None
_I.A.4 ADDITIONAL STUDIES/COMMENTS (ORAL RfD)
McKeown-Eyssen et al. (1983) have provided a report of neurologic abnormalities in four
communities of Cree Indians in northern Quebec. A group of 241 children first exhibited clinical signs
consistent with MeHg exposure between 12 and 30 months of age. An attempt was made to account for
possible confounding factors; the interviewers determined alcohol and tobacco consumption patterns
among the mothers of affected children. Age of the mothers and multiparity was also taken into account
in analysis of the data. The average indices of exposure were the same for boys and girls at 6 ug/g; only
6% had exposure above 20 ug/g. The prevalence of multiple abnormal neurologic findings was about 4%
for children of both sexes. The most frequently observed abnormality was delayed deep tendon reflexes;
this was seen in 11.4% of the boys and 12.2% of the girls. These investigators found that when there was
a positive association between maternal Hg exposure and abnormal neurologic signs in boys, the
incidence rate was 7.2%. The incidence rate for neurologic disorders in daughters was less and was
found to be not statistically significant. Disorders of muscle tone were usually confined to the legs.
Persistence of the Babinski reflex and incoordination due to delayed motor development were seen with
equal frequency for both sexes. The discriminant analysis conducted for the boys to distinguish the 15
cases with abnormal muscle tone or reflexes from the 82 normal controls was unable to separate
differences between these groups based on confounding variables. The prevalence of abnormality of
muscle tone or reflexes was found to increase 7 times with each increase of 10 ug/g of the prenatal
exposure index. Although this study provides supportive data for the RfD, it is not included with the
principal studies because it was confounded by alcoholism and smoking among
mothers.
Studies performed in New Zealand investigated the mental development of children who had
prenatal exposure to MeHg (Kjellstrom et al., 1986, 1989). A group of 11,000 mothers who regularly ate
fish were initially screened by survey and of these about 1000 had consumed fish in three meals per week
during pregnancy. Working from this large population base, 31 matched pairs were established. For
proper comparison a reference child matched for ethnic group and age of mother, child's birthplace and
birth date was identified for each high Hg child. Retrospective Hg concentrations were determined from
the scalp hair of the mothers to match the period of gestation. The average hair concentration for
high-exposure mothers was 8.8 mg/kg and for the reference group it was 1.9 mg/kg.
The children of exposed mothers were tested at 4 and 6 years of age. At 4 years of age the
children were tested using the Denver Developmental Screen Test (DDST) to assess the effects of Hg.
This is a standardized test of a child's mental development that can be administered in the child's home.
It consists of four major function sectors: gross motor, fine motor, language, and personal-social. A
developmental delay in an individual item is scored as
abnormal, questionable when the child has failed in their response and at least 90% of the children can
pass this item at a younger age. The results of the DDST demonstrated 2 abnormal scores and 14
questionable scores in the high Hg-exposed group and 1 abnormal and 4 questionable scores in the
control group. Analysis of the DDST results by sector showed that developmental delays were most
commonly noted in the fine motor and language sectors but the
differences for the experimental and control groups were not significant. The investigators noted that
differences in performance of the DDST between high Hg-exposed and referent children could be due to
B-40
-------
confounding variables and that DDST results are highly dependent upon the age of the child.
Standardized vision tests and sensory tests were also performed to measure
development of these components of the nervous system. The prevalence for developmental delay in
children was 52% from high Hg mothers and 17% from mothers of the reference group. In comparison to
other studied populations, the hair Hg concentration of the mothers in this study were lower than those
associated with CNS effects in children exposed in Japan and Iraq. Results of the DDST demonstrated 2
abnormal scores and 14 questionable scores in the high
Hg-exposed group and 1 abnormal and 4 questionable scores in the control group. Analysis of the DDST
results by sector showed that developmental delays were most commonly noted in the fine motor and
language sectors but the differences for the experimental and control groups were not significant. The
data obtained from this study is too limited for detailed dose-response analysis. The differences in
performance of the DDST between high Hg-exposed and referent children could be due to confounding
variables. DDST results are highly dependent upon the age of the child. Infants of the Hg-exposed
group more frequently had low birth weights and were more likely to be born prematurely. Use of this
study is also limited by the fact that there was only a 44% participation rate.
A second stage follow-up of the original Kjellstrom study was carried out when the children
were 6 years old to confirm or refute the developmental findings observed at age 4 (Kjellstrom et al.,
1989). In this later study the high exposure children were compared with three control groups with lower
prenatal Hg exposure. The mothers of children in two of these control groups had high fish consumption
and average hair Hg concentrations during pregnancy of 3-6 mg/kg and 0-3 mg/kg, respectively. For this
study the high exposure group was matched for maternal ethnic group, age, smoking habits, residence,
and sex of the child. For this second study, 61 of 74 high-exposure children were available for study.
Each child was tested at age 6 with an array of scholastic, psychological, and behavioral tests which
included the Test of Language Development (TOLD), the Wechsler Intelligence Scale for Children, and
the McCarthy Scale of Children's Abilities. The results of the tests were compared between groups.
Confounding was controlled for by using linear multiple regression analysis. A principal finding was
that normal results of the psychological test variables were influenced by ethnic background and
social class. The high prenatal MeHg exposure did decrease performance in the tests, but it contributed
only a small part of the variation in test results. The investigation found that an average hair Hg level of
13-15 mg/kg during pregnancy was consistently associated with decreased test performance. Due to the
small size of the actual study groups it was not possible to determine if even lower exposure levels might
have had a significant effect on test
results. The Kjellstrom studies are limited for assessing MeHg toxicity because the developmental and
intelligence tests used are not the most appropriate tests for defining the effects of MeHg. Also, greater
significance was seen in differences of cultural origins of the children than the differences in maternal
hair MeHg concentrations.
The initial epidemiologic report of MeHg poisoning involved 628 human cases that occurred in
Minamata Japan between 1953 and 1960 (Tsubaki and Irukayama, 1977). The overall prevalence rate for
the Minamata region for neurologic and mental disorders was 59%. Among this group 78 deaths
occurred and hair concentrations of Hg ranged from 50-700 ug/g. Hair Hg concentrations were
determined through the use of less precise analytic methods than were available for later studies. The
specific values derived from these studies do not contribute directly to quantitative risk assessment for
MeHg. The most common clinical signs observed in adults were paresthesia, ataxia, sensory
disturbances, tremors, impairment of hearing and difficulty in walking. This particular group of
neurologic signs has become known as "Minimata disease." Examination of the brains of severely
affected patients that died revealed marked atrophy of the brain (55% normal volume and weight) with
B-41
-------
cystic cavities and spongy foci. Microscopically, entire regions were devoid of neurons, granular cells in
the cerebellum, golgi cells and Purkinje cells. Extensive investigations of congenital Minamata disease
were undertaken and 20 cases that occurred over a 4-year period were documented. In all instances the
congenital cases showed a higher incidence of symptoms than did their mothers. Severe disturbances of
nervous function were described and the affected offspring were very late in reaching developmental
milestones. Hair concentrations of Hg in affected infants ranged from 10 to 100 ug/g. Data on hair Hg
levels for the mothers during gestation were not available.
Rice (1989) dosed five cynomolgus monkeys (Macaca fascicularis) from birth to 7 years of age
with 50 ug/kg-day and performed clinical and neurologic examinations during the dosing period and for
an additional 6 years. As an indicator of the latent effects of MeHg, objective neurologic examinations
performed at the end of the observation period revealed insensitivity to touch and loss of tactile response.
In addition, monkeys dosed with MeHg were clumsier and slower to react when initially placed in an
exercise cage as well as in the later stages of the observation period.
Gunderson et al. (1986) administered daily doses of 50-70 ug/kg of MeHg to 11 crab-eating
macaques (Macaca fascicularis) throughout pregnancy which resulted in maternal blood levels of
1080-1330 ug/L in mothers and 1410-1840 ug/L in the offspring. When tested 35 days after birth the
infants exhibited visual recognition deficits.
In another study, groups of 7 or 8 female crab-eating macaques (Macaca fascicularis) were dosed
with 0.50 and 90 ug/kg-day of MeHg through four menstrual cycles (Burbacher et al., 1984). They were
mated with untreated males and clinical observations were made for an additional 4 months. Two of
seven high-dose females aborted and three did not conceive during the 4-month mating period; the other
two females delivered live infants. Two of seven females of the 50 ug/kg-day dose group aborted; the
remaining females delivered live infants. All 8 females of the control group conceived and 6 delivered
live infants. These reproductive results approached but did not reach statistical significance.
Reproductive failure within dose groups could be predicted by blood Hg levels. The dams did not show
clinical signs of MeHg poisoning during the breeding period or gestation but when females were dosed
with 90 ug/kg-day for 1 year 4/7 did show adverse neurologic signs.
Bornhausen et al. (1980) reported a decrease in operant behavior performance in 4-month-old
rats whose dams had received 0.005 and 0.05 mg/kg-day of MeHg on days 6 through 9 of gestation. A
statistically significant effect (<0.05) was observed in offspring whose dams had received 0.01 and 0.05
mg/kg during gestation. The authors postulated that more severe effects of in utero exposure would be
seen in humans since the biological half-time of Hg in the brain of humans is 5 times longer than the rat.
In addition, much longer in utero exposure to Hg would occur in humans since gestation is much longer
in chronologic time.
In another investigation groups of Wistar rats (50/sex/dose) were administered daily doses of 2,
10, 50 and 250 ug/kg-day of MeHg for 26 months (Munro et al., 1980). Female rats that received 25
ug/kg-day had reduced body weight gains and showed only minimal clinical signs of neurotoxicity;
however, male rats that received this dose did show overt clinical signs of neurotoxicity, had decreased
hemoglobin and hematocrit values, had reduced weight gains, and showed increased mortality.
Histopathologic examination of rats of both sexes receiving 25 ug/kg-day revealed demyelination of
dorsal nerve roots and peripheral nerves. Males showed severe kidney damage and females had minimal
renal damage. This study showed aNOAEL of 5 ug/kg-day and a LOAEL of 25 ug/kg-day.
B-42
-------
A 2-year feeding study of MeHg chloride was conducted in B6C3F1 mice (60 mice/sex/group) at
doses of 0, 0.4, 2 and 10 ppm (0, 0.04, 0.17, and 0.83 mg/kg-day) to determine chronic toxicity and
possible carcinogenic effects (Mitsumori et al., 1990). The mice were examined clinically during the
study and neurotoxic signs characterized by posterior paralysis were observed in 33 males after 59 weeks
and 3 females after 80 weeks in the 10 ppm group. A marked increase in mortality and a significant
decrease in body weight gain were also observed in the 10 ppm male dose group, beginning at 60 weeks.
Post mortem examination revealed toxic encephalopathy consisting of neuronal necrosis of the brain and
toxic peripheral sensory neuropathy in both sexes of the 10 ppm group. An increased incidence of
chronic nephropathy was observed in the 2 and 10 ppm males. Based upon this study a NOAEL of 0.04
mg/kg-day and a LOAEL of 0.17 mg/kg-day was determined. These results indicated that B6C3F1 mice
are more sensitive to the neurotoxic effects of MeHg than ICR mice.
KINETICS: MeHg in the diet is almost completely absorbed into the bloodstream. Animal studies
indicate (Walsh, 1982) that age has no effect on the efficiency of the gastrointestinal absorption, which is
usually in excess of 90%. From the bloodstream MeHg is distributed to all tissues, and distribution is
complete within 4 days in humans. The time necessary to reach peak brain levels from a single oral dose
is 1 or 2 days longer than other tissues and at this time the brain contains 6% of the total dose. Also at
this time the brain concentration is six times that of the blood.
Methylmercury is converted to inorganic Hg in various tissues at different rates in mammals.
The fraction of total Hg present as Hg++ depends on the duration of exposure and the time after
cessation of exposure. The percentages of total Hg present as inorganic Hg++ in tissues of the Iraqi
population exposed for 2 months were: whole blood 7%, plasma 22%, breast milk 39% and urine 73%.
Measurements in the hepatic tissue of patients that had died was 16-40% of Hg++.
The fecal pathway accounts for 90% of the total elimination of Hg in mammals after exposure to
MeHg. Essentially all Hg in feces is in the inorganic form. The process of fecal elimination begins with
biliary excretion with extensive recycling of both MeHg and Hg++ complexed with glutathione.
Inorganic Hg is poorly absorbed across the intestinal wall, but MeHg is readily reabsorbed such that a
secretion-resorption cycle is established. The intestinal microflora convert MeHg to inorganic Hg.
Whole body half-times determined in human volunteers averaged 70 days with a range of 52-93
days. Observations of blood half-times is 50 days with a range of 39-70 days. Lactating women have a
significantly shorter whole body half-time of 42 days compared with 79 days in nonlactating women.
Selenium is known to bioconcentrate in fish and it is thought that simultaneous ingestion of
selenium may offer a protective effect for the toxicity of MeHg based upon its antioxidant properties.
Selenium has been observed to correlate with Hg levels in blood (Granjean and Weihe, 1992).
B-43
-------
_I.A.5 CONFIDENCE IN THE ORAL RfD
Study ~ Medium
Data Base ~ Medium
RfD - Medium
The benchmark dose approach allowed use of the entire dose-response assessment and the
calculation of a value that was consistent with the traditional NOAEL/LOAEL approach. In addition, the
results of laboratory studies with nonhuman primates support the quantitative estimate of the
NOAEL/LOAEL range of the benchmark dose that was indicated by the human studies. The reported
literature covers detailed studies of human exposures with quantitation of MeHg by analysis of
specimens from affected mother-fetus pairs. A strength of the Marsh study is the fact that the
quantitative data came directly from the affected population and quantitation is based on biological
specimens obtained from affected individuals. Unfortunately, a threshold was not easily defined and
extended application of modeling techniques were needed to define the lower end of the dose-response
curve. This may indicate high variability of response to MeHg in the human mother-fetal pairs or
misclassification in assigning pairs to the cohort. Recent concerns expressed in the research community
relate to the applicability of a dose-response estimate based on a grain-consuming population when the
actual application is likely to help characterize risk for fish-consuming segments of the population.
Confidence in the supporting data base is medium. Confidence in the RfD is medium.
_I.A.6 EPA DOCUMENTATION
Source Document - U.S. EPA, 1995
This IRIS summary is included in The Mercury Study Report to Congress, which was reviewed
by OHEA and EPA's Mercury Work Group in November 1994. An interagency review by scientists from
other federal agencies took place in January 1995. The report was also reviewed by a panel of
non-federal external scientists in January 1995 who met in a public meeting on January 25-26. All
reviewers comments have been carefully evaluated and considered in the revision and finalization of this
IRIS summary. A record of these comments is summarized in the IRIS documentation files.
Other EPA Documentation - U.S. EPA, 1980, 1984, 1987, 1988
Agency Work Group Review ~ 12/02/85, 03/25/92, 02/17/94, 08/04/94, 09/08/94,
09/22/94, 10/13/94, 11/23/94
Verification Date - 11/23/94
_I.A.7 EPA CONTACTS (ORAL RfD)
Rita Schoeny / OHEA - (513)569-7544
Bruce Mintz / OST - (202)260-9569
B-44
-------
REFERENCES
Aberg, B., L. Ekman, R. Falk, U. Greitz, G. Persson and J. Snihs. 1969. Metabolism of methyl mercury
(203Hg) compounds in man. Arch. Environ. Health. 19: 478-484.
Allen B.C., R.J. Kavlock, C.A. Kimmel and E.M. Faustman. 1994a. Dose response assessment for
developmental toxicity. II. Comparison of generic benchmark dose estimates with NOAELS. Fund.
Appl. Toxicol. 23: 487-495.
Allen, B.C., R.J. Kavlock, C.A. Kimmel and E.M. Faustman. 1994b. Dose response assessment for
developmental toxicity. III. Statistical models. Fund. Appl. Toxicol. 23:496-509.
Best, C.H. and N.B. Taylor. 1961. A Text in Applied Physiology, 7th ed. The Williams and Wilkins
Co., Baltimore, MD. p. 19, 29.
Bornhausen, M., H.R. Musch and H. Greim. 1980. Operant behavior performance changes in rats after
prenatal methylmercury exposure. Toxicol. Appl. Pharmacol. 56: 305-310.
Burbacher, T.M., C. Monnett, L.S. Grant and N.K. Mottet. 1984. Methylmercury exposure and
reproductive dysfunction in the nonhuman primate. Toxicol. Appl. Pharmacol. 75: 18-24.
Cox, C., T.W. Clarkson, D.O. Marsh and G.G. Myers. 1989. Dose-response analysis of infants
prenatally exposed to methylmercury: An application of a single compartment model to single-strand hair
analysis. Environ. Res. 49:318-332.
Faustman, E.M., B.A. Allen, R.J. Kavlock and C.A. Kimmel. 1994. Dose-response assessment for
developmental toxicity. 1. Characterization of database and determination of no observed adverse effect
level. Fund. Appl. Toxicol. 23:478-486.
Grandjean, P. and P. Weihe. 1993. Neurobehavioral effects of intrauterine mercury exposure: Potential
sources of bias. Environ. Res. 61:176-183.
Gunderson, V.M., K.S. Grant, J.F. Pagan and N.K. Mottet. 1986. The effect of low-level prenatal
methylmercury exposure on visual recognition memory of infant crab-eating macaques. Child Dev. 57:
1076-1083.
Kershaw, T.G., T.W. Clarkson and P.H. Dhahir. 1980. The relationship between blood levels and the
dose of methylmercury in man. Arch. Environ. Health. 35(1): 28-36.
Kjellstrom, T., P. Kennedy, S. Wallis and C. Mantell. 1986. Physical and mental development of
children with prenatal exposure to mercury from fish. Stage 1: Preliminary test at age 4. National
Swedish Environmental Protection Board, Report 3080 (Solna, Sweden).
Kjellstrom, T., P. Kennedy, S. Wallis amd C. Mantell. 1989. Physical and mental development of
children with prenatal exposure to mercury from fish. Stage 2: Interviews and psychological tests at age
6. National Swedish Environmental Protection Board, Report 3642 (Solna, Sweden).
B-45
-------
Marsh, D.O., T.W. Clarkson, C. Cox et al. 1987. Fetal methylmercury poisoning: Relationship between
concentration in single strands of maternal hair and child effects. Arch. Neurol. 44: 1017-1022.
McKeown-Eyssen, G.E., J. Ruedy and A. Neims. 1983. Methylmercury exposure in northern Quebec.
II: Neurologic finds in children. Am. J. Epidemiol. 118(4): 470-479.
Miettinen, J.K., T. Rahola, T. Hattula, K. Rissanen and M. Tillander. 1971. Elimination of 203-Hg
methylmercury in man. Ann. Clin. Res. 3:116-122.
Mitsumori, K., M. Hiarano, H. Ueda, K. Maiata and Y. Shirasu. 1990. Chronic toxicity and
carcinogenicity of methylmercury chloride in B6C3F1 mice. Fund. Appl. Toxicol. 14: 179-190.
Munro, I.C., E.A. Nera, S.M. Charbonneau, B. Junkins and Z. Zawidzka. 1980. Chronic toxicity of
methylmercury in the rat. J. Environ. Pathol. Toxicol. 3(5-6): 437-447.
Phelps, R.W., T.W. Clarkson, T.G. Kershaw and B. Wheatley. 1980. Interrelationships of blood and
hair mercury concentrations in a North American population exposed to methylmercury. Arch. Environ.
Health. 35: 161-168.
Rice, B.C. 1989. Delayed neurotoxicity in monkeys exposed developmentally to methylmercury.
Neurotox. 10: 645-650.
Marsh, D.O., T.W. Clarkson, C. Cox et al. 1987. Fetal methylmercury poisoning: Relationship between
concentration in single strands of maternal hair and child effects. Arch. Neurol. 44: 1017-1022.
Seafood Safety. 1991. Committee on Evaluation of the Safety of Fishery Products, Chapter on
Methylmercury: FDA Risk Assessment and Current Regulations, National Academy Press, Washington,
DC. p. 196-221.
Sherlock, J.C., D.G. Lindsay, J. Hislop, W.H. Evans and T.R. Collier. 1982. Duplication diet study on
mercury intake by fish consumers in the United Kingdom. Arch. Environ. Health. 37(5): 271-278.
Snyder, W.S., M.T. Cook, L.R. Karhausen et al. 1975. International Commission of Radiological
Protection. No. 23: Report of a Task Group on Reference Man. Pergamon Press, NY.
Suzuki, T., T. Kongo, J. Yoshinaga et al. 1993. The hair-organ relationship in mercury concentration in
contemporary Japanese. Arch. Environ. Health. 48:221-229.
Tsubaki, T.K. and K. Irukayama. 1977. Minamata Disease: Methylmercury Poisoning in Minamata and
Niigata, Japan. Elsevier Science Publishers, New York. p. 143-253.
U.S. EPA. 1980. Ambient Water Quality Criteria Document for Mercury. Prepared by the Office of
Health and Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH
for the Office of Water Regulation and Standards, Washington, DC. EPA/440/5-80/058. NTIS PB
81-117699.
U.S. EPA. 1984. Mercury Health Effects Update: Health Issue Assessment. Final Report. Prepared by
the Office of Health and Environmental Assessment, Environmental Criteria and Assessment Office,
B-46
-------
Cincinnati, OH for the Office of Air Quality Planning and Standards, Research Triangle Park, NC.
EPA/600/8-84/019F. NTIS PB81-85-123925.
U.S. EPA. 1987. Peer Review Workshop on Mercury Issues. Environmental Criteria and Assessment
Office, Cincinnati, OH. Summary report. October 26-27'.
U.S. EPA. 1988. Drinking Water Criteria Document for Inorganic Mercury. Prepared by the Office of
Health and Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH
for the Office of Drinking Water, Washington, DC. EPA/600/X-84/178. NTIS PB89-192207.
U.S. EPA. 1995. Mercury Study Report to Congress. Office of Research and Development,
Washington, DC. External Review Draft. EPA/600/P-94/002Ab.
Walsh, C.T. 1982. The influence of age on the gastrointestinal absorption of mercuric chloride and
methylmercury chloride in the rat. Environ. Res. 27:412-420.
WHO (World Health Organization). 1990. Environmental Health Criteria 101: Methylmercury.
Geneva.
B-47
-------
_II. CARCINOGENICITY ASSESSMENT FOR LIFETIME EXPOSURE
Substance Name ~ Methylmercury
CASRN - 22967-92-6
Preparation Date ~ 5/24/94
_II.A EVIDENCE FOR CLASSIFICATION AS TO HUMAN CARCINOGENICITY
_II.A.l WEIGHT-OF-EVIDENCE CLASSIFICATION
Classification ~ C; possible human carcinogen
Basis ~ Based on inadequate data in humans and limited evidence of carcinogenicity in animals. Male
ICR and B6C3F1 mice exposed to methylmercuric chloride in the diet had an increased incidence of
renal adenomas, adenocarcinomas and carcinomas. The tumors were observed at a single site and in a
single species and single sex. The renal epithelial cell hyperplasia and tumors were observed only in the
presence of profound nephrotoxicity and were suggested to be a consequence of reparative changes in the
cells. Several nonpositive cancer bioassays were also reported. Although genotoxicity test data suggest
that methylmercury is capable of producing chromosomal and nuclear damage, there are also nonpositive
genotoxicity data.
_II.A.2 HUMAN CARCINOGENICITY DATA
Inadequate. Three studies were identified that examined the relationship between methylmercury
exposure and cancer. No persuasive evidence of increased carcinogenicity attributable to methylmercury
exposure was observed in any of the studies. Interpretation of these studies, however, was limited by
poor study design and incomplete descriptions of methodology and/or results.
Tamashiro et al. (1984) evaluated the causes of death in 334 subjects from the Kumamoto
Prefecture who had been diagnosed with Minamata disease (methylmercury poisoning) and died between
1970 and 1980. The subjects involved fishermen and their families who had been diagnosed with the
disease; thus, Minamata disease was used as a surrogate for methylmercury exposure. The controls were
selected from all deaths that had occurred in the same city or town as the cases and were matched on the
basis of sex, age at death (within 3 years) and year of death; two controls were matched to each subject.
Malignant neoplasms were designated as the underlying cause of death in 14.7% (49/334) of the subjects
and 20.1% (134/668) of the controls. For 47 subjects in which Minamata disease was listed as the
underlying cause of death, the investigators reanalyzed the mortality data and selected one of the
secondary causes to be the underlying cause of death in order to allow examination of the subjects and
controls under similar conditions and parameters. The three subjects for which Minamata disease was
listed as the only cause of death were excluded from further analysis. Using the Mantel-Haenzel method
to estimate odds ratios, no significant differences were observed between the subjects and controls with
respect to the proportion of deaths due to malignant neoplasms among males, females or both sexes
combined. The estimated odds ratios and 95% confidence intervals were 0.84 (0.49-1.43), 0.58
(0.28-1.21) and 0.75 (0.50-1.11) for males, females and both sexes combined. Similarly, no increases in
odds ratio were observed among the subjects relative to the controls when malignant neoplasms were
B-48
-------
identified as a secondary cause of death or were listed on death certificates as one of the multiple causes
of death. These data suggest that cancer incidence was not increased in persons with overt signs of
methylmercury poisoning when compared with persons for whom no diagnosis of methylmercury
poisoning had been made. Interpretation is limited by potential bias in designating the cause of death
among patients with known Minamata disease and by the uncertainty regarding the extent of
methylmercury exposure and undiagnosed Minamata disease among the controls. In a subsequent study,
Tamashiro et al. (1986) compared the mortality patterns (between 1970 and 1981) among residents of the
Fukuro and Tsukinoura districts in the Kumamoto Prefecture (inhabited mainly by fishermen and their
families) with that of age-matched residents of Minamata City (also in the Kumamoto Prefecture) who
died between 1972 and 1978. In this study, high exposure to methylmercury was inferred from residence
in a district believed to have higher intake of local seafood. By contrast, in the 1984 study described
above, high methylmercury exposure was inferred from a diagnosis of Minamata disease. A total of 416
deaths were recorded in the Fukuro and Tsukinoura districts in 1970-1981, and 2325 deaths were
recorded in Minamata City in 1972-1978. No statistically significant increase in the overall cancer
mortality rate was observed; however, an increase in the mortality rate due to liver cancer was observed
(SMR, 207.3; 95% CI, 116.0-341.9). Analysis of mortality by sex showed a statistically significant
increase in the rate of liver cancer only among males (SMR, 250.5; 95% CI, 133.4-428.4). Males also
had statistically significant higher mortality due to chronic liver disease
and cirrhosis. The authors note that these results should be interpreted with caution because the
population of the Fukuro and Tsukinoura districts had higher alcohol consumption and a higher
prevalence of hepatitis B (a predisposing factor for hepatocellular cancer). Interpretation of these results
is also limited by an incomplete description of the methodology used to calculate the SMRs; it is unclear
whether the study authors used appropriate methods to compare mortality data collected over disparate
time frames (12 years for exposed and 7 years for controls).
In a study from Poland, Janicki et al. (1987) reported a statistically significant increase in
mercury content of hair in leukemia patients (0.92 +/-1.44 ppm [sic]; n=47) relative to that in healthy
unrelated patients (0.49 +/-0.41 ppm; n=79). Similarly, the mercury content in the hair of a subgroup of
leukemia patients (0.69 +/- 0.75; n=19) was significantly greater than that in healthy relatives who had
shared the same residence for at least 3 years preceding the onset of the disease (0.43 +/- 0.24 ppm;
n=52). When patients with specific types of leukemia were compared with the healthy unrelated subjects
(0.49 +/- 0.41 ppm; n=79), only those with acute leukemia (type not specified; 1.24 +/- 1.93 ppm; n=23)
had a significantly increased hair mercury content. No significant differences in hair mercury content
were observed in 9 patients with chronic granulocytic leukemia or 15 patients with chronic lymphocytic
leukemia when compared with the unrelated, healthy controls. The authors inferred that acute leukemia
was associated with increased level of mercury in hair. This study is of limited use for cancer risk
assessment because of the following: uncertainty regarding the correlation between the chronology of
incorporation of mercury in the hair and onset of the disease; the small population studied; the failure to
describe adequately the characteristics of the leukemia patients or healthy controls (age
distribution, length of residence in the region, criteria for inclusion in the study); uncertainty regarding
the source of mercury exposure (the authors presumed that exposure was the result of use of
methylmercury-containing fungicides); and the failure to address exposure to other chemicals or adjust
for other leukemia risk factors. Furthermore, the variability of hair mercury
content was large, and the mean hair mercury levels were within normal limits for all groups. Thus, the
statistical significance may have been due to chance.
The carcinogenic effects of organomercury seed dressing exposure were investigated in a series
of case-control studies for incidence of soft-tissue sarcomas (Eriksson et al., 1981, 1990; Hardell and
B-49
-------
Eriksson, 1988) or malignant lymphomas (Hardell et al., 1981). These studies were conducted in
Swedish populations exposed to phenoxyacetic acid herbicides or chlorophenols (the exposures of
primary interest in the studies), organomercury seed dressings, or other pesticides. Exposure frequencies
were derived from questionnaires and/or interviews. Control groups from the same region of the country
were matched to cases based on vital status. A total of 402 cases of soft-tissue sarcoma and (among
persons not exposed to phenoxyacetic acid herbicides) 128 cases of malignant lymphoma were reported.
In each study, the odds ratio for exposure to organomercury in seed dressings and the incidence of
sarcoma or lymphoma was either <1.0 or the range of the 95% confidence interval for the odds ratio
included 1.0; therefore, no association was indicated for organomercury exposure and soft-tissue sarcoma
or malignant lymphoma. The study subjects were likely to have experienced exposures to the other
pesticides and chemicals.
_II.A.3 ANIMAL CARCINOGENICITY DATA
Limited. Three dietary studies in two strains of mice indicate that methylmercury is
carcinogenic. Interpretation of two of the positive studies was complicated by observation of tumors
only at doses that exceeded the MTD. A fourth dietary study in mice and four dietary studies in rats
failed to indicate carcinogenicity associated with methylmercury exposure. Interpretation of four of the
nonpositive studies was limited because of deficiencies in study design or failure to achieve an MTD.
Methylmercuric chloride (>99% pure) was administered in the diet at levels of 0, 0.4, 2 or 10
ppm (0, 0.03, 0.15 and 0.73 (mg/kg)/day in males and 0, 0.02, 0.11 and 0.6 (mg/kg)/day in females) to 60
ICR mice/sex/group for 104 weeks (Hirano et al., 1986). Interim sacrifices (6/sex/group) were conducted
at 26, 52 and 78 weeks. Complete histopathological examinations were performed on all animals found
dead, killed in extremis or killed by design. Mortality, group mean body weights and food consumption
were comparable to controls. The first renal tumor was observed at 58 weeks in a high-dose male, and
the incidence of renal epithelial tumors (adenomas or adenocarcinomas) was significantly increased in
high-dose males (1/32, 0/25, 0/29 and 13/26 in the control, low-, mid- and high-dose groups,
respectively). Ten of the 13 tumors in high-dose males were adenocarcinomas. These tumors were
described as solid type or cystic papillary types of adenocarcinomas. No invading proliferation into the
surrounding tissues was observed. The incidence of renal epithelial adenomas was not significantly
increased in males, and no renal adenomas or adenocarcinomas were observed in any females studied.
Focal hyperplasia of the tubular epithelium was reported to be increased in high-dose males (13/59; other
incidences not reported). Increases in non-neoplastic lesions in high-dose animals provided evidence that
an MTD was exceeded. Non-neoplastic lesions reported as increased in treated males included the
following: epithelial degeneration of the renal proximal tubules; cystic kidney; urinary cast and pelvic
dilatation; and decreased spermatogenesis. Epithelial degeneration of the renal proximal tubules and
degeneration or fibrosis of the sciatic nerve was reported in high-dose females.
Methylmercuric chloride (>99% pure) was administered in the diet at levels of 0, 0.4, 2 or 10
ppm (0, 0.3, 0.14 and 0.69 (mg/kg)/day in males and 0, 0.03, 0.13 and 0.60 (mg/kg)/day in females) to 60
B6C3F1 mice/sex/group for 104 weeks (Mitsumori et al., 1990). In high-dose males, a marked increase
in mortality was observed after week 60 (data presented graphically; statistical analyses not performed by
authors). Survival at study termination was approximately 50, 60, 60 and 20% in control, low-, mid- and
high-dose males, respectively, and 58, 68, 60 and 60% in control, low-, mid- and high-dose females,
respectively. The cause of the high mortality was not reported. At study termination, the mean body
B-50
-------
weight in high-dose males was approximately 67% of controls and in high-dose females was
approximately 90% of controls (data presented graphically; statistical analyses not performed by study
authors). The incidence of focal hyperplasia of the renal tubules was significantly increased in high-dose
males (14/60; the incidence was 0/60 in all other groups). The incidence of renal epithelial carcinomas
(classified as solid or cystic papillary type) was also significantly increased in high-dose males (13/60;
the incidence was 0/60 in all other groups). The incidence of renal adenomas (classified as solid or
tubular type) was also significantly increased in high-dose males; the incidence was 0/60, 0/60, 1/60 and
5/60 in control, low-, mid- and high-dose males, respectively, and 0/60, 0/60, 0/60 and 1/60 in control
low-, mid- and high-dose females, respectively. No metastases were seen in the animals. The incidences
of a variety of non-neoplastic lesions were increased in the high-dose mice including the following:
sensory neuropathy; neuronal necrosis in the cerebrum; neuronal degeneration in the cerebellum; and
chronic nephropathy of the kidney. Males exhibited tubular atrophy of the testis (1/60, 5/60, 2/60 and
54/60 in control, low-, mid- and high-dose, respectively) and ulceration of the glandular stomach (1/60,
1/60, 0/60 and 7/60 in control, low-, mid- and high-dose males, respectively). An MTD was achieved in
mid-dose males and high-dose females. High mortality in high-dose males indicated that the MTD was
exceeded in this group.
Mitsumori et al. (1981) administered 0, 15 or 30 ppm of methylmercuric chloride (>99% pure) in
the diet (0, 1.6 and 3.1 (mg/kg)/day) to 60 ICR mice/sex/group for 78 weeks. Interim sacrifices of up to
6/sex/group were conducted at weeks 26 and 52. Kidneys were microscopically examined from all
animals that died or became moribund after week 53 or were killed at study termination. Lungs from
mice with renal masses and renal lymph nodes showing gross abnormalities were also examined.
Survival was decreased in a dose-related manner; at week 78 survival was 40, 10 and 0% in control, low-
and high-dose males, respectively, and 55, 30 and 0%, in control, low- and high-dose females,
respectively (statistical analyses not performed). The majority of high-dose mice (85% males and 98%
females) died by week 26 of the study. Examination of the kidneys of mice that died or were sacrificed
after 53 weeks showed a significant increase in renal tumors in low-dose males (13/16 versus 1/37 in
controls). The incidence of renal epithelial adenocarcinomas in control and low-dose males was 0/37 and
11/16, respectively. The incidence of renal epithelial adenomas in control and low-dose males was 1/37
and 5/16, respectively. No renal tumors were observed in females in any group. No metastases to the
lung or renal lymph nodes were observed. Evidence of neurotoxicity and renal pathology were observed
in the treated mice at both dose levels. The high mortality in both groups of treated males and in
high-dose females indicated that the MTD was exceeded in these groups. (Note: Hirano et al. (1986)
was a followup to this study.)
Mitsumori et al. (1983, 1984) administered diets containing 0, 0.4, 2 or 10 ppm of
methylmercuric chloride (0, 0.011, 0.05 and 0.28 (mg/kg)/day in males; 0, 0.014, 0.064 and 0.34
(mg/kg)/day in females) to 56/sex/group Sprague-Dawley rats for up to 130 weeks. Interim sacrifices of
10/group (either sex) were conducted at weeks 13 and 26 and of 6/group (either sex) at weeks 52 and 78.
Mortality was increased in high-dose males and females. At week 104, survival was approximately 55,
45, 75 and 10% in control, low-, mid-and high-dose males, respectively, and 70, 75, 75 and 30% in
control, low-, mid- and high-dose females, respectively (data presented graphically). All males in the
high-dose group had died by week 119. Body weight gain was significantly decreased in high-dose
males starting after week 44 and females
after 44 weeks (approximately 10-20%, data presented graphically). No increase in tumor incidence was
observed in either males or females. Noncarcinogenic lesions that were significantly increased in
high-dose rats included the following: degeneration in peripheral nerves and the spinal cord (both
sexes); degeneration of the proximal tubular epithelium (both sexes); severe chronic nephropathy
B-51
-------
(females); parathyroid hyperplasia (both sexes); polyarteritis nodosa and calcification of arterial wall
(females); fibrosis of bone (females); bile duct hyperplasia (males); and hemosiderosis and
extramedullary hematopoiesis in the spleen (males). Mid-dose males exhibited significantly increased
degeneration of the proximal tubular epithelium and hyperplasia of the parathyroid. An MTD was
achieved in mid-dose males and exceeded in high-dose males and high-dose females.
No tumor data were reported in a study using Wistar rats (Munro et al., 1980). Groups of 50
Wistar rats/sex/dose were fed diets containing methylmercury; doses of 2, 10, 50 and 250 (ug/kg)/day
were fed for 26 months. High-dose female rats exhibited reduced body weight gains and showed minimal
clinical signs of neurotoxicity; however, high-dose male rats showed overt clinical signs of neurotoxicity,
decreased hemoglobin and hematocrit values, reduced weight gains and significantly increased mortality.
Histopathologic examination of the high-dose rats of both sexes revealed demyelination of dorsal nerve
roots and peripheral nerves. Males showed severe dose-related kidney damage, and females had minimal
renal damage.
No increase in tumor incidence or decrease in tumor latency was observed in another study using
rats of an unspecified strain (Verschuuren et al., 1976). Groups of 25 female and 25 male rats were
administered methylmercuric chloride at dietary levels of 0, 0.1, 0.5 and 2.5 ppm (0, 0.004, 0.02 and 0.1
(mg/kg)/day) for 2 years. No significant effects were observed on growth or food intake except for a 6%
decrease (statistically significant) in body weight gain at 60 weeks in high-dose females. Survival was
72, 68, 48 and 48% in control, low-, mid- and high-dose males, respectively, and 76, 60, 64 and 56% in
control, low-, mid- and high-dose females, respectively (statistical significance not reported). Increases
in relative kidney weights were observed in both males and females at the highest dose. No effects on
the nature or incidence of pathological lesions were observed, and tumors were reported to have been
observed with comparable incidence and latency among all of the groups. This study was limited by the
small sample size.
No increase in tumor incidence was observed in a study using white Swiss mice (Schroeder and
Mitchener, 1975). Groups of mice (54/sex/group) were exposed until death to methylmercuric acetate in
the drinking water at two doses. The low-dose group received 1 ppm methylmercuric acetate (0.19
(mg/kg)/day). The high-dose group received 5 ppm methylmercuric acetate (0.95 (mg/kg)/day) for the
first 70 days and then 1 ppm thereafter, due to high mortality (21/54 males and 23/54 females died prior
to the dose reduction). Survival among the remaining mice was not significantly different from controls.
Significant reductions in body weight were reported in high-dose males (9-15% lower than controls) and
high-dose females (15-22% lower than controls) between 2 and 6 months of age. After dying, mice were
weighed and dissected; gross tumors were counted, and limited histopathologic sections were made of
heart, lung, liver, kidney and spleen for microscopic examination. This study is limited because
complete histological examinations were not performed.
No increase in tumor incidence was observed in a multiple-generation reproduction study using
Sprague-Dawley rats (Newberne et al., 1972). Groups of rats (30/sex) were given semisynthetic diets
supplemented with either casein or a fish protein concentrate to yield dietary levels of 0.2 ppm
methylmercury (0.008 (mg/kg)/day). Another group of controls received untreated rat chow. Rats that
received diets containing methylmercury during the 2-year study had body weights and hematology
comparable to controls. Detailed histopathological analyses revealed no lesions of the brain, liver, or
kidney that were attributable to methylmercury exposure. Mortality data were not presented.
Interpretation of these data is limited by the somewhat small group sizes and failure to achieve an MTD.
B-52
-------
_II.A.4 SUPPORTING DATA FOR CARCINOGENICITY
Blakley (1984) administered methylmercuric chloride to female Swiss mice (number/group not
specified) in drinking water at concentrations of 0, 0.2, 0.5 or 2.0 mg/L for 15 weeks (approximately 0,
0.03, 0.07 and 0.27 (mg Hg/kg)/day). At the end of week 3, a single dose of 1.5 mg/kg of urethane was
administered intraperitoneally to 16-20 mice/group. No effects on weight gain or food consumption were
observed. Lung tumor incidence in mice not administered urethane (number/group not specified) was
less than one tumor/mouse in all groups. Statistically significant trends for increases in the number and
size of lung adenomas/mouse with increasing methylmercury dose were observed; the number of
tumors/mouse was 21.5, 19.4, 19.4 and 33.1 in control, low-, mid- and high-dose mice, respectively, and
the tumor size/mouse was 0.70, 0.73, 0.76 and 0.76 mm in control, low-, mid- and high-dose mice,
respectively. The study authors suggest that the increase in tumor number and size may have been
related to the immunosuppressive activity of methylmercury. It should be noted that this study is
considered a short-term bioassay, and pulmonary adenomas were the only tumor type evaluated.
Humans ingesting methylmercury-contaminated foods have been reported to experience
chromosomal aberrations (Skerfving et al., 1970, 1974) or SCE (Wulf et al., 1986); however,
interpretation of these studies is limited by methodological deficiencies.
As reviewed in WHO (1990), methylmercury is not a potent mutagen but appears to be capable
of causing chromosome damage and nuclear perturbations in a variety of systems. In Bacillus subtilis,
methylmercury produced DNA damage (Kanematsu et al., 1980). Methylmercury produced
chromosomal aberrations and aneuploidy in human peripheral lymphocytes (Betti et al., 1992), SCE in
human lymphocytes (Morimoto et al., 1982), and DNA damage in human nerve and lung cells as well as
Chinese hamster V-79 cells and rat glioblastoma cells (Costa et al., 1991).
Bone marrow cells of cats treated with methylmercury in a study by Charbonneau et al. (1976)
were examined by Miller et al. (1979). The methylmercury treatment resulted in an increased number of
nuclear abnormalities and an inhibition of DNA repair capacity. Methylmercury induced a weak
mutagenic response in Chinese hamster V-79 cells (Fiskesjo, 1979). Methylmercury also induced
histone protein perturbations and influenced factors regulating nucleolus-organizing activity (WHO,
1990). Moreover, methylmercury has been reported to interfere with gene expression in cultures of
glioma cells (WHO, 1990). Mailhes (1983) reported a significant increase in the number of hyperploid
oocysts in Lak:LVG Syrian hamsters fed methylmercury; however, no evidence of chromosomal damage
was reported. Suter (1975) concluded that strain-specific differences exist with respect to the ability of
methylmercury to produce dominant lethal effects in mice. Nondisjunction and sex-linked recessive
lethal mutations were observed in Drosophila melanogaster treated with methylmercury (Ramel, 1972 as
cited in U.S. EPA, 1985). Methylmercury produced single strand breaks in DNA in cultured L5178Y
cells (Nakazawa et al., 1975).
Negative studies have also been reported. Methylmercury acetate was reported to be negative in
a Salmonella typhimurium assay and a mouse micronucleus assay (Heddle and Bruce, 1977, as reported
in Jenssen and Ramel, 1980). Methylmercury was not mutagenic and did not cause recombination in
Saccharomyces cerevisiae but did slightly increase chromosomal nondisjunction (Nakai and Machida,
1973). Matsumoto and Spindle (1982) reported no significant increase in SCE in developing mouse
embryos; they did report, however, that the developing mouse embryos were highly sensitive to in vitro
treatment with methylmercury.
B-53
-------
_II.B QUANTITATIVE ESTIMATE OF CARCINOGENIC RISK FROM ORAL EXPOSURE
None. The two studies by Mitsumori et al. (1981, 1990) were limited by high mortality in the
high-dose males, the only group to exhibit a statistically significant increase in tumor incidence. Tumors
were observed only in those dose groups in which the MTD had been exceeded. The study by Hirano et
al. (1986) was not limited by low survival, but the tumors were observed in conjunction with
nephrotoxicity and, thus, their incidence may have been a high-dose phenomenon that would not be
expected to occur at low doses. The tumors appeared to originate from focal hyperplasia of the tubular
epithelium induced as a reparative change. The hyperplasia was not observed in tubular epithelium that
was undergoing early degenerative changes. Thus, the tumors may not occur where degenerative
changes do not occur. The genotoxicity data indicate that methylmercury is not a potent mutagen but
may produce chromosomal damage; these data do not support a hypothesis that methylmercury is a
genotoxic carcinogen. It appears, rather, that methylmercury exerts its carcinogenic effect only at high
dose, at or above an MTD. Because the linearized multistage procedure is based on the assumption of
linearity at low doses, the relevance of deriving a slope factor based on data for which a threshold may
exist is questionable.
It is likely that systemic non-cancer effects would be seen at methylmercury exposures lower
than those required for tumor formation. Long-term administration of methylmercury to experimental
animals produces overt symptoms of neurotoxicity at daily doses an order of magnitude lower than those
required to induce tumors in mice.
_II.C QUANTITATIVE ESTIMATE OF CARCINOGENIC RISK FROM INHALATION
EXPOSURE
None.
_II.D EPA DOCUMENTATION, REVIEW, AND CONTACTS (CARCINOGENICITY
ASSESSMENT)
_II.D.l EPA DOCUMENTATION
Source Documents -U.S. EPA, 1995
This IRIS summary is included in The Mercury Study Report to Congress which was reviewed
by OHEA and EPA's Mercury Work Group in November 1994. An interagency review by scientists from
other federal agencies took place in January 1995. The report was also reviewed by a panel of
non-federal external scientists in January 1995 who met in a public meeting on January 25-26. All
reviewers comments have been carefully evaluated and considered in the revision and fmalization of this
IRIS summary. A record of these comments is summarized in the IRIS documentation files.
B-54
-------
_II.D.2 REVIEW (CARCINOGENICITY ASSESSMENT)
Agency Work Group Review ~ 03/03/94
Verification Date - 03/03/94
_II.D.3 U.S. EPA CONTACTS (CARCINOGENICITY ASSESSMENT)
Rita Schoeny / OHEA - (513)569-7544
REFERENCES
Betti, C., T. Davini and R. Barale. 1992. Genotoxic activity of methyl mercury chloride and dimethyl
mercury in human lymphocytes. Mutat. Res. 281(4): 255 260.
Blakley, B.R. 1984. Enhancement of urethane-induced adenoma formation in Swiss mice exposed to
methylmercury. Can. J. Comp. Med. 48: 299 302.
Charbonneau, S.M., I.C. Munro, E.A. Nera et al. 1976. Chronic toxicity of methylmercury in the adult
cat. Toxicology. 5: 337-349.
Costa, M., N.T. Christie, O. Cantoni, J.T. Zelikoff, X.W. Wang and T.G. Rossman. 1991. DNA damage
by mercury compounds: An overview. In: Advances in Mercury Toxicity, T. Suzuki, N. Imura and
T.W. Clarkson, Ed. Plenum Press, New York, NY. p. 255-273.
Eriksson, M., L. Hardell, N.O. Berg, T. Moller and O. Axelson. 1981. Soft-tissue sarcomas and
exposure to chemical substances: A case-referent study. Br. J. Ind. Med. 38:2733.
Eriksson, M., L. Hardell and H.-O. Adami. 1990. Exposure to dioxins as a risk factor for soft-tissue
sarcoma: A population-based case-control study. J. Natl. Cancer Inst. 82(6): 486 490.
Fiskesjo, G. 1979. Two organic mercury compounds tested for mutagenicity in mammalian cells by use
of the cell line V 79-4. Hereditas. 90: 103-109.
Hardell, L. and M. Eriksson. 1988. The association between soft-tissue sarcomas and exposure to
phenoxyacetic acids. A new case-referent study. Cancer. 62: 652 656.
Hardell, L., M. Eriksson, P. Lenner and E. Lundgren. 1981. Malignant lymphoma and exposure to
chemicals, especially organic solvents, chlorophenols and phenoxy acids: A case-control study. Br. J.
Cancer. 43: 169 176.
Heddle, J.R. and W.R. Bruce. 1977. Comparison of the micronucleus and sperm assay for mutagenicity
with the carcinogenic activities of 61 different agents. In: Origins of Human Cancer, H.H. Hiatt, J.D.
Watson, J.A. Winsten, Ed. Vol. 4. Cold Spring Harbor Conferences.
B-55
-------
Hirano, M., K. Mitsumori, K. Maita and Y. Shirasu. 1986. Further carcinogenicity study on
methylmercury chloride in ICRmice. Jpn. J. Vet. Sci. 48(1): 127 135.
Janicki, K., J. Dobrowolski and K. Krasnicki. 1987. Correlation between contamination of the rural
environment with mercury and occurrence of leukemia in men and cattle. Chemosphere. 16: 253 257.
Jenssen, D. and C. Ramel. 1980. The micronucleus test as part of a short-term mutagenicity test
program for the prediction of carcinogenicity evaluated by 143 agents tested. Mutat. Res. 75: 191 202.
Kanematsu, N., M. Hara and T. Kada. 1980. Rec assay and mutagenicity studies on metal compounds.
Mutat. Res. 77: 109-116.
Mailhes, J.B. 1983. Methylmercury effects on Syrian hamster metaphase II oocyte chromosomes.
Environ. Mutagen. 5: 679-686.
Matsumoto, N. and A. Spindle. 1982. Sensitivity of early mouse embryos to methylmercury toxicity.
Toxicol. Appl. Pharmacol. 64: 108-117.
Miller, C.T., Z. Zawidska, E. Nagy and S.M. Charbonneau. 1979. Indicators of genetic toxicity in
leukocytes and granulocytic precursors after chronic methylmercury ingestion by cats. Bull. Environ.
Contain. Toxicol. 21:296-303.
Mitsumori, K., K. Maita, T. Saito, S. Tsuda and Y. Shirasu. 1981. Carcinogenicity of methylmercury
chloride in ICRmice: Preliminary note on renal carcinogenesis. Cancer Lett. 12:305310.
Mitsumori, K., K. Takahashi, O. Matano, S. Goto and Y. Shirasu. 1983. Chronic toxicity of
methylmercury chloride in rats: Clinical study and chemical analysis. Jpn. J. Vet. Sci. 45(6): 747-757.
Mitsumori, K., K. Maita and Y. Shirasu. 1984. Chronic toxicity of methylmercury chloride in rats:
Pathological study. Jpn. J. Vet. Sci. 46(4): 549-557.
Mitsumori, K., M. Hirano, H. Ueda, K. Maita and Y. Shirasu. 1990. Chronic toxicity and
carcinogenicity of methylmercury chloride in B6C3F1 mice. Fund. Appl. Toxicol. 14: 179 190.
Morimoto, K., S. lijima and A. Koizumi. 1982. Selenite prevents the induction of sister-chromatid
exchanges by methyl mercury and mercuric chloride in human whole-blood cultures. Mutat. Res. 102:
183-192.
Munro, I., E. Nera, S. Charbonneau, B. Junkins and Z. Zawidzka. 1980. Chronic toxicity of
methylmercury in the rat. J. Environ. Pathol. Toxicol. 3: 437-447.
Nakai, S. and I. Machida. 1973. Genetic effect of organic mercury on yeast. Mutat. Res. 21(6): 348.
Nakazawa, N., F. Makino and S. Okada. 1975. Acute effects of mercuric compounds on cultured
mammalian cells. Biochem. Pharmacol. 24:489-493.
Newberne, P.M., O. Glaser and L. Friedman. 1972. Chronic exposure of rats to methyl mercury in fish
protein. Nature. 237:40-41.
B-56
-------
Ramel, C. 1972. Genetic effects. In: Mercury in the Environment - An Epidemiological and
Toxicological Appraisal, L. Friberg and J. Vostal, Ed. CRC Press, Cleveland, OH. p. 169-181. (Cited in
U.S. EPA, 1985).
Schroeder, H. and M. Mitchener. 1975. Life-time effects of mercury, methylmercury, and nine other
trace metals in mice. J. Nutr. 105: 452 458.
Skerfving, S., K. Hansson and J. Lindsten. 1970. Chromosome breakage in humans exposed to methyl
mercury through fish consumption. Arch. Environ. Health. 21: 133-139.
Skerfving, S., K. Hansson, C. Mangs, J. Lindsten andN. Ryman. 1974. Methylmercury-induced
chromosome damage in man. Environ. Res. 7:83-98.
Suter, K.E. 1975. Studies on the dominant-lethal and fertility effects of the heavy metal compounds
methylmercuric hydroxide, mercuric chloride, and cadmium chloride in male and female mice. Mutat.
Res. 30: 365-374.
Tamashiro, H., M. Arakaki, H. Akagi, M. Futatsuka and L.H. Roht. 1984. Causes of death in Minamata
disease: Analysis of death certificates. Int. Arch. Occup. Environ. Health. 54:135-146.
Tamashiro H., Arakaki M., Futatsuka M. and E.S. Lee. 1986. Methylmercury exposure and mortality in
southern Japan: A close look at causes of death. J. Epidemiol. Comm. Health. 40: 181-185.
U.S. EPA. 1980. Ambient Water Quality Criteria Document for Mercury. Prepared by the Office of
Health and Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH
for the Office of Water Regulation and Standards, Washington, DC. EPA/440/5-80/058. NTIS
PB81-117699.
U.S. EPA. 1984a. Mercury Health Effects Update: Health Issue Assessment. Final Report. Prepared by
the Office of Health and Environmental Assessment, Environmental Criteria and Assessment Office,
Cincinnati, OH for the Office of Air Quality Planning and Standards, Research Triangle Park, NC.
EPA/600/8-84/019F. NTIS PB81-85-123925.
U.S. EPA. 1984b. Health Effects Assessment for Mercury. Prepared by the Office of Health and
Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH for the
Office of Emergency and Remedial Response, Washington, DC. EPA/540/1086/042. NTIS
PB86-134533/AS.
U.S. EPA. 1988. Drinking Water Criteria Document for Inorganic Mercury. Prepared by the Office of
Health and Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH
for the Office of Drinking Water, Washington, DC. EPA/600/X-84/178. NTIS PB89-192207.
U.S. EPA. 1993. Summary Review of Health Effects Associated with Mercuric Chloride: Health Issue
Assessment (Draft). Prepared by the Office of Health and Environmental Assessment, Environmental
Criteria and Assessment Office, Cincinnati, OH for the Office of Air Quality Planning and Standards,
Research Triangle Park, NC. EPA/600/R-92/199.
B-57
-------
U.S. EPA. 1995. Mercury Study Report to Congress. Office of Research and Development,
Washington, DC. External Review Draft. EPA/600/P-94/002Ab.
Verschuuren, H.G., R. Kroes, E.M. Den Tonkelaar et al. 1976. Toxicity of methylmercury chloride in
rats. III. Long-term toxicity study. Toxicology. 6:107123.
WHO (World Health Organization). 1990. Methylmercury. Vol.101. Geneva, Switzerland: World
Health Organization, Distribution and Sales Service, International Programme on Chemical Safety.
Wulf, H.C., N. Kromann, N. Kousgaard, J.C. Hansen, E. Niebuhr and K. Alboge. 1986. Sister
chromatid exchange (SCE) in Greenlandic eskimos: Dose-response relationship between SCE and seal
diet, smoking, and blood cadmium and mercury concentrations. Sci. Total Environ. 48: 81-94.
B-58
-------
APPENDIX C
ATTENDEES OF U.S. EPA PEER REVIEW WORKSHOP
ON MERCURY ISSUES
-------
LIST OF WORKSHOP ATTENDEES
Giuseppe Andres
State University of New York
206 Farber Hall
Main Street Campus
Department of Pathology
Buffalo, NY 14214
716-831-2846
Karen Blackburn
Environmental Criteria and Assessment Office
U.S. EPA
26 West Martin Luther King Drive
Cincinnati, OH 45268
513-569-7569
Harlal Choudhury
Environmental Criteria and Assessment Office
U.S. EPA
26 West Martin Luther King Drive
Cincinnati, OH 45268
513-569-7536
Tom Clarkson
University of Rochester
P.O. Box EHSC
University of Rochester Medical School
Rochester, NY 14642
716-275-3911
Michael Dieter
NTP, NIEHS, NIH
P.O. Box 12233
Research Triangle Park, NC 27709
919-541-3368
Michael Dourson
Environmental Criteria and Assessment Office
U.S. EPA
26 West Martin Luther King Drive
Cincinnati, OH 45268
513-569-7544
Ernest Foulkes
University of Cincinnati College of Medicine
Department of Environmental Health
Cincinnati, OH 45267-0056
513-872-5769
Kris Khanna
Office of Drinking Water
U.S. EPA
401M Street, S.W.
Washington, DC 20460
202-382-7588
Loren D. Koller
Oregon State University College of Veterinary
Medicine
Corvallis, OR 973 31
503-754-2098
W. Bruce Peirano
Environmental Criteria and Assessment Office
U.S. EPA
26 West Martin Luther King Drive
Cincinnati, OH 45268
513-569-7540
David Reisman
Environmental Criteria and Assessment Office
U.S. EPA
26 West Martin Luther King Drive
Cincinnati, OH 45268
513-569-7588
Paul L. Richter
New Jersey Department of
Environmental Protection
CN413
401 East State Street - Sixth Floor
Trenton, NJ 08625
609-984-9759
C-l
-------
Linda Saunders Cindy Sonich-Mullin
Eastern Research Group, Inc. Environmental Criteria and Assessment Office
6 Whittemore Street U.S. EPA
Arlington, MA 02174 26 West Martin Luther King Drive
617-648-7800 Cincinnati, OH 45268
513-569-7523
Heidi Schultz
Eastern Research Group, Inc. Bob Vanderslice
6 Whittemore Street Office of Drinking Water
Arlington, MA 02174 U.S. EPA
617-648-7800 401 M Street, S.W.
Washington, DC 20460
202-475-6711
C-2
-------
APPENDIX D
HEALTH EFFECTS OF MERCURY AND MERCURY COMPOUNDS
UNCERTAINTY ANALYSIS OF THE METHYLMERCURY RfD
-------
D.I Introduction and Background
The purpose of the analysis in this appendix is two-fold: first, to determine plausible bounds on
uncertainty associated with the data and dose conversions used to derive the methylmercury Reference
Dose (RfD); second, to compare the RfD to estimated distributions of human population thresholds for
adverse effects. The analysis presented in this appendix is a modeled estimate of the human threshold for
specific health effects attributable to methylmercury exposure. The basis for the analysis and the RfD is
the data from the 1971 Iraqi methylmercury poisoning incident, specifically the data from the Marsh et
al. (1987) study. The analysis also includes studies pertinent to the conversion of mercury
concentrations in hair to estimated ingestion levels. The population studied in Marsh et al. (1987) is
hereafter referred to as the Iraqi cohort. The methylmercury RfD was based on a benchmark dose
calculated from the combined developmental effects of late walking, late talking, mental effects, seizures
and neurological effects (scores greater than 3 on a test) in children of women exposed during pregnancy;
benchmark doses for the individual developmental effects and for adult paresthesia were also calculated.
All the benchmark doses for developmental endpoints were calculated from the Iraqi cohort data. The
adult paresthesia benchmark dose was calculated from data presented in Bakir et al. (1973). The studies
and their use in the calculation of the RfD for methylmercury are described in detail in chapters 3 and 6
of Volume IV of this Report.
The approach used in this analysis and the EPA's RfD methodology presuppose the existence of
thresholds for certain health effects. The RfD is defined by the U.S. EPA (U.S. EPA, 1995) as
an estimate (with uncertainty spanning perhaps an order of magnitude) of a daily
exposure to the human population (including sensitive subgroups) that is likely to be
without an appreciable risk of deleterious effects during a lifetime.
This definition implies that the RfD is an exposure level that is below the threshold for adverse effects in
a sensitive subpopulation. For purposes of this analysis, the human population threshold is defined as the
threshold for the most sensitive individual of an identified sensitive subpopulation. The definition of
sensitive subpopulations excludes hypersensitive individuals whose susceptibilities fall far outside the
normal range. A threshold is defined as the level of exposure to an agent or substance below which a
specific effect is not expected to occur. The definition of threshold does not include concurrent exposure
to other agents eliciting the same effect by the same mechanism of action. In other words, there is an
assumption that the induced response is entirely a result of exposure to a single agent. The adverse
health endpoints for the methylmercury RfD as determined by the RfD/RfC Workgroup are the specific
clinically-observed endpoints reported in Marsh et al. (1987). The uncertainty analysis was confined to
those endpoints. The 81 pregnant female/offspring pairs comprising the Iraqi cohort were taken as a
surrogate for the most sensitive subpopulation expected in the general U.S. fish consuming population.
The sensitive subpopulation was specifically identified as humans exposed to methylmercury in utero.
Other analyses of the Iraqi cohort data are available in the literature but are not directly
applicable to the estimation of threshold distributions. An analysis presented in the Seafood Safety
report (NAS, 1991) groups the Iraqi cohort observations by ranges of measured mercury concentrations
in hair in order to estimate the cumulative response distribution. The response data grouped by hair
mercury concentrations groupings were used to calculate the benchmark dose levels on which the
methylmercury RfD was based. As any grouping of data introduces an additional level of uncertainty,
this threshold analysis was based on the ungrouped observations.
D-l
-------
Cox et al. (1989) presented estimates of thresholds based on the ungrouped observations of the
Iraqi cohort for two of the five developmental endpoints considered by the U.S. EPA in the derivation of
the methylmercury RfD. Cox et al. (1989) used a threshold model that included the threshold as a
parameter. The value of the threshold parameter was estimated by a statistical method that optimized the
likelihood at different values of the threshold. The estimated threshold for late walking in offspring (first
walking after 18 months) was 7.3 ppm mercury in hair with an upper 95% confidence limit of 14 ppm.
This threshold value was based on the best (optimized likelihood) estimate for background incidence of
late walking of 0%. The upper 95% confidence limit was highly sensitive to the value of the background
parameter, increasing to 190 ppm mercury in hair for a background of 4%. The optimized likelihood
threshold for neurological effects (neurological scores > 3) based on a background incidence of 9% was
10 ppm mercury in hair with an upper 95% confidence limit of 287 ppm.
The analysis examined the major sources of uncertainty explicitly and implicitly inherent to the
methylmercury RfD and attempted to bound them quantitatively. There are a number of sources of
uncertainty in the estimation of either a human threshold or an RfD from the Iraqi cohort data and from
the dose conversion used to estimate ingestion dose levels from hair mercury concentrations. The
principal uncertainties arise from the following sources: the variability of susceptibilities within the Iraqi
cohort; population variability in the pharmacokinetic processes reflected in the dose conversion; response
classification error; and exposure classification error.
The data show a very broad range of susceptibilities in the 81 individuals of the Iraqi cohort. An
analysis of the response rates based on hair mercury concentrations showed up to a 10,000-fold span
between the 5th and 95th percentiles when projected to the general population (Hattis and Silver, 1994).
Uncertainty in threshold estimates arising from the variability in individual susceptibilities was estimated
by calculating a distribution of thresholds from a regression model for repeated bootstrap samples of the
original Iraqi cohort data set. The bootstrap procedure and regression model are described in section
D.2.1. The bootstrap procedure results in a distribution of population thresholds for specific effects in
units of ppm mercury in hair.
The methylmercury RfD used a dose conversion formula (section 6.3.1.1 of Volume IV of this
report) to estimate the ingestion dose in mg methylmercury per kg body weight per day (mg/kg-day) that
would result in a specified mercury concentration in hair. This formula comprises a number of variables
that are associated with biological processes. There are measured ranges for each variable which can be
attributed to interindividual variability in pharmacokinetics and to experimental variation.
The response classification is the assignment of an individual observation to one of two
categories ~ responder or nonresponder. The response classification for each of the developmental
endpoints reported in Marsh et al. (1987) is based on a fixed value (response decision point) that
constitutes a response when exceeded. It is possible that some observations, particularly those that
represent responses in the immediate vicinity of the response decision point, were misclassified; a
responder may have been classified as a nonresponder or vice versa. The response classifications for late
walking and late talking are particularly susceptible to this type of error. The response estimates were
based on subject recall in members of a population that does not traditionally record these events. The
classification of neurological test battery scores is more objective but still susceptible to some degree of
investigator interpretation and misclassification.
Exposure classification error is the inclusion of individuals in the exposure group who had been
exposed outside a critical period. This type of error is a source of uncertainty for all developmental
endpoints that have a critical period of exposure combined with uncertainty about the actual timing of the
D-2
-------
gestational period. The result of this type of error is the misclassification of an unexposed individual as
an exposed individual. The consequence of this misclassification is an overestimation of the exposure
level associated with a given response or nonresponse and subsequent overestimation of population
variability. For example, in the Iraqi cohort it is noted that an individual with the highest estimated
mercury exposure is a non-responder for developmental effects on the nervous system. This may be due
to differences in individual susceptibility to methyl mercury toxicity, or it may be a consequence of
misclassification; the individual may have been exposed during a period of time which is not critical to
development. There is potential for misclassification as the determination of the correspondence of
gestational period and exposure was dependent on subject recall. Data pertaining to this type of
uncertainty are not yet available.
Other areas of uncertainty are those directly related to the RfD methodology. Specifically, it was
concluded by an Agency Work Group that there were no adequate chronic or reproductive studies. An
uncertainty factor of 10 is generally applied when chronic studies are not available. This uncertainty
factor is based on an assumption inherent to the RfD methodology that increased exposure duration will
lower the dose required for observation of the effect. Support for this assumption has been published
(Weil and McCollister, 1963; Dourson and Stara, 1989) and is discussed in section D.2.2.2 of this
Appendix. An uncertainty factor of 3 is generally applied if reproductive studies are not available. No-
Observed-Adverse-Effect Levels (NOAELs) for reproductive studies are generally 2-fold to 3-fold higher
than NOAELs for chronic studies and are not expected to be the basis for the RfD more than 5% of the
time (Dourson, Knauf and Swartout, 1992).
D.2 Methods
Thresholds were estimated in a two-stage process. The first stage was the estimation of
threshold distributions based on hair mercury concentrations, which was accomplished by applying a
regression model to successive bootstrap samples of the observations in Marsh etal. (1987). This
process is detailed in section D.2.1. The second stage was the conversion of the thresholds expressed as
ppm mercury in hair to mg methylmercury per kg body weight per day (mg/kg-day); this involved a
Monte Carlo analysis of the variability of the underlying biological processes. The dose-conversion
model is described in section D.2.2.
For the uncertainty analysis thresholds for four of the six endpoints evaluated for the
methylmercury RfD and for combined developmental effects were estimated. The developmental effects
included in the threshold analysis were late walking, late talking and neurological effects. Thresholds for
seizures and mental symptoms were not estimated because these effects occurred at 5-fold higher hair-
mercury concentrations than did the other effects. As the resulting thresholds would be much higher than
the others they would not be expected to contribute significantly to the combined developmental effects
threshold distribution as defined for this analysis (the lowest of the individual effect thresholds for each
bootstrap sample). Response rates for seizures and mental symptoms would be expected to influence the
benchmark dose, however, as the benchmark dose is a function of all responses. The data used to
estimate thresholds for adult paresthesia were not the same as those used to calculate the benchmark dose
in the derivation of the methylmercury RfD. The benchmark dose was calculated from the data presented
in Bakir et al. (1973). The threshold estimates were calculated from the Iraqi cohort data so that all
thresholds would be estimated from the same group of individuals to enable a direct comparison. The
Iraqi cohort data are summarized in Table D-l. A plus (+) in Table D-l indicates a positive response. A
positive response for neurological effects was a neurological score greater than 3 as defined in Marsh et
al. (1987). Positive responses for late walking and late talking were 18 months and 24 months (after
birth), respectively (Marsh etal, 1987).
D-3
-------
Table D-l
Incidence of Developmental and Adult Effects as reported in Marsh, et a/., 1987
max ppm
mercury in
hair
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
3
3
5
6
6
7
8
9
10
10
12
12
14
16
18
1°
neuro test
scores > 3
+
+
+
+
late
walking *
+
+
late
talking b
+
+
+
adult
pares-
thesia
+
max ppm
mercury in
hair
23
26
38
45
48
52
59
60
62
72
74
75
78
86
98
104
114
118
154
196
202
242
263
269
294
336
339
357
362
376
399
404
405
418
443
468
557
568
598
674
neuro test
scores > 3
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
late
walking
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
late
talking
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
adult
pares-
thesia
+
+
+
+
+
+
+
+
+
+
+
+
+
a defined as
b defined as
first walking after 18 months
first talking after 24 months
All threshold calculations and Monte Carlo simulations were performed in S-PLUS® (ver 3.2) for
Microsoft® Windows® (ver 3.1) on several microprocessors based on the Intel® 486DX2/66
microprocessors. Sensitivity analyses were performed in Crystal Ball® (ver 3.0) and Excel® (ver 4.0) for
Windows®.
D-4
-------
D.2.1 Estimation of Thresholds Based on ppm Mercury in Hair
Hair mercury concentrations at the thresholds for adult paresthesia, three developmental
endpoints and combined developmental effects were estimated from the Iraqi cohort data. Threshold
estimation was accomplished by applying a regression model to successive bootstrap samples of the 81
observations. The bootstrap method (Efron and Tibshirani, 1991; 1993) is a nonparametric approach that
can be used for estimation of confidence intervals on a given variable without assuming a specific
parametric distribution for that variable. The bootstrap method is based on the assumption that the
observed sample is a random sample of a larger population and provides an estimate of sample-size
uncertainty. The bootstrap process consists of taking a random sample of the same size as the observed
sample distribution from the original sample distribution. The sampling is conducted with replacement
of selected observations prior to the next random selection such that individual observations may appear
more than once in any given sample. In this case, the bootstrap approach was applied in order to allow
estimation of confidence intervals on the dose associated with a given response. For this analysis, 5000
bootstrap samples were generated. Thresholds, in units of ppm mercury in hair, for each endpoint were
calculated for each bootstrap sample. The output was a distribution of 5000 thresholds for each endpoint
representing the variability in the population threshold as estimated from the Iraqi cohort data. The
threshold distribution for combined effects was defined as the minimum of the single-effect threshold
hair-mercury concentrations calculated at each bootstrap iteration; this definition is based on the
assumption that the endpoints were independent. The stability of the bootstrap was evaluated by
determining the change in the 5th and 95th percentiles, and their ratio, for each doubling of the number of
iterations. The bootstrap was considered to be stable if successive estimates were within 5% of each
other.
The threshold model used in this analysis treated background incidence and response related to
exposure (induced response) independently. The procedure is illustrated in Figure D-l, which shows the
Marsh etal, 1987 data and regression lines for developmental neurological effects (neurological score >
3). Figure D-l is an example of a threshold determination from a single iteration of the bootstrap
procedure. All of the response data were binary; that is, individuals were either responders or
nonresponders for a given effect. The binary responses associated with each hair mercury concentration
are indicated by a "+" at the top and bottom of the chart for responders and nonresponders, respectively.
The mean background and fitted regression lines for induced response are shown. The threshold was
defined in the regression model as the concentration of mercury in hair corresponding to the fitted
response equal to the background incidence. This is equivalent to the point of intersection of the
background and induced response line indicated as point A in Figure D-l. This model was chosen so that
the threshold estimate was a consequence of, rather than a contribution to, the estimation of background
and induced response. Other threshold models that could have been used, such as the one used by Cox et
al. (1989), include the threshold as a parameter to be simultaneously estimated with all other model
parameters. Also the maximum likelihood approach for parameter estimation was not used because of
the apparent extreme sensitivity of the upper 95% confidence limit on the threshold to variation in the
background estimate (Cox etal., 1989).
D-5
-------
Figure D-l
Regression Model for Determination of Bootstrap Thresholds
oq
d
cp
d
c\i
d
p
d
+ -H--H-H- •++•
+ + + + observations
— - - background incidence
- - - - background = 0
log-linear fit
— — log-probit fit
+ + + +++ + + -H- + +
++-H-* + -H- + -H- +++ +H+ + +
I
10
50
ppm mercury in hair
100
500
lognormal (GM = 250, GSD = 1.35)
pd = probability density
In the regression model, response was regressed on the logarithm of dose using the probit
function (log-probit model). In those cases where the log-probit model-predicted responses were always
greater than background, a log-linear regression model was used to determine the threshold (point B in
Figure D-l). The log-probit and log-linear fitted regression lines are shown for one bootstrap sample in
Figure D-l.
Hair mercury concentrations of 1 ppm were assumed to represent background exposure levels
(Katz and Katz, 1992). All other observations, the first of which was at 14 ppm for any effect, were
included in the estimation of background and induced response rate as follows. Background incidence
for each effect was estimated directly from each bootstrap sample by performing repeated regressions of
response on hair mercury concentrations, starting with the assumed background range and successively
adding data points at the next higher hair mercury concentration until the regression slope was near zero
and was least statistically significant. Background incidence was defined as the mean of the fitted values
of the resulting regression. The induced response regression slope was calculated in a similar fashion,
starting with all observations above concentrations of 12 ppm hair-mercury and successively adding data
points at the next lower hair mercury concentration until the regression slope was maximized and
statistically significant (p < 0.05).
D-6
-------
D.2.2 Estimation of Ingestion Dose Levels in mg/kg-day
D.2.2.1 Estimation of Dose Conversion Uncertainty
The uncertainty arising from the calculation of ingestion dose levels, in mg/kg-day, corresponding to
measured concentrations of mercury in hair was estimated through analysis of the dose conversion
formula. The formula, which estimates ingested dose levels corresponding to the measured
methylmercury concentration in hair, incorporates the formula used in the derivation of the RfD with the
inclusion of an additional term to account for the hair to blood concentration ratio for methylmercury and
the conversion of elimination constants to their equivalent half-lives. The latter was done as a matter of
convenience as most of the studies reported half-lives rather than elimination constants. The formula
used in the derivation of the RfD is described in Chapter 6 (section 6.3.1.1) of Volume IV of this report
and is reproduced here as equation 1.
C x b x V (1)
A x f x bw
where
d is the daily dietary intake in mg/kg-day,
C is the concentration of methylmercury in the blood in (ig/liter,
b is the elimination constant (of methylmercury from the blood) in days"1,
V is the volume of blood in the body in liters,
A is the fraction of mercury in the diet that is absorbed,
f is the fraction of absorbed mercury that is found in the blood.
bw is body weight in kg.
Variable C in formula 1 can be related to the concentration of mercury in hair by the formula given in
equation 2.
(2)
where
Hgh is the concentration of mercury in hair in ppm (|ig mercury/g hair),
hb is the hair to blood concentration ratio for methylmercury in (ig mercury/g hair/((ig mercury/ml
blood).
Variable b in formula 1, which is a first-order elimination rate constant, and the clearance half-life
arerelated by the formula
log 2 given in equation 3.
b =—^~
f/2
D-7
-------
(3)
where
b is the elimination constant,
loge 2 is the natural logarithm of 2 (= 0.693),
t,/2 is the half-life of methylmercury in the blood.
Substituting for C and b in equation 1 from equations 2 and 3, respectively, gives the formula for
ingestion levels based on mercury concentrations in hair (equation 4).
log. 2 X Hgh X V
hb x tVz x f x A x bw
Dividing both sides of equation 4 by Hgh gives the dose conversion factor, which when multiplied by
a hair mercury concentration gives the corresponding ingestion level in mg/kg-day (equation 5).
log 2 x V
DCF = - ^ - (5)
hb x tVi x f x A x bw
where
DCF is the dose conversion factor in ppm mercury in hair/(mg/kg-day).
Lower levels of exposure to methylmercury were expected to be associated with the observation of
effects in adults for exposure durations longer than those observed for the Iraqi cohort (U.S. EPA, 1995;
Barnes and Dourson, 1988). The potential effect of exposure duration on the dose eliciting chronic
effects is given in equation 6.
DCF
= ~U (6)
where
DCFeda is the exposure-duration adjusted dose conversion factor in ppm mercury in
hair/(mg/kg-day),
DCF is the dose conversion factor (from equation 5),
UD is a unitless adjustment for uncertainty arising from limited exposure duration.
Monte Carlo simulations were conducted for equations 5 and 6. The output of these simulations
were used to calculate a family of ingestion threshold distributions (in mg/kg-day) for each endpoint.
This was done by multiplying the bootstrap threshold distribution for a given endpoint by specific
percentiles of the appropriate dose conversion distribution. Each member of a family of distributions was
associated with a specific probability dependent on the relative likelihood of the DCF.
D-8
-------
D.2.2.2 Input Variable Distributions
Distributions were assigned to each variable in equations 5 and 6 based on the data available in
the literature. The general form of the distribution, whether triangular, normal or lognormal, was
determined by examination of the shape of the distribution of empirical data and by consideration of the
underlying biological and physical processes. A triangular distribution was used when a judgement was
made that the value of the variable fell within identifiable absolute limits. Many of the variables reflect
underlying exponential processes and would be distributed as the logarithm of the nominal values. Such
variables were described as being distributed in log space. In these cases a lognormal, or log-triangular
distribution was chosen to represent the variable.
Determination of the distribution parameters focused on identifying the median (50thpercentile)
and extreme percentiles from the available data. The focus was on the median, rather than the mean, in
order to specify percentiles of the distribution. In general, the mean value of the Monte Carlo output is
more closely related to the median, rather than the mean, of the inputs. The extreme percentiles were
those corresponding to the lowest and highest observations and were determined by multiplying the rank
order of the observation by 100/(n + 1), where n is the total number of observations. For example, the
lowest and highest observations in a sample size of 9 define the 10th and 90th percentiles (80% confidence
interval), respectively. Calculating the percentiles on the basis of n + 1, rather than n, allowed for the
possibility of obtaining more extreme values in additional samples. The median and extreme percentiles
were preserved in the final distribution whenever possible by adjusting the distribution parameters
accordingly. The distribution assigned to each variable is given in Table D-2.
Table D-2
Dose Conversion Monte Carlo Simulation Input Variable Distributions
Variable
hbc
t*
Vd
fe
A
bw
Form
lognormal
log triangular
triangular
log triangular
triangular
lognormal
Nominal a
Value
250 f
53 days
5 liters
0.059s
0.95s
55kg
Parameters b
GM=250, GSD = 1.5
min= 1.455, mode = 1.676, max = 2. 085 (Iog10)
min = 3.63, mode = 5.0, max = 6. 37
min = -1.41, mode = -1.30, max = -0.934 (Iog10)
min = 0.90, mode = 0.95, max =1.0
GM = 55, GSD= 1.13
a median or geometric median
b Key: GM - geometric mean, GSD - geometric standard deviation; min - absolute minimum, mode
most likely value, max = absolute maximum
0 correlated with tVl [correlation coefficient (r) = -0.5]
d correlated with bw (r = -0.47)
e correlated with bw (r = +0.57)
f (ig Hg/g hair/mg Hg/1 blood
8 unitless ratio
Hair to Blood Concentration Ratio for methylmercury (hb)
D-9
-------
This variable represents variation in a population of the ratio of the concentration of
methylmercury in hair to the concentration of methylmercury in blood. The distribution for this variable
was based on the EPA RfD/RfC Work Group's analysis of the available data, which is presented in
Chapter 6 (section 6.3.1.1) of Volume IV of this report. The EPA RfD/RfC Work Group has judged that
the most appropriate value for this variable lies between 200 and 300 based on results published by
Phelps et al. (1980), with 250 selected as the point estimate. The value of 250 was used as the geometric
mean of the distribution for hb. The data given in Phelps et al. (1980) were not detailed enough to allow
determination of the shape or variance of the distribution. The lognormal form for the distribution was
chosen as most representative of the empirical data reported in Sherlock et al. (1982). The geometric
standard deviation (GSD) of 1.5 was estimated in this analysis from the same data (Sherlock et al, 1982).
The distribution is shown in Figure D-2.
A correlation coefficient (r) of-0.5 was assumed between hb and ti/2in the Monte Carlo
simulation of equation 5. The amount of mercury in the hair should be at least partially dependent on
how quickly methylmercury is eliminated from the blood; that is, the faster that methylmercury is
eliminated from the blood, the greater would be the difference between the concentration of mercury in
the hair and mercury in the blood. The relationship between t,/2 and hb would be expected to be inverse
(negative correlation); high values of tVl would be associated with low values of hb. The magnitude of
the correlation coefficient was judged by the U.S. EPA to be at least as strong as -0.5. The data available
for the calculation of the correlation between t,/2 and hb are extremely limited. A correlation coefficient
between hairblood concentration and half-life of about -0.3 was calculated in this analysis from data on
four individuals (Kershaw et al., 1980).
Figure D-2
Probability Density Function for Hair-to-Blood Concentration Ratio (hb)
300 400
hairblood ratio
lognormal (GM = 250, GSD = 1.35)
pd = probability density
D-10
-------
Half-Life ofMethylmercury in the Blood (ty)
Several human studies reported clearance half-lives for methylmercury from blood in the range
of 32-105 days with averages of 45-70 days (Miettinen etal., 1971; Bakir etal., 1973; Greenwood et al.,
1978; Kershaw etal, 1980; Smithed al., 1994). An average elimination constant for methylmercury
from the blood of 0.014 with a range of 0.0099 to 0.0165 was reported by Sherlock, etal. (1984)
corresponding to an average half-life of 50 days with a range of 42-70 days. Table D-3 gives the average
and range of reported half-lives or half-lives calculated from equation 3 for each of the five studies. The
values in Table D-3 were for male and female adults, combined. Average half-lives of methylmercury in
the blood, as reported by Greenwood, et al. (1984), were somewhat shorter for lactating women (46 days)
and longer for children (90 days) than for the adult average of 70 days. A histogram of the combined
data was highly skewed and roughly triangular in shape. The log-triangular distribution (Figure D-3) was
chosen as best representative of the empirical data. The median value of this distribution was 53 days;
this was slightly higher than that used in the derivation of the methylmercury RfD (50 days), which
would result in slightly lower dose conversion values.
Table D-3
Half-Life of Methylmercury in the Blood (days)
Reference
Smith etal. 1994
Sherlock et al. 1984
Kershaw etal. 1980
Bakir etal. 1973
Greenwood et al. 1978
Low
31.9
42
46.7
40
49
Average
45.3
49.5
51.9
65
70
High
60
70
66.5
105
95
D-ll
-------
Figure D-3
Probability Density Function for the Half Life of Methylmercury in the Blood (ty)
50 60 70 80 90 100
half-life (days)
log triangular {min = 1.455, mode = 1.676, max = 2.085 (Iog10)}
pd = probability density
Volume of Blood in the Body (V)
This variable represents the variation in a population of the total volume of blood in the body.
The distribution was based on published values of estimated whole blood volumes for a cohort of 20
pregnant Nigerian women (Harrison, 1966). Whole blood volumes in the third trimester of pregnancy
ranged from 4.0 to 6.0 liters ; the mean and median values were both 5.0 liters (Harrison, 1966). The
distribution of empirical data was roughly triangular and symmetrical. The minimum and maximum
values were adjusted so that the range of observed values fell within the 90% confidence interval (n =
20). The distribution is shown in Figure D-4.
Blood volume was assumed to be positively correlated with body weight; larger blood
volumes would be associated with higher body weights. For this analysis a correlation coefficient of
0.57 between V and bw from the data given in Harrison (1966) was calculated. This correlation
coefficient was assumed for the standard Monte Carlo simulation of equation 5.
D-12
-------
Figure D-4
Probability Density for Blood Volume (V)
0.035 0.040 0.045 0.050 0.055 0.060 0.065
blood volume
triangular {min =3.5 liters, mode = 5.0 liters, max = 6.5 liters}
pd = probability density
Fraction of Absorbed methylmercury in the Blood (f)
This variable reflects the distribution and dilution of the absorbed methylmercury in all
compartments of the body. The distribution on f was based on several human studies showing values in
the range of 5-10% of the absorbed dose of methylmercury in the blood (Miettinen etal, 1971; Kershaw
etal, 1980; Sherlock etal., 1984). All of the measured values have been adjusted for an assumed total
blood volume of 5 liters. The studies are summarized in Chapter 6 (section 6.3.1.1) of Volume IV of this
report. The distribution is shown in Figure D-5. The median value of this distribution of 5.9% was
higher than that used in the derivation of the methylmercury RfD (5%), which would result in lower dose
conversion values.
Sherlock et al. (1984) reported that the fraction of methylmercury in the blood was negatively
correlated with body weight; smaller fractions of methylmercury in the blood were associated with
larger body weights. For this analysis the U.S. EPA calculated a correlation of -0.47 between f and bw
from the data given in Sherlock et al. (1984). This correlation was assumed for the Monte Carlo
simulation of equation 5.
D-13
-------
Figure D-5
Probability Density for Fraction of Absorbed Methylmercury in the Blood (f)
0.05 0.06 0.07 0.0!
fraction MeHg in the blood
log triangular {min= -1.41, mode = -1.30, max = -0.934 (Iog10)}
pd = probability density
Fraction of methylmercury in the Diet that is Absorbed (A)
This distribution was based on the results of two human studies showing uptake of radio-
labeled methylmercury of 95% and greater (Aberg etal, 1969) and 94% and greater (Miettinen etal,
1971) and animal studies showing 90% or greater absorption (summarized in Walsh, 1982). The
distribution reflects the expectation that this value is close to 100 and will not vary much. The
distribution is shown in Figure D-6.
Figure D-6
Proba
bility
Densit
yfor
Fracti
on of
Methy
Imerc
ury
Absor
bed
from
the
Gut
(A)
0.92 0.94 0.96
absorption factor
triangular {min = 0.90, mode = 0.95, max = 1.0}
pd = probability density
D-14
-------
Body Weight (bw)
The distribution for body weight was based on Harrison (1966), previously described for the
definition of V. The observed body weights during the third trimester of pregnancy ranged from 49.5 kg
to 73.9 kg with a geometric mean of 55 kg (Harrison, 1966). A lognormal distribution was visually
fitted to the data. The distribution is shown in Figure D-7. The median value of this distribution of 55 kg
was lower than that used in the derivation of the methylmercury RfD (60 kg). Use of the lower value for
bw which would result in higher dose conversion values.
Figure D-7
Probability Density Distribution for Body Weight (bw)
lognormal (GM = 55 kg, GSD = 1.13)
pd = probability density
Uncertainty Arising from Limited Exposure Duration
This factor is an adjustment for the uncertain effects of exposure duration on the magnitude of
the effective dose. It is based on the assumption that continuing exposure will result in the appearance of
effects at exposure levels wherein there were no effects observed following shorter exposure durations.
The U.S. EPA commonly applies an uncertainty factor of 10 when calculating a chronic RfD from a
study of subchronic duration (U.S. EPA, 1995). In concept, the value of 10 for this uncertainty factor
represents a high estimate of the uncertainty in order to maintain the protective nature of the RfD. An
empirical analysis of the Weil and McCollister (1963) data by Dourson and Stara (1983) supports the use
of an uncertainty factor of 10 as protective. About 50% of ratios of subchronic NOAELs to chronic
NOAELS for rats exposed to a variety of substances other than methylmercury (as reported by Weil and
McCollister, 1963) were below 3.5 and 95% were below 10 (Dourson and Stara, 1983).
The published data were insufficient for the estimation of a distribution for UD. A point
estimate of 4.7 was made for UD from a few studies of methylmercury toxicity in nonhuman primates.
These studies are summarized in Table D-4. Table D-4 gives NOAELs and Lowest-Observed-Adverse-
Effect Levels (LOAELs) for studies of short-term and long-term duration in monkeys. The neurologic
endpoints were limited to clinically-observable effects in order to maintain approximate equivalence of
effects across exposure durations. UD was estimated by dividing the short-term LOAEL of 0.21 mg/kg-
D-15
-------
day (Sato and Ikuta, 1975) by 0.045, the average of the two long-term exposure LOAELS of 0.04 and
0.05 mg/kg-day (Burbacher etal, 1988; Rice and Gilber, 1992).
UD was used in equation 6 to adjust the dose conversion factor for the estimation of the
exposure level associated with chronic effects. Specifically, the exposure duration adjusted dose
conversion factor (DCFeda) from equation 6 was multiplied by the adult paresthesia bootstrap threshold
distribution to obtain an ingestion threshold distribution for chronic neurologic effects. The precise
nature of the chronic effects was not specified because the effects observed in the monkey studies used to
define UD included a number of different neurologic effects. In this case the paresthesia observed in the
Iraqi cohort was used as a surrogate for all possible adult neurologic effects that might occur following
short-term exposure to methylmercury.
Table D-4
Methylmercury Toxicity in Animals
Reference
Sato & Ikuta 1975
Burbacher etal.,
1988
Rice & Gilbert,
1992
Exposure
Duration
36-132 days
3 years
6.5-7 years
Effects
ataxic gait, myoclonic
seizures
slight tremor, motor
incoordination,
blindness
decreased fine motor
performance, other
NOAEL LOAEL
0.07 0.21
none 0.04
none 0.05
D.2.2.3 Correlation of Input Variables
Apart from the assumptions of correlation between individual input variables described
previously (standard correlations), a simplifying assumption was that the susceptibility of any individual
was independent of the value of the dose-conversion factor. It is very likely, however, that susceptibility
and ti/2 are correlated. Longer residence times of methylmercury in the blood, corresponding to longer
half-lives, should have a direct effect on toxicity. Thus, there would be some likelihood that the
susceptibility of the individual at the population threshold (the most sensitive individual) would be
related to larger values of t,/2. Monte Carlo analyses of equation 5 limiting t,/2 to values greater than 53
days (the median of the tVl distribution) or greater than 84 days (the 90th percentile of the ti/2 distribution)
were also conducted. The latter simulation was included only to determine the sensitivity of the output to
changes in the assumption and was not considered to be a realistic scenario. Standard input variable
correlations were assumed for these simulations. Results of this simulation are presented in section D.3.
A sensitivity analysis was conducted to examine the effect of different correlation assumptions
on the relative contribution of each input variable distribution to the variance of the Monte Carlo
simulation output. The sensitivity analysis was performed for standard correlations, no correlations and
for the alternate half-life (t,/2 > 53 days or t,/2 > 84 days) scenarios.
D-16
-------
D.2.3 Estimation of Uncertainty Arising from Response Classification Error
A variable-response model was constructed to assess response classification error. The
variable-response model is identical to the general threshold model except that responses presumed to be
uncertain were given fractional values rather than 0 or 1 (for nonresponders and responders,
respectively). Values of 0 or 1 were generated at each bootstrap sampling by comparison of the
fractional value with random numbers drawn from a uniform distribution between 0 and 1. The uncertain
observations were defined as those that fell close to the defined minimum response. For late walking the
observations that fell into this category were those of 18 - 20 months (late walking = not walking by 18
months). For late talking the uncertain observations were 24 - 26 months (late talking = not talking by 24
months). A value of 0.5 was assigned to each of the observations for late walking and late talking that
were designated uncertain; this represented the largest possible uncertainty in classification (50%
classification error). A 50% classification error was judged to be plausible, given the highly variable
factors involved in the original classifications in Marsh et al. (1987). A separate analysis was conducted
for late walking assuming a 25% classification error. This was done to allow for the possibility that the
large number of observations at exactly 18 months (22), was a result of 18 months being used as an upper
bound, rather than an exact estimate. This could have occurred for observations that were uncertain but
judged by the authors (Marsh et al., 1987) to be 18 months or less.
The determination of neurological scores in Marsh et al. (1987) was considered to be more
objective than the determination of late walking or late talking. There is, however, a possibility that the
distinction between adjacent scores is not absolute. The uncertain observations for neurological effects
were scores of 3 or 4. Although there was no clear basis for determining classification error for this
endpoint, the error was judged likely to be considerably less than for late walking and late talking.
Simulations were run assuming a 10% or 20% classification error. A classification error rate of 20% was
considered to be an upper bound.
There was no basis on which to determine the extent of classification error for adult
paresthesia. In addition, there was no way to determine which responses (or nonresponses) were
marginal. A 5% classification error was assumed for all observations to determine the sensitivity of the
threshold simulation to small error rates.
D.3 Results
D.3.1 Bootstrap Analysis
Figures D-8 to D-10 show the frequency distributions of thresholds for the individual
developmental endpoints resulting from the bootstrap analysis. Figure D-l 1 is the threshold distribution
for the occurrence of any developmental effect. The figures are histograms of the frequency of
occurrence of calculated threshold mercury concentrations resulting from 5,000 iterations of the
bootstrap procedure. The threshold values on the abscissa are given in Iog10 units. These distributions
represent uncertainty in the estimation of population thresholds calculated from hair mercury
concentrations. The small separate peaks at about 600 ppm in Figures D-8 to D-10 represent bootstrap
samples that result in a nonsignificant (p > 0.05) log-probit regression slope. A nonsignificant slope
implies that there was no relationship between hair mercury concentrations and observed response for
that bootstrap sample. In these cases the threshold was defined as the largest hair mercury concentration
in the sample. The frequency with which nonsignificant slopes occurred was interpreted as a measure of
the reliability of the endpoint as a measure of methylmercury toxicity. This occurred in 0.4% of the
bootstrap samples for neurological effects, 0.1% of the late walking bootstrap samples and in 0.2% of the
D-17
-------
late walking bootstrap samples. These percentages effectively limit the upper end of the bootstrap
confidence intervals that can be defined meaningfully. As an example, the upper limit on the bootstrap
confidence interval for the neurological effects threshold was 99.6%. The maximum two-tailed
symmetrical confidence interval on the median threshold for this endpoint that excluded nonsignificant
slopes was 99.2%.
At the other end of the distributions, the log-probit model fails to estimate a threshold when the
lowest log-probit response is greater than the background incidence (log-probit regression line fails to
intersect background in Figure D-l). This would occur in all bootstrap samples where a zero background
incidence was estimated or when the calculated log-probit response at 3 ppm mercury in hair was greater
than the background incidence for that sample. The threshold was calculated from the log-linear
regression model in these cases (point B in Figure D-l); this occurred in 6%, 13.5% and 100% of the
bootstrap samples for neurological effects, late talking and late walking, respectively. The average
background incidence, as estimated from the Iraqi cohort data for each bootstrap sample, was 10.7% for
neurological effects and 8.6% for late talking. Background incidence for late walking was 0% for all
samples.
Figure D-8
Bootstrap Threshold Distribution for Developmental Neurological Effects
5000 iterations
ppm Hg in hair (loglO)
D-18
-------
Figure D-9
Bootstrap Threshold Distribution for Late Walking
5000 iterations
o
° 1
ppm Hg in hair (loglO)
Figure D-10
Bootstrap Threshold Distribution for Late Talking
5000 iterations
ppm Hg in hair (loglO)
D-19
-------
Figure D-ll
Bootstrap Threshold Distribution for Combined Developmental Effects
5000 iterations
0 1
ppm Hg in hair (loglO)
Figure D-12 shows the distribution of bootstrap thresholds for adult paresthesia . The
distribution is a result of 5000 iterations of the bootstrap procedure. Nonsignificant regression slopes
occurred in 7.5% of the samples as shown by the peak at around 2.8 (600 ppm) in Figure D-12. The
largest confidence interval for the adult paresthesia threshold that excluded nonsignificant slopes was
85%. Background incidence for adult paresthesia was 0% for all samples.
Figure D-12
Bootstrap Threshold Distribution for Adult Paresthesia
5002 iterations
° ~1
-.11
ppm Hg in hair (loglO)
D-20
-------
Table D-5 gives selected percentiles from the cumulative bootstrap threshold distributions for
the developmental endpoints and adult paresthesia. The threshold values given in Table D-5 are given in
units of ppm mercury in hair. The values given for the 5th and 95th percentiles in Table D-5 define the
90% bootstrap confidence interval for each threshold. The adult paresthesia thresholds were the lowest
of all the endpoints modeled but showed the greatest variability. The late walking threshold was the
lowest and most variable of the individual developmental endpoint thresholds. The combined-effects
threshold was the least variable of all the thresholds as would be expected from the method of calculation
(minimum of the three individual endpoint thresholds). For the combined developmental effects, the late
walking threshold was the lowest of the three thresholds most often (45%) with neurological effects and
late talking contributing the lowest threshold 31% and 24% of the time, respectively.
Table D-5
Bootstrap Threshold Distributions in ppm Mercury in Hair for All Effects
Endpoint
neurological effects a
late walking b
late talking °
combined developmental effects 24 4
adult paresthesia
Bootstrap Percentile
5th 25th 50th 75th 95th
3.8
3.6
5.5
2.5
0.64
10
8.0
13
5.3
1.5
19
14
20
8.7
2.8
33
25
31
14
5.9
63
58
57
24
>500
a neurological test scores > 3 in children exposed in utero
b walking after 18 months
0 talking after 24 months
d threshold for the occurrence of any developmental effect
All bootstraps stabilized within 4000 iterations as measured by the change in the 5th and 95th
percentiles and the ratio of those percentiles. The largest change from 2000 to 4000 iterations in any of
the stability measurements was 3.5%.
D.3.2 Response Classification Uncertainty
Table D-6 gives percentiles of the cumulative bootstrap threshold distributions resulting from
the consideration of response classification error for each of the endpoints. The distributions were a
result of 5000 iterations of the bootstrap procedure. Frequency plots for these distributions are shown in
Figures D-13 and D-14 for late walking and adult paresthesia, respectively.
D-21
-------
Table D-6
Bootstrap Threshold Distributions in ppm Mercury in Hair with Inclusion of
Response Classification Error
Endpoint
neurological effects
(CEa = 10%)
neurological effects
(CE = 20%)
late walking
(CE = 25%)
late walking
(CE = 50%)
late talking15
adult paresthesia0
Bootstrap Percentile
5th 25th 50th 75th 95th
2.6
2.3
0.74
0.79
2.1
0.50
8.4
7.4
2.5
2.6
5.9
1.5
16
15
5.7
5.0
12
3.3
30
30
15
12
25
15
71
>600
>600
>600
99
>600
a classification error assumption for responses at boundary of minimum value defining a positive response
b 50% classification error (boundary responses)
0 5% classification error assumed for all responses above background
Figure D-13
Bootstrap Threshold Distribution for Late Walking with Response-Classification Error
5000 iterations
0 1
ppm Hg in hair (loglO)
D-22
-------
Figure D-14
Bootstrap Threshold Distribution for Adult Paresthesia with Response-Classification Error
5000 iterations
0 1
ppm Hg in hair (loglO)
The primary result of the assumptions of response classification error was an increase in the
number of bootstrap samples resulting in a nonsignificant log-probit regression slope as shown in Table
D-7. Late walking and adult paresthesia, for which over 20% of the regression slopes were
nonsignificant, were the most sensitive to classification error. Bootstrap confidence intervals of less than
60% were the largest that excluded nonsignificant log-probit slopes for both endpoints. 6.1% of the
slopes for neurological effects were nonsignificant when a 20% classification error was assumed; there
was little effect when the error estimate was reduced to 10%. Only 4.6% of the slopes for late walking
were nonsignificant with a classification error of 50%; the width of the 90% confidence interval,
however, increased by a factor of 4. Eliminating late walking from the combined developmental effects
resulted in a 53% increase in the median bootstrap threshold to 12 ppm; the 5th and 95thpercentiles were
increased to 2.7 and 32 ppm, respectively.
D-23
-------
Table D-7
Percentage of Bootstrap Thresholds Resulting from
Nonsignificant Log-Probit Regression Slopes
Endpoint
neurological effects
late walking
late talking
adult paresthesia
standard1
regression
0.4
0.8
0.2
7.5
classification error
regression
CE2 = 10%
CE = 20%
CE = 25%
CE = 50%
CE = 50%
CE = 5%3
2.1
6.1
10.9
21.9
4.6
20.9
a no classification error
b classification error assumption
0 all observations
D.3.3 Dose Conversion Monte Carlo Simulation and Sensitivity Analysis
Table D-8 shows the results of the Monte Carlo simulation of the dose conversion factor (DCF
in equation 5) for different correlation assumptions. Standard assumptions (scenario 1, Table D-8) were
the distribution assignments and correlations described in section D.2.2.2 and summarized in Table D-2.
Scenario 2 in Table D-8 assumed that all the variables in equation 5 were independent. Scenarios 3 and
4 included the standard assumptions and an additional assumption that an increased residence time of
methylmercury in the blood contributed to the susceptibility of the most sensitive individual. Scenario 3
restricted tVl to the upper half of the standard distribution, representing a moderate association of half-life
and susceptibility, while scenario 4 represented a stronger association, restricting t,/2to the upper 10% of
the standard distribution.
Figure D-15 shows the dose conversion frequency distributions for the standard dose
conversion factor distribution (scenario 1, Table D-8). The values on the abscissa are given in ppm
mercury in hair/(mg/kg-day) in Iog10 units. The distributions in Table D-8 and Figure D-15 represent the
uncertainty in the ratio of the exposure level (in mg/kg-day) to hair mercury concentration for the most
sensitive individual of the exposed population. The nominal dose conversion factor is defined here as the
median value of the standard simulation (scenario 1, Table D-8). The median value for the standard
simulation was 8.0 x 10"5 with a 90% confidence interval spanning a 3.57-fold range. The corresponding
dose conversion value used in the derivation of the methylmercury RfD was 9.8 x 10"5. That is, the
methylmercury RfD would change very little if calculated using the median of the simulated dose
conversion distribution. Using the nominal dose conversion factor, an exposure level of 1 x 10"4
mg/kg-day corresponds to a hair mercury concentration of 1.25 ppm, with a 90% confidence interval of
0.69 ppm to 2.36 ppm.
D-24
-------
Table D-8
Dose Conversion Factor Monte Carlo Simulation Output for Different Correlation Assumptions in
mg/kg-day
Scenario
1) standard correlations8
2) no correlations
3) t,/2 > 53 daysb (std. correlations)
4) t,/2 > 84 days0 (std. correlations)
Percentile
5th 25th 50th 75th 95th
4.2 x 10'5
2.6 x 10'5
3.0 xlO'5
2.2 x 10'5
6.2 x 10'5
5.0 x 10'5
4.7 x 10'5
3.4 x 10'5
8.0 x 10'5
7.9 x 10'5
6.2 x 10'5
4.6 x 10'5
1.0 xlO'4
1.2 xlO'4
8.3 x 10'5
6.1xlO'5
1.5 x 10'4
2.3 x 10'4
1.3 x 10'4
9.4 x 10'5
a hb correlated with tVl (r = -0.5); f correlated with bw (r = -0.47); V correlated with bw (r = +0.57)
^O'percentileofty,
'QO^percentileofty,
Figure D-15
Dose Conversion Distribution (standard assumptions)
10000 iterations
-4.5 -4.0
mg/kg-day (logl 0)
Monte Carlo simulation of equation 4 with HgH = 1
Standard assumptions as in Table D-2
The result of assuming correlations between input variables in equation 5 (Table D-8, scenario 1)
is a 60% reduction of the width of the 90% confidence interval compared to assuming total independence
of inputs (scenario 2). Conversely, restricting ti/2to the upper half of the distribution (correlating
susceptibility and ty) resulted in increased uncertainty around the DCF and lower dose conversion
estimates. Reductions in the median dose conversion estimate were 22% and 42% for a moderate (t,/2>
53 days) and a strong (t% > 84 days) association of tVz and susceptibility, respectively;
D-25
-------
the increase in the width of the 90% confidence interval, as measured by the ratio of the 95th and 5th
percentiles, was about 20% in both cases.
Table D-9 shows the relative contribution of each dose conversion input variable to the variance
of the Monte Carlo simulation output for selected scenarios from Table D-8. It can be seen from Table
D-9 that hb contributed the most to the variance of the output across the scenarios, while bw, V and A
contributed relatively little. The relative contribution to the output variance of tVz and f was highly
sensitive to the correlation assumptions.
Table D-9
Sensitivity Analysis for Dose Conversion Monte Carlo Simulation:
Contribution of Each Input Variable to Output Variance (%)
Input Variable
hb
t*
f
V
bw
A
Scenario
no correlations standard alternate t,/2b
correlations"
47.9
26.7
16.3
4.7
4.3
0
46.5
7.9
33.3
9.6
2.4
0.3
60.4
0.0
29.0
8.3
1.9
0.4
a hb correlated with t,/2 (r=-0.5); f correlated with bw (r=-0.47); V correlated with bw (r=+0.57)
b t,/2 > 53 days
The variability of the dose conversion simulation was somewhat less than the contribution from
the bootstrap procedure. The widths of the 90% bootstrap confidence intervals on the thresholds (in ppm
mercury in hair) ranged from 1.1 to 1.3 orders of magnitude (12-20 fold difference in the 5 * and 95 *
percentiles from Table D-5). The width of the 90% confidence interval for the standard dose conversion
simulation spanned 0.55 orders of magnitude (Table D-8), or about 18-30% of that for the bootstrap
confidence intervals.
D.3.4 Estimation of Ingestion Thresholds
The distributions given in Table D-7 were used to obtain dose-conversion confidence intervals
for specific threshold estimates. Table D-10 gives values at selected percentiles for the distribution of
dose-conversion uncertainty around the median ingestion threshold estimates. The values in Table D-10
are given in units of 10"4 mg/kg-day as a convenience for comparison with the RfD of 1 x 10"4 mg/kg-day.
The distributions in Table D-10 were determined by multiplying the appropriate dose conversion
distribution from Table D-7, as noted in Table D-10, by the median bootstrap threshold estimates for
each of the endpoints given in Table D-5. That is, the distributions in Table D-10 represent the output of
equations 5 or 6 with Hgh equal to the median of the indicated bootstrap distribution. As an example, the
distribution for developmental neurological effects in Table D-10 was a result of multiplying the 5th, 25th,
50th, 75th and 95th percentiles of the standard DCF distribution by the median bootstrap threshold (19
D-26
-------
ppm mercury in hair) for developmental neurological effects. The duration-adjusted adult paresthesia
distribution was the dose conversion distribution for adult paresthesia given in Table D-l 1 divided by 4.7
(UD). Table D-l 1 is the equivalent of Table D-8 for the S^percentile bootstrap threshold estimates from
Table D-5.
Table D-10
Dose Conversion Distributions for Median Ingestion Threshold Estimates
in mg/kg-day (x 104)
Endpoint
neurological effects8
late walking8
late talking8
combined developmental effects8
combined developmental effects (t,/2
> 53 days)b
adult paresthesia8
duration-adjusted adult paresthesia0
Percentile
5th 25th 50th 75th 95th
7.9
6.1
8.4
3.7
2.6
1.2
0.26
12
9.0
12
5.4
4.0
1.7
0.36
15
12
16
7.0
5.4
2.2
0.47
19
15
20
8.7
7.2
2.8
0.60
27
21
29
13
11
4.0
0.86
8 standard assumptions (scenario 1, Table D-9)
b scenario 3, Table D-9
0 adult paresthesia distribution divided by UD
Table D-ll
Dose Conversion Distributions for 5th Percentile Ingestion Threshold Estimates
in mg/kg-day (x 104)
Endpoint
neurological effects8
late walking8
late talking8
combined developmental effects8
combined developmental effects'5
(t,/2 > 53 days)
adult paresthesia8
duration-adjusted adult paresthesia0
Percentile
5th 25th 50th 75th 95th
1.6
1.5
2.3
1.0
0.74
0.27
0.058
2.4
2.2
3.4
1.5
1.1
0.40
0.086
3.1
2.9
4.4
2.0
1.5
0.52
0.11
3.8
3.6
5.5
2.5
2.0
0.65
0.14
5.6
5.2
8.0
3.6
3.1
0.94
0.20
1 standard assumptions (scenario 1, Table D-9)
D-27
-------
b scenario 3, Table D-9
0 adult paresthesia distribution divided by UD
The 5% and 95% columns in Tables D-10 and D-l 1 represent the 90% confidence intervals for
specific percentiles of the thresholds expressed as ingestion levels in mg/kg-day. For example, with
respect to the neurological effects distribution in Table D-10, there is 90% confidence that the true
median threshold for neurological effects is between 7.8 x 10"4 and 2.7 x 10"3 mg/kg-day. Similarly, there
is 90% confidence that the true 5th percentile of the neurological effects threshold distribution is between
1.4 x 10"4 and 4.8 x 10"4 mg/kg-day (Table D-l 1). The median ingestion threshold for duration-adjusted
adult paresthesia was 1 x 10"4 mg/kg-day, with a 90% confidence interval of 2.3 x 10"5 mg/kg-day to 4 x
10"5 mg/kg-day.
Figure D-16 is a plot of the cumulative bootstrap threshold distribution for combined
developmental effects multiplied by values of the dose conversion distribution at selected percentiles.
The plots from left to right in Figure D-16 represent different realizations of the distribution of ingestion
thresholds based on the relative likelihood of specific values of the dose conversion factor (5th, 50th and
95th percentiles). The horizontal box and whisker plot corresponds to the dose conversion distribution
multiplied by the median of the bootstrap threshold distribution for combined developmental effects as
given in Table D-10; the box is the interquartile range (25th to 75th percentiles) and the whiskers are the
5th and 95th percentiles. Figure D-l7 is the same plot for combined developmental effects with the
assumption that t,/2 is greater than 53 days (Table D-8, scenario 3). Figures D-18 to D-22 are the
equivalent plots for the individual-effect thresholds. Ingestion threshold distributions for adult effects
are shown in Figures D-21 and D-22. Figure D-21 is the ingestion threshold distribution for the adult
paresthesia observed in the Iraqi cohort. Figure D-22 is the ingestion threshold distribution for duration-
adjusted adult paresthesia resulting from dividing the adult paresthesia ingestion threshold distribution by
UD. That is, Figure D-22 is the distribution in Figure D-21 shifted to the left by a factor of 4.7.
Figure D-16
Cumulative Combined Developmental Effects Ingestion Threshold Distribution
D-28
-------
ingestion threshold distribution
' TT1 =• dose conversion distribution
- RtD
0.00005 0.0001
0.00050 0.001
threshold (mg/kg-day)
p = cumulative probability
Figure D-17
Cumulative Combined Developmental Effects Ingestion Threshold Distribution
(t,/2 > 53 days)
ingestion threshold distribution
H I I =< dose conversion distribution
0.00050 0.001
threshold (mg/kg-day)
0.00500
p = cumulative probability
Figure D-18
Cumulative Neurological Effects Ingestion Threshold Distribution
D-29
-------
ingestion threshold distribution
1 iTI =* dose conversion distribution
- RtD
0.00005 0.0001
0.00050 0.001
threshold (mg/kg-day)
p = cumulative probability
Figure D-19
Cumulative Late Walking Ingestion Threshold Distribution
ingestion threshold distribution
°iT1 =• dose conversion distribution
0.00050 0.001
threshold (mg/kg-day)
0.00500
p = cumulative probability
Figure D-20
Cumulative Late Talking Ingestion Threshold Distribution
D-30
-------
ingestion threshold distribution
1 iTI =* dose conversion distribution
- RID
0.00005 0.0001
0.00050 0.001
threshold (mg/kg-day)
p = cumulative probability
Figure D-21
Cumulative Adult Paresthesia Ingestion Threshold Distribution
(no exposure duration adjustment)
ingestion threshold distribution
iTI =* dose conversion distribution
0.00001 0.00005 0.0001 0.00050 0.001
threshold (mg/kg-day)
p = cumulative probability
Figure D-22
D-31
-------
Cumulative Duration-Adjusted Adult Paresthesia Threshold Distribution
ingestion threshold distribution
H I I =• dose conversion distribution
1e-6
5e-6 5e-5 1e-4
threshold (mg/kg-day)
5e-4 1e-3
p = cumulative probability
The uncertainty around the location of the RfD within each of the distributions shown in Figures
D-17 to D-22 is indicated by the vertical line at 1 x 10~4 mg/kg-day; this uncertainty came from the dose
conversion variability. As an example, the RfD fell between the 0.035 and 4.5 percentiles of the
combined developmental effects threshold distribution with 90% confidence as determined by the
intersection of the RfD line with the 5 % and 95% ingestion threshold curves (Figure D-16). The median
estimate of the location of the RfD in this distribution was the 0.25 percentile. Similarly, the RfD fell
between the 39th and 91st percentile of the duration-adjusted adult paresthesia threshold distribution with
the median at the 75th percentile (Figure D-22). The RfD was at the 18th percentile of the adult
paresthesia threshold distribution and below the 1st percentile for all of the other threshold distributions.
D.4 Discussion of Uncertainty Analysis
Because the Iraqi cohort is considered to be a sensitive subgroup, as defined in the RfD
methodology, the output distributions of the analysis are meant to reflect the uncertainty around an
estimate of the thresholds for effects in humans including sensitive individuals. The results for each
endpoint should be interpreted as the distribution of the uncertainty around the human population
threshold. The results should not be interpreted as the distributions of individual thresholds within the
population. Estimates of risk above the threshold cannot be obtained from this analysis.
This analysis has attempted to incorporate all areas of uncertainty involved in the derivation of
the methylmercury RfD in Chapter 6 of Volume IV of this report. The 10-fold uncertainty factor (UF)
includes a 3-fold (1005) factor for human variability and a 3-fold factor for the lack of reproductive and
chronic studies. The UF for human variability includes a consideration of both susceptibility and
variation in methylmercury blood half-lives. The bootstrap threshold confidence intervals incorporate
the former, and the latter is explicitly modeled (t,/2) in the dose conversion Monte Carlo analysis.
Uncertainty arising from the lack of chronic data is estimated by UD; this uncertainty was a point estimate
only, as the data were inadequate for defining a distribution for UD. Because UD was derived as a scaling
D-32
-------
factor for adult effects it is not directly comparable to the UF for chronic effects used in the derivation of
the RfD, which was based on developmental endpoints. The only uncertainty included in the RfD and
not addressed here is the uncertainty attributed to the lack of a reproductive study; there are no
appropriate data for the estimation of this uncertainty. In general, reproductive NOAELs are slightly
lower than developmental NOAELs for other substances, but much higher than chronic NOAELs
(Dourson, Knauf and Swartout, 1992). That is, the uncertainty in the thresholds is expected to be much
less for lack of a reproductive study than for lack of a chronic study.
The uncertainty analysis presented in this appendix was limited to only those data and formulae
directly related to the derivation of the methylmercury RfD. Other data sets or models were not
considered. A few sources of uncertainty in the data used to derive the methylmercury RfD have not
been included in this analysis. Exposure classification error arising from uncertainty as to the
correspondence of actual exposure and critical exposure period cannot be estimated from the data as
published in Marsh et al., 1987. This source of uncertainty could be a major contributor to the apparent
extreme variability of susceptibilities in the Iraqi cohort. Variability in the interpretation of the definition
of a response was not estimated in this analysis. That is, there would have been some differences in how
individuals interpreted what constituted first walking or first talking, probably more so for the latter. The
classification errors assumed for this analysis only account for uncertainty in the timing of the event
given an unequivocal positive response. Also, the response decision points defining an adverse effect
were accepted uncritically. For example, changing the definition of late walking to either greater than 16
months or greater than 20 months would have a significant effect on the analysis. Measurement error for
hair mercury concentrations has not been estimated for this analysis; the necessary data are unavailable
in the published reports (Marsh et al., 1987; Cox et al., 1989). In addition, the results of this analysis are
conditional on a specific representation of population variability in the parameters of the dose conversion
variables. That is, the choice of the form, and parameters for the distributions assigned to each of the
variables is largely a matter of judgement. The particular set of parameters chosen for each distribution
is only one option of a number of possible choices; uncertainty as to the value of the parameters is not
included in the analysis.
The threshold analysis shows that adult paresthesia was the most sensitive individual effect
observed for the Iraqi cohort, particularly when adjusted for the effects of continuing exposure. That is,
in this analysis, paresthesia in adults was estimated to be observable at a lower exposure than the
developmental endpoints. The absence of an observed background incidence for paresthesia in the Iraqi
cohort partially contributed to the low threshold estimates. A background incidence for paresthesia
would be expected in the general population. The adult paresthesia bootstrap thresholds were also the
most unstable as measured by the frequency of nonsignificant slopes. The RfD fell between the 39th and
91st percentiles of the duration-adjusted adult paresthesia threshold distribution, a considerably larger
range than that for any of the developmental effects. On the average, the RfD fell below the 1st percentile
for all developmental effects, with only a 5% chance that it was as high as the 16th percentile.
The response-classification uncertainty analyses were based on hypothetical classification error
rates. Assumptions of 50% response-classification error for late walking and late talking were worst-case
for those values immediately adjacent to the response decision point value for any given effect. That is,
for late walking, the values of 18 or 20 months for first walking and 24 or 26 months for first talking
were assumed to be equally likely, resulting in misclassification 50% of the time. This would require an
uncertainty in recall of these events of at least 2 months, which is not unlikely in this particular situation.
The actual classification error was likely to be somewhat less than 50%, particularly as the large number
of observations for late walking at 18 months (22 of the 81 individuals) suggests that 18 months may
have been used as an upper bound in some cases. The response-classification error assumptions for late
walking and late talking were best-case for all other values as no error is assumed. Even with a 25%
D-33
-------
classification error, however, the results of the response-classification uncertainty analysis indicate that
the late walking endpoint was unreliable as a measure of methylmercury toxicity. The exclusion of this
endpoint would not have a very large impact on the combined developmental effects threshold
distribution, increasing the thresholds by about 50%. Although the late talking threshold distribution is
not grossly affected by response-classification error, variability in interpretation of the definition of the
endpoint (first talking) likely would have been greater than that for walking; this uncertainty was not
estimated in this analysis. The neurological effects thresholds were least sensitive to classification error,
assuming that the true error was closer 10% than 20%. The assumption seems reasonable given the much
greater objectivity of the measurement of the effect. Adult paresthesia was the most sensitive to
classification error, showing extreme variability in the threshold estimates with a classification error rate
as low as 5% (all observations). These results suggest that strong conclusions based on the late walking
and adult paresthesia endpoints are unwarranted.
Results of the alternate scenarios (Table D-8) show that the primary effect of the correlation
assumptions among the dose conversion input variables was a fairly large reduction in the variance of the
Monte Carlo simulation output. The assumption of correlation of individual susceptibility and half-life
of methylmercury in the blood did not have a marked effect on the simulation except for a 42% reduction
in the median when a strong correlation was assumed (t,/2> 84 days). The latter scenario probably
represented a worst-case situation although no data were found that directly address the magnitude of the
hypothetical correlation.
The sensitivity analysis indicates that the variables that contribute the most to the dose
conversion simulation variability are the hairblood ratio (hb), the half-life of methylmercury in the blood
(t,/2) and the fraction of absorbed methylmercury found in the blood (f). There is very little that can be
done to reduce the uncertainty in these variables because appropriate data directly applicable to the Iraqi
cohort are not available. These results could be of use in the experimental design and collection of data
for estimates of ingestion levels from hair concentrations in the future.
2.4.5 Conclusions of Analysis of Uncertainty Around Human Health effects of Methylmercury
A major source of the variability was in the estimation of bootstrap thresholds from the Iraqi
cohort data as evidenced by the 12-20 fold difference in the 5th and 95th percentiles of the bootstrap
threshold distributions. The uncertainty arising from limited exposure duration contributed almost as
much, with a 12.5-fold difference in the 5th and 95th percentiles. The corresponding spreads in the dose
conversion distributions were 2.4-4.2 fold. Correlations between variables were important with respect
to the variance of the Monte Carlo simulations but were not well-defined by empirical data. Additional
areas of uncertainty remain to be modeled.
Of the developmental endpoints, the neurological effects, which are determined by a battery of
tests and do not depend on subject recall, would seem to be the most objective measure of methylmercury
toxicity. Late walking was not a reliable endpoint because of sensitivity to classification error.
The RfD of 1 x 10"4 mg/kg-day is very likely well below the threshold for developmental effects
but may be above the threshold for exposure duration-adjusted adult paresthesia. Strong conclusions
based on the latter result are not warranted because of the sensitivity of the adult paresthesia threshold to
classification error and the general lack of data addressing the effects of exposure duration.
D.5 References
D-34
-------
Aberg, B., L. Ekman, R. Falk, U. Greitz, G. Persson and J. Snihs. 1969. Metabolism of methyl mercury
(203Hg) compounds in man. Arch. Environ. Health. 19: 478-484.
Bakir, F., S.F. Damluji, L. Amin-Zaki, M. Murtadha, A. Khalidi, N.Y. Al-Rawi, S. Tikriti, and H.E.
Dhahir (1973). Methylmercury poisoning in Iraq. Science 181: 230-240.
Barnes, D.G. and M.L. Dourson. 1988. Reference dose (RfD): description and use in health risk
assessments. Reg. Tox. Pharm. 3:224-238.
Berglund, F., M. Berlin, G. Birke, et al. 1971. Methyl mercury in fish. A toxicologic-epidemiologic
evaluation of risks. Report from an expert group. Nordisk Hygienisk Tidskrift. Supplementum 4.
Stockholm.
Burbacher, T.M., C. Monnett, K.S. Grant and N.K. Mottet. 1988. Methyl mercury effects on
reproduction and offspring size at birth. Reprod Toxicol l(4):267-278.
Burbacher, T.M., M.K. Mohamed and N.K. Mottet. 1984. Methylmercury exposure and reproductive
dysfunction in the nonhuman primate. Toxicol Appl Pharmacol 75:18-24.
Cox, C., T.W. Clarkson, D.E. Marsh, et al. 1989. Dose-response analysis of infants prenatally exposed
to methyl mercury: An application of a single compartment model to single-strand hair analysis. Environ
Res 49(2):318-332.
Dourson, M.L., L. Knauf and J. Swartout. 1992. On Reference Dose and its underlying toxicity data
base. Toxicol. Indus. Health. 8:171-189.
Dourson, M.L. and J. Stara. 1983. Regulatory history and experimental support of uncertainty factors.
Reg. Tox. Pharm. 3:224-238.
Draper, N.R. and H. Smith. 1966. Applied Regression Analysis. John Wiley & Sons, Inc., New York,
NY.
Efron, B. and R. Tibshirani. 1991. Statistical Data Analysis in the Computer Age. Science. 253:390-
395.
Efron, B. and Tibshirani, R. J. 1993. An Introduction to the Bootstrap. Chapman & Hall, San Francisco.
Evans, H.L., R. Garman and B. Weiss. 1977. Methylmercury: Exposure duration and regional
distribution as determinants of neurotoxicity in nonhuman primates. Toxicol Appl Pharmacol 41:15-33.
Evans, M., N. Hastings and B. Peacock. 1993. Statistical Distributions, second edition, John Wiley &
Sons, Inc., New York, NY.
Harrsion, KA. 1966. Blood volume changes in normal pregnant Nigerian women. J. Obstet. Gynaec.
Brit. Cwlth. 73:717-723.
Hattis, D. and K. Silver. 1994. Human Interindividual Variability ~ A Major Source of Uncertainty in
Assessing Risks for Noncancer Health Effects. Risk Analysis 14:421-432.
D-35
-------
Katz, S.A. and R.B. Katz. 1992. Use of hair analysis for evaluating mercury intoxication of the human
body: a review. J Appl Toxicol 12(2):79-84.
Kershaw, T.G., T.W. Clarkson and P.H. Dhahir. 1980. The relationship between blood levels and dose
of methyl mercury in man. Arch Environ Health 35:28-36.
Marsh, D.O., T.M. Clarkson, C. Cox, G.J. Myers, L. AminZaki, and S. Al-Tikriti. 1987. Fetal
Methylmercury Poisoning: Relationship Between Concentration in Single Strands of Maternal Hair and
Child Effects. Arch. Neurol. 44:1017-1022.
McCullagh, P. and J. A. Nelder. 1983. Generalized Linear Models. Chapman and Hall, London.
Miettinen, J.K., T. Rahola, T. Hattula, et al. 1971. Elimination of 203Hg-methyl mercury in man. Annals
of Clinical Research 3:110-122.
Phelps, R.W., T.W. Clarkson, T.G. Kershaw, et al. 1980. Interrelationships of blood and hair mercury
concentrations in a North American population exposed to methyl mercury. Arch Environ Health
35:161-168.
Rice, B.C. 1989b. Brain and tissue levels of mercury after chronic methyl mercury exposure in the
monkey. J Toxicol Environ Health 27(2): 189-198.
Rice, B.C. and S.G. Gilbert. 1992. Exposure to methyl mercury from birth to adulthood impairs high-
frequency hearing in monkeys. Toxicol Appl Pharmacol 115(1):6-10.
Rice, B.C. and S.G. Gilbert. 1982. Early chronic low-level methyl mercury poisoning in monkeys
impairs spatial vision. Science 216:759-761.
Sato, T. and F. Ikuta. 1975. Long-term studies on the neurotoxicity of small amount of methyl mercury
in monkeys (first report). In: Tsubaki T, ed. Studies on the Health Effects of Alkylmercury in Japan.
Japan: Environment Agency, 63-70.
National Academy of Science. 1991. Methylmercury: FBA Risk Assessment and Current Regulations.
Committee on Evaluation of the Safety of Fishery Products, F.E. Ahmed, Ed. Food and Nutrition Board
Institute of Medicine, National Academy Press, Washington, BC. p. 196-221.
Sherlock, J.C., B.G. Lindsay, J.E. Hislop, W.E. Evans, T.R. Collier. 1982. Buplication diet study on
mercury intake by fish consumers in the United Kingdom. Arch Environ Health 37(5):271-278.
Sherlock, J.C., J. Hislop, B. Newton, G. Topping and K. Whittle. 1984. Elevation of mercury in human
blood from controlled chronic ingestion of methylmercury in fish. Human Toxicol. 3:117-131.
U.S. EPA. 1995. Integrated Risk Information System (IRIS). On-line. Office of Health and
Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH.
Walsh, C.T. 1982. The influence of age on the gastrointestinal absorption of mercuric chloride and
methyl mercury chloride in the rat. Environ. Res. 27: 412-420.
Weil, C. S. and B.B. McCollister. 1963. Relationship between Short- and Long-Term Feeding Studies
in Besigning an Effective Toxicity Test. Agricultural and Food Chemistry 11:486-491.
B-36
------- |